BobboFitos
Joined: 21 Feb 2009
Posts: 200
Location: Cambridge, MA
PostPosted: Wed Aug 11, 2010 8:35 pm Post subject: 2010-11 NBA Win Predictions Reply with quote
Not sure if others have worked on their projections or whatnot, but seems like this thread should start soon. (Reason I thought of this was http://sports.espn.go.com/nba/news/stor ... tStandings - ESPN released their expert picks) I wont have mine up for a little while, but just testing the waters for whenever people want to make their picks.
_________________
http://pointsperpossession.com/
@PPPBasketball
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Fri Aug 13, 2010 5:18 am Post subject: Reply with quote
I think the list from ESPN is quite solid. I don't think it'll be easy to make a substantially better predition. My (very early) guesses where that list can be improved the most:
Milwaukee(-)
New York(-)
Washington(-)
Cleveland(+)
Portland(+, might be a gamble though)
San Antonio(+)
Houston(-)
Phoenix(+)
Sacramento(+)
Golden State(+)
Back to top
View user's profile Send private message
BobboFitos
Joined: 21 Feb 2009
Posts: 200
Location: Cambridge, MA
PostPosted: Fri Aug 13, 2010 8:35 am Post subject: Reply with quote
back2newbelf wrote:
I think the list from ESPN is quite solid. I don't think it'll be easy to make a substantially better predition. My (very early) guesses where that list can be improved the most:
Milwaukee(-)
New York(-)
Washington(-)
Cleveland(+)
Portland(+, might be a gamble though)
San Antonio(+)
Houston(-)
Phoenix(+)
Sacramento(+)
Golden State(+)
Miami for sure. New Orleans for sure. But the latter is probably a little unfair since ESPN made their picks before the acquisition of Ariza.
I'd throw Charlotte in as well. Your list is solid, but I was struck at their underrating of the first two and overrating of the last. (By a substantial margin)
_________________
http://pointsperpossession.com/
@PPPBasketball
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 412
PostPosted: Wed Aug 18, 2010 3:01 pm Post subject: Reply with quote
Well, let me try to enliven in some small way these summer doldrums and be a bit mean to approximately half of "ESPN's best basketball minds". Looking at the forecasted Eastern Conference standings, I see very little wisdom coming from the crowd in expecting the Heat to win only 61 games. And this is not just to quibble with the predicted value (I take the over, by the way.) What I am having trouble imagining is any plausible reasoning that led the more pessimistic half (approximate) of their 93 experts to an expected win total that must have been what? 55 games? 57? And more generally, it is a bit saddening to contemplate the implication that there is little statistical reasoning whatsoever that influenced this crowd.
The average over the last 20 years (a reasonable reference, I think) for the highest regular season win total in the NBA is approximately 64 games. So, straight off, for perspective, what we are being told, is that a team that was 47-35 last year and then acquired possibly the greatest crop of free agents, in the prime of their careers, in the history of the league won't really even approach the average best record in the league. Even a "naive" prediction, for those pessimistic, simply applying the historic second best record in the league (60 games), say, would surely have raised the overall estimate.
Instead some reasoning was used that drove down the average. But what could it have been? I am going to suppose that the basic consensus among the pessimists was the same as the optimists, that as a result of the transactions, Miami would be no better on defense but considerably better on offense. The difference being that the pessimists believed that offensive improvements would only add 10 to 12 wins, max.
It is this implication that seems completely daft. And on this point, neither APM voodoo nor Wins Produced magical regressions need to be invoked. If one believes that the current collection of offensive talent on the Heat is greater than that of last year's Heat roster as well as that of the Cavs, Raptors, and Wizards (and everyone believes this, no?) the inference is that 64 wins should be an approximate lower bound.
Fleshing out the point, LBJ, Bosh, and Miller all had a TS% of about 60%. And collectively they used 50% of their teams' scoring possessions (FGA + 0.44 FTA). Miami, by contrast, jettisoned half their roster, players that utilized 49% of last year's scoring possessions at an average TS% of 53%. As a first cut then, one should expect that half the offense should improve by 7%age points. In terms of Pythogorean Wins, on a base of last year's Miami points scored and allowed, this yields 64 wins (64.2, if it matters).
Now, one can argue that the Juwan Howards of the world are going to get some shots, driving down the average, against which is the more persuasive argument, that with LBJ, Bosh, and Miller on the court, that they and others are going to see their TS%s increase.
The bottom line however is that approximately 64 wins is a huge difference from what the ESPN pessimists were predicting.
What is the larger point?
It is not unreasonable to expect, what with the ubiquitous internet sources of basketball statistics, and public discussions dating back close to twenty years, that all NBA-related media would avail themselves of what is on offer and thereby improve their craft.
Perhaps I am reading too much into a silly poll, however. Maybe the participants weren't really paying much attention, for the anonymitiy? How about ESPN shows the predictions of each participant? And to further incentivize excellence, a modest proposal: How about mimicking the organizational structure of English Football and dividing the 93 experts into three leagues (or more). The experts that best predict the Heat record can form the Premier League, the following 31 would be in the Championship League, followed by League One, or whatever. Then each year thereafter, the bottom and top of each table would be relegated and promoted accordingly. That should better focus the imagination.
But then again, the rant aside, maybe I am just plain wrong about the Heat. How does one get to 58 or so wins with that roster, though? What is the argument?
P.S. And how does the ESPN average for the middling improvement of the Heat coherently relate to the predicted collapse of the James-less Cavs?
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 817
PostPosted: Wed Aug 18, 2010 8:52 pm Post subject: Reply with quote
If one was trying to make the expert panel's average a better predictor, I'd think it would help over years to vary the weights given to individual predictors or even knock some out based on performance as you did. Your structure might well be helpful- market just "the Premier League" prediction average or a variably weighted blend of the leagues.
Or there might be a more flexible or complicated or just different theory driven structure that would be even better that other trained stat experts are more able to suggest, describe and possibly apply.
I haven't really thought much about what my best guess for the Heat would be, at least yet. At first blush though I am not particularly bothered by an average prediction of 61 and I'd probably move fairly cautiously if I moved up.
Royce, do you feel you could post (this time or in the future with the notice and consent of the voters) some fairly high roll-up of the team prediction distributions for the predicted wins? Say the number of votes for over and under the average? Or the number of votes for 1 to 5 wins under or over the predicted average and the same for greater than 5 over or under the predicted average? You could even use large bands for reporting the results- say less than 10 votes in that win prediction range, or 10-30, or more than 30 or whatever to give some detail but not put anyone on a small island. I'd find that both interesting and helpful with regard to the Heat and other teams, if it were considered acceptable.
back2newbelf, can you clarify what your + and - signs mean? That ESPN is too high or low in your opinion or that you personally would adjust the prediction up or down as that sign indicates? I am not clear.
Last edited by Crow on Wed Aug 18, 2010 10:50 pm; edited 3 times in total
Back to top
View user's profile Send private message
Royce
Joined: 13 Feb 2005
Posts: 3
PostPosted: Wed Aug 18, 2010 10:31 pm Post subject: Reply with quote
I might be able to provide more info later, when I have more time.
As for the Heat ... a quick count reveals that 26 voters predicted fewer than 61 wins ... another 35 predicted fewer than 65 ... just to provide a couple of benchmarks that suggest in this case it's a little "easier" to pull the mean down than to lift it ... which is fairly obvious.
High: 71
Low: 50
Average prediction: 62.4
Adjusted average: 61.2
_________________
Royce Webb
NBA Editor | ESPN.com
Back to top
View user's profile Send private message
erivera7
Joined: 19 Jan 2009
Posts: 184
Location: Chicago, IL
PostPosted: Wed Aug 18, 2010 10:51 pm Post subject: Reply with quote
I'd like to know who predicted 50 wins for the Heat on the ESPN panel. For the record, I was one of the panelists and predicted 65 wins.
50 wins? That's ... a little absurd, even from the most pessimistic view.
_________________
@erivera7
I cover the Orlando Magic - Magic Basketball
Back to top
View user's profile Send private message
Kevin Pelton
Site Admin
Joined: 30 Dec 2004
Posts: 979
Location: Seattle
PostPosted: Wed Aug 18, 2010 11:05 pm Post subject: Reply with quote
I'm trying to come up with a technical term that explains why projections have to be regressed to the mean, but I can't think of such a thing. Suffice it to say that just because we know a team will win 64 games doesn't mean that whoever the best team is should be predicted to win 64 games.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Crow
Joined: 20 Jan 2009
Posts: 817
PostPosted: Wed Aug 18, 2010 11:16 pm Post subject: Reply with quote
Since 1980-81 there have been 41 teams at or above 61 wins, 11 at or above 65 wins, 6 at or above 67 wins.
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Wed Aug 18, 2010 11:52 pm Post subject: Reply with quote
Kevin Pelton wrote:
I'm trying to come up with a technical term that explains why projections have to be regressed to the mean, but I can't think of such a thing. Suffice it to say that just because we know a team will win 64 games doesn't mean that whoever the best team is should be predicted to win 64 games.
Yeah, that's a well known effect. We know about how good the best team will be, but we don't know which team that will be.
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Crow
Joined: 20 Jan 2009
Posts: 817
PostPosted: Thu Aug 19, 2010 1:03 am Post subject: Reply with quote
If the Heat get to 67 wins that would at least tie for 3rd best out of 800 some team seasons in the last 30 years or within the top 0.5th percentile. 65 will at least tie for 9th or close to top 1%. 61 would only be about top 5% but if they hit 65 you only miss by 4 games, which is still good for any one prediction.
If one of the big 3 missed 15+ games more than expected or if they collectively missed 15+ more than expected that might knock a couple games off the predicted based on individual data, or worse than average luck in close games could, or perhaps other things (chemistry on offense and / or defense or whatever) might.
Going to 63 - 65 for Miami instead of 61 and being right would have a modest impact on the overall average error contest. It would knock about 0.1 off the average error. But if you are trying to get to an average error of 7 or significantly below 7 every 0.1 is valuable.
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Thu Aug 19, 2010 5:27 am Post subject: Reply with quote
I don't think the Heat care about more than having home-court advantage and might rest some players once they feel they're "safe" in #1. That would result in
max(# of wins of all other teams) + 2 or so
the "+2" is obviously debatable, since they're so good, they might even win 60% of their games when two stars are resting. Miami's strength of schedule for the last ~10 games might be important here for the final projection
Quote:
back2newbelf, can you clarify what your + and - signs mean? That ESPN is too high or low in your opinion or that you personally would adjust the prediction up or down as that sign indicates?
+ means I expect more wins
some comments:
Golden State, after having had 32 PythWins, is finally getting their center back (Biedrins) and they get David Lee (which should definitely outweigh losing Watson, Maggette, Turiaf, Randolph) is projected to 29 wins. Hm.
I think New York has good potential to implode. My tea leaves tell me that Stoudemire will be pissed after 10 games because he's not used to not being spoonfed and all the losses, leading to a whole lot of chemistry issues (even if all goes well I don't expect them to win 37)
Portland(51 PythWins) probably gets more games out of -Oden, Camby, Przybilla, Batum-, all of which are very good players. I don't see them winning 49 games(more like.. 59).
Sacramento(29 PythWins, projected by ESPN to win 30): Just losing Nocioni would have made me predict them win 35 or so. They also get Cousins and Dalembert
Back to top
View user's profile Send private message
mtamada
Joined: 28 Jan 2005
Posts: 377
PostPosted: Thu Aug 19, 2010 4:40 pm Post subject: Reply with quote
Kevin Pelton wrote:
I'm trying to come up with a technical term that explains why projections have to be regressed to the mean, but I can't think of such a thing. Suffice it to say that just because we know a team will win 64 games doesn't mean that whoever the best team is should be predicted to win 64 games.
Yeah, there's undoubtedly a term out there, but I don't know what it is. By saying "regression to the mean", we're 90% there already. A more specific term in the context of forecasting, but which unfortunately doesn't quite fit this situation, is "winner's curse". An example, which also shows how it's in reality yet another manifestation of regression to the mean: if my best estimate of how the Rockets are going to do is 41-41, then I should (using just about any loss function) predict 41 wins for them. But if my best estimate for the Heat is 66 wins, then (for a wide range of reasonable loss functions) I should NOT predict 66 wins, I should regress my forecast a bit downward, to 63 or whatever. Because people who bid high for an item of unknown value are more likely to find that the item's true value is less than they had predicted.
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 412
PostPosted: Wed Aug 25, 2010 11:26 pm Post subject: Reply with quote
I am hoping that Kevin and Mike will expand a bit upon their remarks. In particular, I don't see what "regression to the mean" or "loss functions" have to do with the issue at hand...at least the issue I raised.
The stable of ESPN writers were asked to estimate the mean of the ex ante distribution of wins for each team in the NBA. That 64 out of 93 came up with a number for Miami that was 64 or below is a bit disturbing, as it would seem to betray a lack of familiarity with or interest in statistics and statistical reasoning.
What can explain this result? It cannot be based upon what Crow seems to suggest, that one of the Heat stars misses an unexpected number of games. Because....that is unexpected, part of one tail of the distribution and not the mean.
Unless someone has some inside info, I see no evidence whatsoever that last year's numbers (in terms of games played and productivity, as a baseline) shouldn't be the basis for next. Opinions on the shape of that earth shouldn't differ.
Unless.
Unless there is really a widespread belief that the coaching staff of the Heat stinks. Or, put more kindly, that James and Wade are basically the same player necessarily implies that there is no way to have them on the court at the same time without significant decreases in their productivity (compared to previous years). And that no teammates will see better shots for their combined presence on the floor. But there is no such widespread belief.
And a final point. Estimating the Heat's record matters. Not in the sense that the ESPN poll matters. It matters because it provides as good of a test as one could possibly hope for. A test of one's understanding of the game of NBA basketball, offensive basketball in particular. There is no pretense of defense being a big part of the equation, so no black box there to worry about. And there are no age issues to inject additional uncertainty.
It will be an interesting season.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 817
PostPosted: Thu Aug 26, 2010 12:50 am Post subject: Reply with quote
What specific assumptions are you making about injuries? The injury impact distribution will vary by team and the Heat could have more performance reducing injury impact than the average team.
James looks heavier to me and he is older. These things are both documented to increase injury risk. 2 of the last 3 years he has missed or passed on 6 and 7 games, the most of his career. 10 or more wouldn't surprise me. Bosh has missed 12 or more games 4 of the last 5 years. Wade has missed 20+ games in 3 of 7 years for a career average of 15. On quick inspection, that seems to add up to more than an average injury impact risk for leading players to me.
Diminishing return impacts for likely usage cuts for the big 3, shared rebounding, etc. can reasonably vary. Your in / out TS% change estimate for the offense is pretty rudimentary. How good will they be as a team from 3 pt land in quantity and FG%? None of the big 3 is anywhere close to even average on 3 point accuracy for their careers. That could matter in a way that overall TS% doesn't pick up exactly. They have other 3 point shooters but how many will they get up and how will the team averages turn out?
How well will they accept their roles and get along on the court? I can see a number of possible frictions that might hold performance somewhat below what a simple projection of past stats would predict.
Each additional win after 55 and 60 is probably increasingly tough (likely needing to win more against good teams to achieve this). A linear win estimate system might well predict too high.
Defense is hard to predict. Counterpart data at 82 games suggests Wade and James to be a little better than average 1 on 1 when facing a lot of starters, Bosh moderately worse. Another tool out there for individual impact estimates, but considering both 1 on 1 and help dimensions, was defensive 4 factors and it shows 2 of the big 3 to be weak (40th percentile or less) on 3 of the 4 defensive factors and 4 of those 12 ratings were in the weakest 16% of the league. It is hard to say how good the team defense will be with them in together at the core with a bit less room for defensive assistants to help than the Cavs or Heat could use as desired or needed with James and Wade. It is hard to say how quick the defense gets to its maximum or how consistent they will be.
How intensely focused and successful will they be in Miami-South Beach with all they want to do and will be asked to do off the court and with the constant media scrutiny? No other team will face this level of scrutiny and potential distraction or harm. It is also not unreasonable to expect that almost every team they face will try very hard to keep up or beat them, perhaps more so than against just any routine preseason #1 or #2.
The average wins of the two recent super teams (Boston and LA) who have won titles over their 2 peak was 62.5.
The average for the best team in the 11 seasons post the Jordan championships and the lockout was 62.7 actual wins. 5 times the leader was at 61 or less. 4 times at 65 or more. 62 was the median.
61 or 65. They are different. I am not sure yet which is better bet but I don't think it is so easy and obvious that 65 is better. It would take more a detailed presentation to convince me that 65 is clearly the better estimate and better than 63 as well. 63 seems a bit safer at this point to me on the surface but it would take very little in a detailed analysis to swing the estimate 2 games either way.
page 2 missing
page 3
back2newbelf
Joined: 21 Jun 2005
Posts: 260
PostPosted: Wed Sep 22, 2010 6:01 pm Post subject: Reply with quote
Could you post the entire vegas predictions or link to them, please. Thanks
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 806
PostPosted: Wed Sep 22, 2010 7:55 pm Post subject: Reply with quote
Links for General odds to win the title and
season win propositions for just the Heat and Lakers (at least so far) both listed here.
http://www.sportsbook.com/livesports/in ... basketball
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 409
PostPosted: Thu Sep 23, 2010 9:17 am Post subject: Reply with quote
OK, time to revise and extend my remarks again, about expected injury/participation. A little voice in my head kept saying "What about Penny?".
Basically, the averages I presented didn't best address the circumstance facing DW this year. The assumption was that he would continue be a 35 mpg player at age 29. This served to screen out the "would be" 35 mpg players who dipped below that in year 29 for injury reasons. Back to the drawing board.
So, querying the lovely basketball-reference database, for 35 mpg 28 year old guards over the most recent 20 year range (1999-00 to 2008-09) we then follow these results to their 29 year old season. This picks up broken Penny and leads to a slightly different picture. (And again, for consistency, I culled out those players from the search results that either weren't career starters, had their 29th year in the lockout shortened season, or had the misfortune of perishing in European car accidents: Craig Ehlo, George McCloud, Scott Skiles, Spud Webb, Steve Smith, Drazen Petrovic, Doug Christie, and Tyrone Corbin).
The overall results:
(1) These guards which participated in 73.4 games when they were 28, saw a drop-off the following year, playing in 70.3 games (and playing 35.4 mpg).
(2) For the eight players over the past 20 years having Wade's history of 67.3 games played per season by the age of 28 (these are: Vince Carter, Jerry Stackhouse, Eric Snow, Baron Davis, Anfernee Hardaway, Ron Harper, Tim Hardaway, and Larry Hughes) things look grimmer, their average year 29 consists of 60.5 games (and 33.2 mpg). (And if you add in the two players just above the cut-off, Lindsey Hunter and Chauncey Billups, who averaged 67.3 and 67.7 games per year by age 28, the average of games played in year 29 moves up significantly to 64.7.)
(3) For what it's worth (not much) the player, actually whose experience most looks like Wade's in this framework is Vince Carter, who missed significant amount of games at ages 22, 25 and 26 (though I am certainly not saying the injury history of these two players was similar) and he showed up for 79 games at age 29.
But a little context on point (2). What drives down the average are the data for Penny and Stackhouse (4 and 26 games played, when 29, respectively). Without Penny, the average goes to 68.6 and without both, we return to 75.7.
Would it be cherry-picking to remove them in terms of properly gauging expectations? Yes, in a sense that we have a really small sample here (I await Neil's multivariate analysis for better results) and we shouldn't chuck out data. But, no, in the sense that these two guys were injured before their 29th year old season began. Accordingly, if Wade clears training camp healthy enough to start the first game, the unfortunate experiences of Penny and Stackhouse don't logically apply.
According to these revisions, perhaps I should change my beliefs? Heat win between 64 and 69 games? Nah. Never mind.
P.S. More interesting is another apparent implication of the games played data. If you sort the results (again, average games played per season in up to seven years prior to and including 28, but not including the lock-out year data in the average....whew) by total years in the NBA by age 28, there is a more ominous trend apparent.
6 year veteran guards playing 35 mpg or more at age 28 (that is those entering the league who were 23 at midpoint in their first season) played in 76.1 games the following year. 7 year veterans, those who entered at age 22, their season averaged only 71.25 games at age 29 (75.7 if you kick out Penny). 8 year veterans, their next year's participation drops to 65.4 games (sample of eight). 9 year: 65 games (sample of four). 10 year: 55 games (sample of two). Then 11 year: 74 games (sample of one: Kobe!). Taking the 8 year and over average, the drop is to 60.9 games.
What is the point? Well, if someone wishes they can extend the analysis to expectations for 30 year guards, inclusive of the age 29 season, but if the results are similar, I wouldn't bank on a three-peat just yet in Miami.
Back to top
View user's profile Send private message
mtamada
Joined: 28 Jan 2005
Posts: 376
PostPosted: Wed Oct 06, 2010 1:45 am Post subject: Reply with quote
schtevie wrote:
I am hoping that Kevin and Mike will expand a bit upon their remarks. In particular, I don't see what "regression to the mean" or "loss functions" have to do with the issue at hand...at least the issue I raised.
Sorry, late reply here. My expansion starts with an example: one of the major ways of utilizing regression to the mean is with Stein estimators. The classic example is if it's two weeks into the baseball season and you see a player who's currently batting .412, and you want to predict what his batting average for the rest of the season will be, you do not (if you're smart) use the unbiased estimate of .412. You want to purposely use a biased estimate, and regress to the mean, because extreme performances are unlikely to be sustained.
The key word is extreme. Stein estimators are remarkable because "extreme" and "mean" can be used in a variety of situations; the baseball example is intuitively obvious, not so obvious is that the mean that you regress to doesn't even have to be the mean of baseball players' batting averages (although obviously, the more similar the variables are to each other, the better the Stein estimator will perform, e.g. if you have data on that same player from other seasons, that can be better than using the mean of all other baseball players).
Bottom line, it doesn't have to be two-week baseball averages. ANY extreme value is more likely to regress to the mean rather than maintain (or be realized) at its extreme value.
And this includes extreme forecasts. We're not talking about two weeks into the basketball season, but as I described above, it doesn't matter. The same principles hold.
Do NBA teams sometimes win 72 games? Yes sometimes they do. But you don't get rich by betting on it happening.
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 409
PostPosted: Wed Oct 06, 2010 9:11 pm Post subject: Reply with quote
Mike, I am still not getting the relevance. Further advice requested.
I was being unkind to a significant portion of the ESPN electorate because they had provided guesstimates of the expected performance of the 2010-11 Heat that were, to my reckoning, at variance with what the fundamentals suggest. The issue, as I understand it, is what is the best estimate of the mean to which performance should regress?
For my approximation, I estimated the effect of James et al. on points scored then added it in to B-R's Pythagorean Wins formula for a Miami Heat baseline. This is a simple production function, with the additional input being the change in expected points (= 2009-10 free agent TS% * vacated scoring opportunities with the Heat).
If the argument is that using last season's TS%s (basically career highs) is an error akin to a baseball player starting off the season hitting .412, I take the theoretical point. However, the assumption is bolstered by the evidence from aging curves which suggests that, for 25 year old players, the mean to which they will regress is rising.
The important known unknown - to paraphrase an infamous American - is the degree to which there are diminishing returns in offensive efficiency relating to overlapping responsibilities on offense. Against which there is a theoretical offset: having better teammates gives a player better scoring opportunities, all else equal. My "model" called this even: the free agents don't get worse, but the pre-existing Heat don't get better either. Anyway, this assumption has nothing to do with unforeseen regressions to the mean. Does it?
Finally, for completeness, there are no similar issues to be concerned about in translating estimated points to wins using the Pythagorean Wins formula, are there?
Back to top
View user's profile Send private message
mtamada
Joined: 28 Jan 2005
Posts: 376
PostPosted: Wed Oct 06, 2010 10:39 pm Post subject: Reply with quote
schtevie wrote:
Anyway, this assumption has nothing to do with unforeseen regressions to the mean. Does it?
That assumption, and the rest of the assumptions, indeed seemingly have nothing to do with unforeseen regressions to the mean. But that's why they're unforeseen.
I.e. your estimates are based on what the numbers tell you the expected outcome ought to be. Seemingly the best procedure. And indeed, for non-extreme estimates, it is the best procedure.
But when your estimates are extreme, i.e. when they are predicting an extreme outcome, that's when you start putting less weight on your numbers and more weight on the mean. Due simply to regression to the mean, which indeed has nothing to do with your assumptions, it just happens. And it's unforeseen (but foreseen by people who have the foresight to take regression to the mean into account).
If you draw 24 balls out of a barrel and 10 are white (and the other 14 are black), and someone asks you to estimate the percent of white balls in the barrel, then 10/24 = .417 is as good an estimate as you can come up with, yes? If that's all the information that you have, yes. But if you know that we're talking about the first 24 at bats of the season, then you have additional information (this is where the Bayesianism starts creeping in), and you say wait a minute, I do not believe that this batter is going to keep hitting .417 for the rest of the season.
If you run your LeBron James stats and take age curves into account and teammates and etc. etc. and the numbers tell you 72 wins, well that's as good an estimate as you can come up with, yes? If that's all the information that you have, yes. But if you know that we're talking about NBA teams, then you say wait a minute, NBA teams almost never win 72 games.
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 602
Location: Where the wind comes sweeping down the plains
PostPosted: Thu Oct 07, 2010 12:08 pm Post subject: Reply with quote
Well said, Mike. You have to regress!
BTW, I am showing that James (+11.5) and Wade (+8.5) contribute nearly all of their advantage on offense (there's a lot bigger spread on offense in general--kind of like baseball). There is no way that an offense can run in the +19 efficiency range. No way an offense can even sniff that. The best offense ever was +9.2 (Dallas 2004). Now obviously, their teammates are not league average on offense, though Bosh was a +4.8 on offense last year. That said, I expect tremendous regression to the mean--even a + 10 offense would be astounding. And most teams with such an offensive efficiency can't maintain even average on the other end of the court.
For reference, James, Wade, and Bosh show as +2.2, +1.9, and +0 on the defensive end.
_________________
GodismyJudgeOK.com/DStats
Back to top
View user's profile Send private message Send e-mail Visit poster's website
BobboFitos
Joined: 21 Feb 2009
Posts: 192
Location: Cambridge, MA
PostPosted: Thu Oct 07, 2010 10:42 pm Post subject: Reply with quote
DSMok1 wrote:
Well said, Mike. You have to regress!
BTW, I am showing that James (+11.5) and Wade (+8.5) contribute nearly all of their advantage on offense (there's a lot bigger spread on offense in general--kind of like baseball). There is no way that an offense can run in the +19 efficiency range. No way an offense can even sniff that. The best offense ever was +9.2 (Dallas 2004). Now obviously, their teammates are not league average on offense, though Bosh was a +4.8 on offense last year. That said, I expect tremendous regression to the mean--even a + 10 offense would be astounding. And most teams with such an offensive efficiency can't maintain even average on the other end of the court.
For reference, James, Wade, and Bosh show as +2.2, +1.9, and +0 on the defensive end.
Is it really so hard to imagine this years heat team being the best offense ever, though? I agree you can't just simply "merge" their APM rating or SPM rating or whatever, but I don't think it's a bad bet that they'll have a >+10 offense.
_________________
-Rob
Back to top
View user's profile Send private message AIM Address
schtevie
Joined: 18 Apr 2005
Posts: 409
PostPosted: Sun Oct 10, 2010 9:45 am Post subject: Reply with quote
Mike, perhaps one or more of the ESPN anonyms who put the Heat in the 50s feels a bit better about his or her estimate, for your emphasis of the centrality of "regression to the mean" when it comes to extreme estimates. In the instance, however, I don't think there is much succor to be found in the concept.
I am not arguing with its general utility, nor the importance of recognizing its scope. Indeed, returning to the initial focus of this string, for anyone who is a bit lazy but who still wants a good chance at winning (and certainly not losing) this year's prediction competition, a simple rule of thumb will get you far: take last year's team records and regress them to the mean a bit (say -0.2 or -0.3 for each win above average). With an ad hoc adjustment for the Heat (and other teams that had clear changes in strength) and you'll be in the mix at the finish. Guaranteed.
But invoking "regression to the mean" arguments won't help much in justifying low-ball estimates for the Heat, primarily because what appears to be an extreme estimate, isn't really. And to the extent that an ex post adjustment need to be made to an estimated outcome, "regression to the mean" isn't the relevant concept. (I am pretty sure.)
To take the most important example, extrapolating LBJ's past performance is not evidence of an extreme estimate, precisely because he is an extremely good player. His performances year to year are highly correlated. We are not talking here about an average baseball player who starts the season hitting .417 with 24 at bats. For TS% (the basis of my simple model) there is a multi-year record of nearly 2000 scoring possessions per season. And in terms of how consistent we should expect his performance to be, we can get a sense of to where he might regress, by looking at TS% for the appropriate age and position cohort.
Taking as a relevant sample the last six years (the starting year, 2004-05, chosen so as to have a more similar year to year average scoring propensity in the NBA) for forwards playing 25 mpg or more at a specific age, these are the LBJ's TS% percentiles at ages 20 to 25: 82, 76, 75, 71, 86, and 90. And the non-regression story is essentially identical with Bosh.
OK, so perhaps the relevant regression is to the 80th percentile of this distribution, so perhaps a decrease of 1.5 percentage points in TS% on this account? Against which we are looking at expected improvement due to maturity, suggested by LBJ's increasing propensity to get to the free throw line. The bottom line is that, on net, regression to the mean, at least on the account of expected individual performances anyway, shouldn't be the reason for downgrading the "expected" performance of the Heat this year.
If one is looking for such a reason, it can be found by invoking the assumed concavity of the basketball production function. In such an instance, the wins produced by the expected value of inputs is greater than the expected wins generated by the ex ante distribution of such inputs. Maybe the production function is really concave, and this is the argument that the aforementioned section of the ESPN electorate was invoking?
Finally, DSMok1, any chance you might provide a complete estimate for the Heat using your SAPM numbers? Whatever it turns out to be, "regression to the mean" shouldn't be the argument invoked to explain any expected deviation from reality. Any notional overshooting of the estimate will more properly be on account of fundamentals: the fact that James and Wade play functionally similar positions; what an estimate based on simple addition of SAPM doesn't take into account.
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 602
Location: Where the wind comes sweeping down the plains
PostPosted: Sun Oct 10, 2010 3:16 pm Post subject: Reply with quote
ASPM would be the right acronym. Advanced Statistical Plus-Minus. Yes, I intend to try to create projections for every player in the league and then create team projections from that. A major problem, though, is how to forecast playing time. That's pretty tricky.
_________________
GodismyJudgeOK.com/DStats
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Crow
Joined: 20 Jan 2009
Posts: 806
PostPosted: Sun Oct 10, 2010 5:01 pm Post subject: Reply with quote
In a quick scan of predictions for the Heat I find about 1/3rd in the low 60s, 1/3rd in the mid 60s and 1/3rd in the high 60s.
The impact of big 3s on their usage has been covered some but I went back to check before-after on TS% and Win Shares. Using big usage duos and triplets identified from recent years by Neal at B-R, I found, in rounded terms, TS%s went up for about 70% of the time for those stars while Win Shares went up about 60% of the time. Actual team wins went up by an average of just 2 games over the previous season of the team where the new pair or trio played. There were a couple of huge successes (Boston 07-08 and Cavs 08-09) but a couple of big flops and most were small to modest gains.
It also seems worth noting that both Wade and James have seen attempts to create new big 2s and 3s. James-Williams in 08-09 worked for both guys in the regular season and big for the team. James-Williams-Jamison in 2009-10 didn't work out in the 2010 playoffs. Wade-O'Neal-Walker actually slipped from the previous regular season in team wins but won the title. All 3 improve on TS% while 2 of the 3 improved on WinShares. Wade-Davis was a flop for the team and both players but it wasn't the only reason for the team fall.
The 2003-4 season in LA was good for Payton on both TS% and Win Shares, a split decision for Malone, neutral on one, down on the other for Kobe, down on both for O'Neal. Not enough opportunity for efficiency / effectiveness / glory to spread all the way around?
Back to top
View user's profile Send private message
greyberger
Joined: 27 Sep 2010
Posts: 46
PostPosted: Sun Oct 10, 2010 11:00 pm Post subject: Reply with quote
Very interesting. When the focus is on Lebron and Wade, I wonder how they'll be dispatched - if the idea is to keep at least one of them on the court in every meaningful minute, usage will vary a lot from lineup to lineup. For the appearance of it you can't use Dwayne Wade as a sixth-man, but having him in and fresh when Lebron and Bosh are taking their first rest might be the best way to do it.
With Wade and James you have to figure the flexibility is there, both in the general sense and the 'skill curves' sense. I'd be more worried about nailing the rotation than whether the 'best 5 in' lineup is as good as advertised.
Back to top
View user's profile Send private message Send e-mail
Crow
Joined: 20 Jan 2009
Posts: 806
PostPosted: Mon Oct 11, 2010 12:17 am Post subject: Reply with quote
Nailing the rotation...
The first 10-20 games will probably supercede whatever plan comes before it, but let's look into it.
i'll begin with Wade starting. Who should his backcourt partner be? Chalmers, Arroyo (or House, Miller, James or somebody from the deep bench) Let's stick with the Chalmers-Arroyo choice for now. Wade was modestly better with Arroyo than Chalmers on raw +/-. Who has the larger assist% or lower usage? Arroyo slightly larger assist% and lower usage ratio. I might opt for Arroyo not that it is a huge deal either way. Neither is a good 3 point shooter.
Put James at SF, Bosh at PF. Who starts at center? Anthony, Haslem, Big Z or Magliore? Anthony is by far far the lightest defensive rebounder. Is that pivotal? Maybe not given James edge but it depends on how dominant they want to be on that. His has the lowest usage. Maybe they consider that more important so the main guys get early touches. I probably wouldn't go Anthony first unless that is his main block of time and you want to get that out of the way. But I'm guessing they will. I don't know if that will be a big deal or not. Time will tell. They can change and probably will. Or at least I'd think some experiment makes sense in the long-run before the playoffs.
First off bench? Maybe Haslem for Anthony about 6 minutes in? Then to my thinking at 8 minutes in Bosh sits and Haslem slides and another center enters. Chalmers in at the 6 or 8 mark. Who’s next off James or Wade? I’ll say Wade and so Miller enters at the 10 minute mark. At the start of the 2nd quarter James sits. Bosh re-enters. Arroyo or Chambers choice depends a bit on how well each did. But lets say Arroyo is back in. At 4 minutes into the 2nd quarter Wade and James re-enter with Arroyo. Some center is playing and I haven’t a clue which it will be and am not that motivated to select between them or they go small for a bit. Haslem subs Bosh with 4 minutes left. They finish the half.
They probably run the second half a bit differently but I am just going to double the minutes from this first half for now. That would give you:
Arroyo 20-24 minutes
Chalmers 24-28
(Even a little more or less of each is possible depending of how they and team do as it is a pretty key job.)
Wade 36
Miller 12
Other wing 8 (could be partly or all Miller at times depending on if his is going well and if the opponent is tough)
James 36
TBA 4 (Bosh or James as needed, but might be worth keeping them fresh though)
Bosh 32
Anthony 16 max
Haslem 24
Other centers 24 or maybe a bit more
This is just spur of the moment, first draft but how does that sound or compare to other people’s assumptions or ideas?
In the long-run it will of course depend on how these lineups perform and how teams attack them.
Last edited by Crow on Mon Oct 11, 2010 1:36 am; edited 1 time in total
Back to top
View user's profile Send private message
erivera7
Joined: 19 Jan 2009
Posts: 178
Location: Chicago, IL
PostPosted: Mon Oct 11, 2010 1:01 am Post subject: Reply with quote
I really think Mike Miller needs to start for the Heat, but that's my opinion. I've watched the Heat play against the Pistons and Thunder, and with the amount of time the ball will be in James' hands (Wade, too) to be the facilitator of the offense, I think it'd behoove Erik Spoelstra if he didn't have Miller on the court with the starters. I mean, that's why they signed Miller in the first place ... to serve as the team's primary spot-up shooter.
The alignment defensively is another story, but that's my initial impression.
_________________
@erivera7
I cover the Orlando Magic - Magic Basketball
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 806
PostPosted: Mon Oct 11, 2010 1:30 am Post subject: Reply with quote
I included the possibility Miller starts, though I still lean against it. The argument for as you say is spot up 3 point shooting. The argument against would be will the big 3 leave many shots for that.
He probably plays more than I first suggested or maybe even allowed with expansion over the other wings by also reducing Arroyo or Chambers or playing smaller more often (cutting the center time). I do expect he'll play more in the playoffs by some combination of these means.
He'll probably be used with the big 3 more as you suggest in the 2nd half of games than my rough doubling of the hypothetical first half plan suggests.
2010-11 NBA Win Predictions (BobboFitos, 2010)
2010-11 NBA Win Predictions (BobboFitos, 2010)
Last edited by Crow on Thu May 12, 2011 2:33 am, edited 1 time in total.
Re: 2010-11 NBA Win Predictions
gabefarkas
Joined: 31 Dec 2004
Posts: 1313
Location: Durham, NC
PostPosted: Tue Oct 12, 2010 8:06 pm Post subject: Reply with quote
mtamada wrote:
Sorry, late reply here. My expansion starts with an example: one of the major ways of utilizing regression to the mean is with Stein estimators. The classic example is if it's two weeks into the baseball season and you see a player who's currently batting .412, and you want to predict what his batting average for the rest of the season will be, you do not (if you're smart) use the unbiased estimate of .412. You want to purposely use a biased estimate, and regress to the mean, because extreme performances are unlikely to be sustained.
The key word is extreme. Stein estimators are remarkable because "extreme" and "mean" can be used in a variety of situations; the baseball example is intuitively obvious, not so obvious is that the mean that you regress to doesn't even have to be the mean of baseball players' batting averages (although obviously, the more similar the variables are to each other, the better the Stein estimator will perform, e.g. if you have data on that same player from other seasons, that can be better than using the mean of all other baseball players).
Are you referring to the James-Stein estimator, or Stein's unbiased risk estimate?
Back to top
View user's profile Send private message Send e-mail AIM Address
mtamada
Joined: 28 Jan 2005
Posts: 377
PostPosted: Thu Oct 14, 2010 3:24 am Post subject: Reply with quote
schtevie wrote:
[...]
But invoking "regression to the mean" arguments won't help much in justifying low-ball estimates for the Heat
The lowest of the lowball, yeah sure there are always people who make goofy forecasts, they have their own obscure reasons. Regression to the mean is a moderate correction to apply, it doesn't create weird forecasts -- but it applies to everybody (everybody who's extreme that is; people near the mean can be expected to have minimal regression).
schtevie wrote:
primarily because what appears to be an extreme estimate, isn't really.
Well that is indeed part of the art of forecasting: recognizing what is extreme and what is not. .412 is extremely high for a batter but quite low for an NBA field goal shooter. LBJ is if not the top player in the league, one of the top players. Therefore his forecasts ought to be extreme, yes?
Yes. But they still should be regressed. Every extreme forecast should be. (Though I'd guess that it's easier to wait until the very end and regress Miami's forecasts overall, rather than for each individual player.)
schtevie wrote:
If one is looking for such a reason, it can be found by invoking the assumed concavity of the basketball production function.
In other words, diminishing marginal returns, or in basketball terms, "there's only one ball on the court", i.e. too many superstars start reducing each other's production. The harder question is do they reduce each other's efficiency. I do not know what the models or evidence show. But your original point about the unrelatedness of regression to the mean still applies here: regardless of whether we assume concavity or linearity, the bottom line still holds: if your forecast is an extreme one, and you haven't applied any regression to the mean yet, do so now. The fact that they Heat are loaded with superstars means that they should be forecast for a lot of wins. But beware of a doctrine of exceptionalism, which claims that they are immune from the laws of statistics.
The 2008 Celtics rolled to 66 wins and a championship. But what happened in 2009 and 2010? The 1969 Lakes had a Big 3 that makes the Heat's look puny -- and won 55 games. The 1977 Sixers had perhaps the two best forwards in the league plus 4 former or future All-Stars, and won 50 games. The 1977 Knicks had 4 All-Pros, and won 42 games. The 2004 Lakers had 4 future Hall of Famers, and won 56.
Last edited by mtamada on Thu Oct 14, 2010 3:30 am; edited 1 time in total
Back to top
View user's profile Send private message
mtamada
Joined: 28 Jan 2005
Posts: 377
PostPosted: Thu Oct 14, 2010 3:26 am Post subject: Reply with quote
gabefarkas wrote:
Are you referring to the James-Stein estimator, or Stein's unbiased risk estimate?
The former, I was not familiar with the latter, although apparently it can be used to derive the former.
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 275
PostPosted: Thu Oct 14, 2010 4:53 am Post subject: Reply with quote
http://www.sportsbook.com/livesports/in ... basketball
vegas now with predictions for all the teams.
The Suns and Nuggets are predicted to win substantially less. Miami at 64.5 right now. T-Wolves lowest at 23.5
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 414
PostPosted: Thu Oct 14, 2010 11:59 am Post subject: Reply with quote
Dang. I should have gotten in on the action at 63.5...
Mike, I think that there is value in being more precise when employing the term "regression to the mean" (RTTM). There are two distinct types of phenomena generating the results we see with NBA, league-wide win estimates across seasons, but it is only the first phenomenon that should be meaningfully associated with the term.
The first is statistical, the second for the lack of a better term, is fundamental or model-based (and with this, the zero-sum constraint on the estimation exercise is an important factor).
What I am referring to as "statistical" is the RTTM you observe when, on average, the kids of tall parents don't grow up to be quite as tall. The counterpart in the NBA is when a team because of truly transient good fortune generates wins in excess of what its "genes" would predict, and the following year (all else equal, which it never is) tends to regress to its expected ability.
What I am calling fundamental or model-based phenomena are the effects of all factors that would fit in the "true" NBA production function (that what G*d would reveal, were she so inclined).
One thorny unknown of particular importance in this regard is the precise effect of aging and injuries. But the general story is known and agreed upon. Younger players improve as their skills and knowledge of the game improve. Past their peak, athleticism deteriorates and the propensity for injury increases.
And how does this class of variables (and their variability) influence apparent RTTM? Well, in the ways we are used to expecting, but this result has nothing to do with the "laws of statistics". Apparent RTTM obtains because of the additional fact that better teams tend to have more minutes played by players at their peak, hence more prone to aging and injury effects, and also because the NBA is a zero-sum league (i.e. the games "unexpectedly" lost due to injury and aging, disproportionately suffered by the better teams, pad the win totals of the below-average younger, healthier teams)
As a thought experiment, imagine we lived in an alternate universe where human physiology was different, such that the young rather than the old tended to be injury-prone, and the league had a really hard salary cap instead of a soft one. Much of the apparent RTTM would vanish.
In light of these arguments, we can (kind of) interpret the recent Celtics' history that was mentioned. The last three years saw win totals of 66, 62, and 50 games.
The first issue is: to what win total should the Celtics have rolled in the first year of the Big 3? (The same question as for the Heat this year.) And the answer is, of course, that I don't really know the precise answer. Nobody does, but 66 is probably pretty close to what the "one true model" (approximated best by APM) would have predicted. KG was out for a stretch and played relatively poorly, just prior and after, but, then again, maybe Paul Pierce was playing "above his age" during the season. Etc.
But that emphasizes the main point: the expected value of this production function for the year in question is what matters for a prediction, and no adjustment for RTTM should be made ex post. (If one ends up requiring ad hoc adjustments, it necessarily implies that the model being used to generate predictions is incomplete and giving systematically biased results.)
What then of the following two years? All that can be said is that the trajectory is entirely consistent with fundamentals, but beyond that I have no firm idea as to how they should have been expected to perform. The 62 games seems about right, and last year, my sense is that they endured quite a few "excess" injuries. But I don't have any great confidence about that statement. And what to expect this year? I wish I had a clue. My guess is that they will "regress" to something above 50 wins. And that is the point.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 825
PostPosted: Thu Oct 14, 2010 6:37 pm Post subject: Reply with quote
The Heat line might move again depending on whether Wade & James are considered ready & 100% closer to or on opening day or are still that way 2-4-8 weeks in, if they adjust the line and still take bets. At least in the abstract it should.
Back to top
View user's profile Send private message
Rhuidean
Joined: 11 Mar 2010
Posts: 40
Location: East Bay, CA
PostPosted: Thu Oct 14, 2010 7:17 pm Post subject: . Reply with quote
Schtevie: You've build a model that tells you X. You also have prior knowledge Y about the situation. Unless you have the utmost confidence in your model X and its results, then it makes sense to consider Y also.
From what I've seen (certain ad-hoc down-weightings of APM/SPM values that people used for LBJ, Wade, and Bosh, weighted APM/SPM values that lead to something close to 100% wins), it doesn't seem as if we should have utmost confidence in most of the models just yet. So incorporating prior knowledge makes some sense.
Back to top
View user's profile Send private message Send e-mail
back2newbelf
Joined: 21 Jun 2005
Posts: 275
PostPosted: Sat Oct 23, 2010 8:22 am Post subject: Reply with quote
Can anyone explain me what this vegas line means?
Atlanta Hawks Over 46.5 (-115) Under 46.5 (-105)
specifically the "-115" and "-105".
Obviously they didn't like their initial projection anymore and changed it a bit, but in what direction?
Also, I want to take back my earlier comments about the Knicks. They added more players than I remembered and Stoudemire looks good (at least in pre season)
Back to top
View user's profile Send private message
haralabob
Joined: 11 Apr 2007
Posts: 27
PostPosted: Sat Oct 23, 2010 12:35 pm Post subject: Reply with quote
back2newbelf wrote:
Can anyone explain me what this vegas line means?
Atlanta Hawks Over 46.5 (-115) Under 46.5 (-105)
specifically the "-115" and "-105".
Obviously they didn't like their initial projection anymore and changed it a bit, but in what direction?
Also, I want to take back my earlier comments about the Knicks. They added more players than I remembered and Stoudemire looks good (at least in pre season)
Assume these lines started out as;
Over 46.5 -110
Under 46.5 -110
You'd have to bet $110 to win 100 in profit on either the over or the under.
You know have to bet $115 to win 100 on the over and only $105 to win $100 on the under.
Rather than change their forecast to 47 wins, they move the "juice" to discourage over bettors and encourage people to bet the under.
Back to top
View user's profile Send private message
HoopStudies
Joined: 30 Dec 2004
Posts: 706
Location: Near Philadelphia, PA
PostPosted: Sun Oct 24, 2010 3:40 pm Post subject: Reply with quote
Am I mistaken or did this thread not actually have anyone's actual predictions in it? Wink
_________________
Dean Oliver
Author, Basketball on Paper
The postings are my own & don't necess represent positions, strategies or opinions of employers.
Back to top
View user's profile Send private message Visit poster's website
back2newbelf
Joined: 21 Jun 2005
Posts: 275
PostPosted: Mon Oct 25, 2010 10:55 am Post subject: Reply with quote
Code:
Boston 53
NewJersey 28
Toronto 25
NewYork 37
Philly 27
Cleveland 35
Milwaukee 42
Chicago 45
Detroit 26
Indiana 31
Orlando 59
Miami 63
Washington 28
Atlanta 48
Charlotte 41
Utah 49
Minnesota 27
Denver 46
Oklahoma 51
Portland 52
GoldenState 36
Sacramento 36
Lakers 54
Phoenix 48
Clippers 27
Memphis 39
Houston 40
SA 49
Dallas 52
NO 36
I'll add more comments later today
If I write "player X is good", it means my rating said he was good. Obviously the rating system could have been wrong. I'm just too lazy to write "my rating system blabla" everytime.
"+" means I think it's good for the team (even if they lost a player)
Boston: Lost Rasheed (+)
Cleveland: Vegas has them at 29.5. Absolutely do not agree. Better Coach(+), still a solid starting 5.
Dallas: More Haywood, Chandler (+)
Denver: Vegas has them at 43.5. No idea why
Detroit: Lost their second best player from last year (Jerebko) (-)
Indiana: Watson was strong for them last year (-)
Clippers: Lost Camby (-). I think they would win <20 without Griffin
Lakers: Barnes is good (+). Kobe doesn't look very healthy and Bynum doesn't start the season (-)
Bucks: Lost Ridnour (-). Do not like Chris Douglas Roberts (-)
Wolves: Replaced players in their (by far) two worst positions, PG and SG (Ridnour for Flynn, Johnson for Brewer) (+)
Philly: Lost their best player in Dalembert and brought in a bad players in Hawes and Nocioni(!) (-)
Phoenix: Vegas has them at 41.5. I don't know. Last time everybody thought the Suns would take a huge hit in the standings, they were just fine (back when Stoudemire missed an entire season)
Portland: (Maybe?) More minutes for Batum/Przybilla/Oden/Camby (+)
Sacramento: No Nocioni, more Dalembert, Cousins, Landry (+)
Washington: No Brandan Haywood (-)
The overall method looked like this:
1. Compare last season depth chart with projected depth chart. Look for significant changes in projected production using https://docs.google.com/leaf?id=0An1RF2 ... 1&cindex=8
2. Drift everybody a bit to the middle
3. I had the west winning way more than last year. Adjust for that
4. Adjust for significant changes in division strength
Last edited by back2newbelf on Tue Oct 26, 2010 9:48 am; edited 3 times in total
Back to top
View user's profile Send private message
Ed Küpfer
Joined: 30 Dec 2004
Posts: 787
Location: Toronto
PostPosted: Mon Oct 25, 2010 11:14 am Post subject: Reply with quote
The Painted Area compiled forecasts from Vegas, John Hollinger, Kevin Pelton, and Kelly Dwyer.
Code:
EAST Vegas JH KP KD
MIA 64 66 60 70
ORL 55 55 52 60
BOS 53 51 42 50
CHI 46 48 51 49
ATL 46 46 35 48
MIL 45 49 42 48
CHA 38 30 32 38
NYK 35 37 44 41
PHI 35 43 33 35
IND 34 31 34 28
WAS 33 30 22 31
DET 31 32 27 19
CLE 29 29 39 12
TOR 26 22 35 31
NJN 25 26 40 27
WEST Vegas JH KP KD
LAL 56 56 46 57
OKC 51 49 48 47
POR 51 55 55 50
SAS 50 54 49 55
DAL 50 48 48 52
UTH 48 47 41 52
HOU 47 46 36 46
DEN 43 46 49 50
PHX 41 38 37 52
NOH 40 45 49 44
MEM 38 36 39 40
LAC 36 27 27 38
GSW 32 32 49 21
SAC 28 28 43 26
MIN 23 26 30 13
KevinP hates me and loves David Lewin. That's what I'm getting from this.
_________________
ed
Back to top
View user's profile Send private message Send e-mail
Kevin Pelton
Site Admin
Joined: 30 Dec 2004
Posts: 979
Location: Seattle
PostPosted: Mon Oct 25, 2010 11:53 am Post subject: Reply with quote
Ed Küpfer wrote:
KevinP hates me and loves David Lewin. That's what I'm getting from this.
Yes, the "do I like their analyst?" factor is heavy in my projections. There are few differences, but if we track this again, my final offering is here:
http://www.basketballprospectus.com/art ... cleid=1219
Back to top
View user's profile Send private message Send e-mail Visit poster's website
back2newbelf
Joined: 21 Jun 2005
Posts: 275
PostPosted: Mon Oct 25, 2010 3:43 pm Post subject: Reply with quote
haralabob wrote:
Assume these lines started out as;
Over 46.5 -110
Under 46.5 -110
You'd have to bet $110 to win 100 in profit on either the over or the under.
You know have to bet $115 to win 100 on the over and only $105 to win $100 on the under.
Rather than change their forecast to 47 wins, they move the "juice" to discourage over bettors and encourage people to bet the under.
Thank you.
Does anyone have a "chance that they'll blow their team up/have a firesale to tank for the lottery"-factor in their predictions?
Cleveland and New Orleans (if Paul doesn't return to form) seem like likely candidates
Back to top
View user's profile Send private message
BobboFitos
Joined: 21 Feb 2009
Posts: 201
Location: Cambridge, MA
PostPosted: Tue Oct 26, 2010 2:37 am Post subject: Reply with quote
back2newbelf wrote:
haralabob wrote:
Assume these lines started out as;
Over 46.5 -110
Under 46.5 -110
You'd have to bet $110 to win 100 in profit on either the over or the under.
You know have to bet $115 to win 100 on the over and only $105 to win $100 on the under.
Rather than change their forecast to 47 wins, they move the "juice" to discourage over bettors and encourage people to bet the under.
Thank you.
Does anyone have a "chance that they'll blow their team up/have a firesale to tank for the lottery"-factor in their predictions?
Cleveland and New Orleans (if Paul doesn't return to form) seem like likely candidates
I'm assuming this is what caused the Denver free-fall (they opened at 49.5 I believe and now stand at 43.5)
_________________
http://pointsperpossession.com/
page 5
Author Message
back2newbelf
Joined: 21 Jun 2005
Posts: 220
PostPosted: Tue Oct 26, 2010 9:52 am Post subject: Reply with quote
edited in some minor changes in wins. Hopefully no one gets injured in the next couple of hours
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 385
PostPosted: Tue Oct 26, 2010 11:13 am Post subject: Reply with quote
What's up with Kevin's socialistic, anti-American sentiments? It's one thing for Kevin to hate Ed (Canada is a fine neighbor and all, but it is foreign, and these are tough times) but why does he hate success? Only three teams at 50 wins or more (four if he is willing to round up Chicago's 49.6)? The last three years have seen 12, 9, and 11 such standouts. Gimme a break!
(He's probably in favor of a harder salary cap too.)
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 527
Location: Where the wind comes sweeping down the plains
PostPosted: Tue Oct 26, 2010 11:42 am Post subject: Reply with quote
Advanced SPM projections for the league win totals, adjusting for strength of schedule and rest days:
Code:
TM Margin Rest SoS Total Team Wins
MIA 11.84 -0.11 -0.71 12.45 Miami 68.5
POR 5.57 -0.05 0.10 5.43 Portland 54.9
SAS 4.74 0.10 0.06 4.78 San Antonio 53.4
ORL 3.92 -0.18 -0.29 4.04 Orlando 51.5
ATL 3.57 -0.07 -0.36 3.85 Atlanta 51.1
BOS 3.00 0.02 -0.42 3.44 Boston 50.0
LAL 3.49 -0.18 -0.01 3.32 LA Lakers 49.7
DEN 3.27 -0.11 0.01 3.14 Denver 49.3
CHI 2.31 0.23 -0.47 3.01 Chicago 48.9
UTA 2.78 -0.01 0.11 2.66 Utah 48.0
OKC 2.36 0.12 0.00 2.47 Oklahoma City 47.5
NOH 2.36 0.01 0.21 2.15 New Orleans 46.7
DAL 1.06 0.08 0.24 0.91 Dallas 43.4
PHO 0.45 0.01 0.27 0.19 Phoenix 41.5
GSW 0.28 -0.17 0.14 -0.03 Golden State 40.9
MIL -0.78 0.21 -0.10 -0.47 Milwaukee 39.8
HOU -0.64 -0.04 0.12 -0.80 Houston 38.9
MEM -0.88 0.06 0.16 -0.98 Memphis 38.4
CHA -2.32 0.13 -0.01 -2.19 Charlotte 35.2
PHI -2.46 -0.17 -0.21 -2.41 Philadelphia 34.6
NYK -2.81 -0.19 0.00 -3.01 New York 33.1
CLE -3.51 0.24 -0.05 -3.22 Cleveland 32.5
NJN -3.69 -0.03 -0.08 -3.65 New Jersey 31.4
DET -3.78 -0.04 0.02 -3.84 Detroit 31.0
SAC -4.35 0.11 0.38 -4.62 Sacramento 29.0
IND -4.65 -0.09 -0.09 -4.65 Indiana 29.0
WAS -4.61 0.03 0.12 -4.70 Washington 28.8
LAC -5.09 -0.10 0.30 -5.49 LA Clippers 26.9
TOR -5.69 -0.02 0.05 -5.76 Toronto 26.3
MIN -5.74 0.20 0.52 -6.06 Minnesota 25.6
I personally think there will be significant diminishing returns for Miami; nearly all of the player's contributions above average are on the offensive side of the ball. It would require Miami to maintain a near-historic level of offense (rivaling Dallas in, what was it, 2005?) to actually achieve this level. That's what the projections say, though, so I'll stick to it. I think they may lost a point or 2 of that margin in actuality, but they may be able to compensate with greater effort on defense.
I manually put in Tiago Splitter as a league-average player; all other players without projections come in at replacement level, which I estimate at -3.5 (rather than the commonly accepted -3.0).
Full updated projections here:
Advanced SPM Player and Team Projections
_________________
GodismyJudgeOK.com/DStats
Back to top
View user's profile Send private message Visit poster's website
DSMok1
Joined: 05 Aug 2009
Posts: 527
Location: Where the wind comes sweeping down the plains
PostPosted: Tue Oct 26, 2010 11:46 am Post subject: Reply with quote
schtevie wrote:
What's up with Kevin's socialistic, anti-American sentiments? It's one thing for Kevin to hate Ed (Canada is a fine neighbor and all, but it is foreign, and these are tough times) but why does he hate success? Only three teams at 50 wins or more (four if he is willing to round up Chicago's 49.6)? The last three years have seen 12, 9, and 11 such standouts. Gimme a break!
(He's probably in favor of a harder salary cap too.)
schtevie... it's the maximum likelihood projection, the best estimate for each team.
Which means there will not be nearly as many teams in the 50s, because of regression to the mean. We don't know which teams will break out.
I would be quite happy to try his or my projections against yours, based on the error for each team. I am quite confident our projections, with the fewer outliers will be closer for each team on average (take the average squared error for each team's win total and we'll see what happens).
_________________
GodismyJudgeOK.com/DStats
Back to top
View user's profile Send private message Visit poster's website
schtevie
Joined: 18 Apr 2005
Posts: 385
PostPosted: Tue Oct 26, 2010 1:15 pm Post subject: Reply with quote
DSMok1 wrote:
schtevie wrote:
What's up with Kevin's socialistic, anti-American sentiments? It's one thing for Kevin to hate Ed (Canada is a fine neighbor and all, but it is foreign, and these are tough times) but why does he hate success? Only three teams at 50 wins or more (four if he is willing to round up Chicago's 49.6)? The last three years have seen 12, 9, and 11 such standouts. Gimme a break!
(He's probably in favor of a harder salary cap too.)
schtevie... it's the maximum likelihood projection, the best estimate for each team.
Which means there will not be nearly as many teams in the 50s, because of regression to the mean. We don't know which teams will break out.
You would have to go all the way back to never to find such equality in the NBA. And not only in terms of the number of 50 win teams, but also in terms of the strength of the top 3 (his average only 55.6 wins, I think). Now, of course, anything can happen and Kevin's estimates may indeed dominate others, but I am betting that the formula that generates this much regression to the mean is not a winning one.
DSMok1 wrote:
I would be quite happy to try his or my projections against yours, based on the error for each team. I am quite confident our projections, with the fewer outliers will be closer for each team on average (take the average squared error for each team's win total and we'll see what happens).
How can I resist this challenge! I climb out of my bleacher seat, stand on home plate and point back to where I came from, and confidently say that I will not be defeated*
My formula is the following. Today being October 23, and 23 being my birth day, I predict that the records of this year's teams will equal last year's, regressed to the mean by the factor 0.23.
To this, I will offer some idiosyncratic adjustments, based upon homerism, random beliefs, and distaste for certain players. To anticipate, I will probably put the Cs down at 55 games, the Heat at 64, I think I like Minnesota to exceed expectations, and I intend, in the interests of international comity to show Houston a little love. Unfortunately, I don't have time at the moment to give my complete predictions, but hope to before game time tonight.
And the changes these cause teams' records, I will apportion according to the simple regression formula, in such a straightforward manner that it improves my chance of victory.
Let the games begin!
*The competition being for average error divided by the time spent in generating the estimates.
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 527
Location: Where the wind comes sweeping down the plains
PostPosted: Tue Oct 26, 2010 1:28 pm Post subject: Reply with quote
schtevie: I agree that the actual distribution will have more teams in the 50s. However, we don't know who they are! So, if I have 3 teams that I project to have 48 wins, 1 will probably have 53, 1 will have 47, and another 44. So 48 was about right, but one or two of those 3 will break 50. I just don't know which. If I projected 1 to get 53, 1, to get 47, and 1 to get 44 when my best information has them at about 48, then I will have worse error, because I will likely have the wrong team getting lucky and the wrong team getting unlucky (where luck includes trades, health, and random variation.)
_________________
GodismyJudgeOK.com/DStats
Back to top
View user's profile Send private message Visit poster's website
Crow
Joined: 20 Jan 2009
Posts: 712
PostPosted: Tue Oct 26, 2010 3:29 pm Post subject: Reply with quote
Atlanta 46
Boston 56
Charlotte 35
Chicago 47
Cleveland 37
Dallas 47
Denver 46
Detroit 25
Golden State 38
Houston 43
Indiana 33
LA Clippers 29
LA Lakers 55
Memphis 42
Miami 63
Milwaukee 42
Minnesota 27
New Jersey 25
New Orleans 39
New York 33
Oklahoma City 47
Orlando 61
Philadelphia 29
Phoenix 44
Portland 53
Sacramento 28
San Antonio 50
Toronto 29
Utah 53
Washington 28
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 220
PostPosted: Tue Oct 26, 2010 5:17 pm Post subject: Reply with quote
Crow wrote:
Could you please post a quick summary of the method(s) used.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 712
PostPosted: Tue Oct 26, 2010 5:36 pm Post subject: Reply with quote
I put together a database of 12 predictions (including yours), applied subjective and varying weights based on my sense of their strength and in some cases past performance, found the weighted average, then made a quick and high level review of last season, pre-season, changes and coaching and tweaked the numbers up to 6 wins either way according to on-balance gut feelings.
Last edited by Crow on Wed Oct 27, 2010 12:43 pm; edited 1 time in total
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 385
PostPosted: Wed Oct 27, 2010 10:48 am Post subject: Reply with quote
OK, a day late. These are my picks starting with a baseline of regression to the mean of 0.23 followed by ad hoc adjustments to make average wins come to 41. (Ordered by division and last year's finish within.)
Boston 57
Toronto 30
New York 32
Philadelphia 28
New Jersey 30
Cleveland 41
Milwaukee 45
Chicago 46
Indiana 25
Detroit 20
Orlando 61
Atlanta 50
Miami 64
Charlotte 35
Washington 19
Utah 54
Denver 50
Portland 50
Oklahoma 50
Minnesota 33
Los Angeles 52
Phoenix 42
Clippers 36
Golden State 30
Sacramento 29
Dallas 46
San Antonio 46
Houston 42
Memphis 45
New Orleans 42
P.S. Ack! Epic modeling fail. I looked at the November calendar to determine my regression coefficient. It should have been 0.26. First excuse for failure lined up.
Back to top
View user's profile Send private message
kief
Joined: 11 May 2009
Posts: 5
PostPosted: Wed Oct 27, 2010 2:27 pm Post subject: Reply with quote
schtevie wrote:
OK, a day late. These are my picks starting with a baseline of regression to the mean of 0.23 followed by ad hoc adjustments to make average wins come to 41. (Ordered by division and last year's finish within.)
Boston 57
Toronto 30
New York 32
Philadelphia 28
New Jersey 30
Cleveland 41
Milwaukee 45
Chicago 46
Indiana 25
Detroit 20
Orlando 61
Atlanta 50
Miami 64
Charlotte 35
Washington 19
Utah 54
Denver 50
Portland 50
Oklahoma 50
Minnesota 33
Los Angeles 52
Phoenix 42
Clippers 36
Golden State 30
Sacramento 29
Dallas 46
San Antonio 46
Houston 42
Memphis 45
New Orleans 42
P.S. Ack! Epic modeling fail. I looked at the November calendar to determine my regression coefficient. It should have been 0.26. First excuse for failure lined up.
Is Miami 64, or 54? If the T-Wolves somehow win more games than Golden State, I'll be amazed.
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 385
PostPosted: Wed Oct 27, 2010 4:10 pm Post subject: Reply with quote
No clue about the West, just a few irrational prejudices and the zero sum constraint. But I'm in it to win it.
If only there were recent 6-year APMs available.......oh well.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 712
PostPosted: Wed Oct 27, 2010 4:26 pm Post subject: Reply with quote
I was just checking on what Adjusted +/- might say about the Heat to follow up on that issue.
Using 4 year RAPM (that went thru 2008-9) and basketball reference's minute projection slightly modified for late roster changes, I got a projected 6.0 point team differential. Using the RAPM from the first two-thirds of 2009-10 I got a projected 5.2 point team differential.
If you looked at RAPM trends at player level and applied age adjustments it might vary some from this.
But, first cut, without putting too fine a point on it, these initial estimates would put team wins somewhere between about 52-60. The average of the two (to give more weight to the most recent) would be 56 wins.
APM gets knocked mildly or heavily for fluctuation (and that may be more accurate for comparison of 1 year to 1 year traditional APM) but I'll note that only 2 of 12 players with usable data varied by more than 1 point on RAPM from their previous 4 year average to last season and it was by less than 2 points in each case.
But, despite best efforts, RAPM is still somewhat context influenced and how players perform in the new team offensive and defensive contexts could vary. RAPM (and other metrics and subjective predictions too) probably would do a better job of predicting Miami's 2011-12 team wins off the 2010-11 player data produce mostly in that team context than it will predict 2010-11 team wins off the 2009-10 player data from several contexts.
The most optimistic predictions based off player boxscore stats top out at 68-70 team wins. Take the blended RAPM estimate of 56 and average it with the most optimistic boxscore predictions and you'd get 60-61. Coincidentally in line with the ESPN panel's average. Which pole will it end up closer to?
A prediction of 64 wins for a team would put that team behind 11 others since 1980. Regression back some from the extreme...
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 712
PostPosted: Wed Oct 27, 2010 5:43 pm Post subject: Reply with quote
There are many tools, many answers.
Use Steve Ilardi's old-version of 6 year APM and you'd get a win projection of 65 or a bit more.
The regression built into RAPM for player evaluation may or may not be a good thing for using it to predict at team level.
The 4 year RAPM and the 6 year APM values have a .85 correlation across the rated Heat players, but at expected they vary most at top and bottom and the variance at the top ends up mattering more as they play much more than the bottom guys.
4 year RAPM combined for Wade-James-Bosh is +10.3 per 48 minutes. For 6 year APM, +18.0. How much will diminishing returns impact star player Adjusted +/-? How will it affect the role players? Just one case but an interesting one. Blend these two estimates equally and again you are just a bit above 60 wins.
6 year APM with ridge regression does sound like a good thing to check for veterans. Using it and say 2 year data might help with building decent league level curves for age / experience and by position or role.
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 220
PostPosted: Thu Nov 04, 2010 5:57 pm Post subject: Reply with quote
Code:
Miami Heat 69.92
Orlando Magic 63.39
LosAngeles Lakers 62.5
Denver Nuggets 61.01
NewOrleans Hornets 60.61
PortlandTrail Blazers 54.57
Dallas Mavericks 53.73
Atlanta Hawks 52.29
Boston Celtics 50.52
SanAntonio Spurs 46.53
NewYork Knickerbockers 45.97
Phoenix Suns 45.64
Philadelphia 76ers 43.11
Houston Rockets 42.9
Memphis Grizzlies 41.49
GoldenState Warriors 40.28
Utah Jazz 37.72
Chicago Bulls 36.2
Milwaukee Bucks 34.61
Washington Wizards 34.23
Indiana Pacers 34.18
Toronto Raptors 34.01
Sacramento Kings 29.87
Charlotte Bobcats 26.68
Cleveland Cavaliers 24.03
LosAngeles Clippers 22.56
NewJersey Nets 22.27
Minnesota Timberwolves 21.59
OklahomaCity Thunder 19.37
Detroit Pistons 18.23
interestingly, the majority of teams is already projected by SRS to win roughly as many games as most people thought, with the obvious exceptions being NOH, NYK, Philly, Utah and most notably Oklahoma
Code:
b2n 8.68
Vegas 8.62
JH 8.21
KP 9.43
KD 8.83
DSMOK 7.68
Crow 9.29
schtevie 9.25
page 6
Crow
Joined: 20 Jan 2009
Posts: 821
PostPosted: Sat Nov 06, 2010 12:19 am Post subject: Reply with quote
Before tonight's game:
Wade- Shots down, eFG% close to unchanged, TS% close to unchanged, FTAs down about 10%, usage close to unchanged, TR% down by about a third, assists per game down by a third.
James- Shots down, eFG% down, TS% down, FTAs down about 20%, usage down slightly, TR% down by a third, assists per game down by 1.
Bosh- Shots down, eFG% down, TS% down, FTAs down about 45%, usage down by almost 30%, TR% down by a third, assists per game down by a third.
So far no jumps up for the stars on these stats from playing together. Plenty of givebacks from sharing the ball / court / shot clock / best looks and passing and rebounding opportunities.
At least 5 supporting players are shooting better early though and the team as a whole is second best on eFG%. But it is not other-worldly.
And they are 10th on turnover rate, 27th on OR% and 17th on FT/FGA.
from one of my earlier posts
"... can the Heat offense get to 115- 117.5 on offense? ... In the last 30 years no team has gotten above 116 on offensive efficiency and only 6 of probably 160-170 team seasons (or less than 4%) broke 115. "
After tonight they will move closer toward 109. Denver probably moves past them so the Heat may tick down from 7th best offensive efficiency to 8th.
Very good team defense though. Coaching matters and contributed significantly to that. Didn't expect as much there or as quick. Will be interesting to see what level they post for the season.
Last edited by Crow on Sat Nov 06, 2010 1:05 pm; edited 1 time in total
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Sat Nov 06, 2010 8:50 am Post subject: Reply with quote
Here's what's happened so far for the Heat:
Code:
Player %Min Proj. ASPM ASPM (regr) Updated Change
Dwyane Wade 68.7% 9.04 6.95 8.86 -0.18
LeBron James 73.9% 11.72 6.04 10.90 -0.82
James Jones 56.9% -3.39 2.77 -2.40 0.99
Joel Anthony 40.2% -1.70 2.44 -1.38 0.33
Chris Bosh 66.6% 3.65 1.92 3.39 -0.26
Zydrunas Ilgauskas 33.3% -2.65 1.74 -2.04 0.60
Eddie House 45.8% -2.73 1.26 -1.90 0.83
Mario Chalmers 8.7% -0.51 0.94 -0.48 0.03
Carlos Arroyo 46.5% -2.15 -0.21 -1.96 0.19
Udonis Haslem 48.5% -1.64 -1.32 -1.62 0.02
Juwan Howard 3.5% -3.46 -3.56 -3.47 0.00
Jamaal Magloire 2.8% -4.41 -5.65 -4.45 -0.04
Jerry Stackhouse 4.9% -3.26 -6.75 -3.35 -0.09
Wade, James, and Bosh are all producing below their projections, but all of the secondary players are performing above their expectations. Lebron James, in particular, is performing below expectations. This is not consistent with other superstars of the top 20 in projected ASPM; it is pretty even between those improving on projected and those not.
Since the rest of Miami players are improving their ASPM's and James, Wade, and Bosh are decreasing, I would say that this strongly shows the handicap of ASPM--true APM would pick up on that change for the other players and trace it back to the 3 stars.
Code:
Player Proj. SPM RegrSPM Upd. SPM Change
LeBron James 11.72 6.04 10.90 -0.82
Chris Paul 10.17 12.49 10.36 0.19
Dwyane Wade 9.04 6.95 8.86 -0.18
Kevin Durant 5.86 -1.29 5.22 -0.65
Dwight Howard 5.59 10.39 5.84 0.25
Rajon Rondo 4.78 6.73 4.99 0.21
Kobe Bryant 4.62 6.36 4.77 0.15
Tim Duncan 4.30 4.74 4.33 0.02
Manu Ginobili 4.28 2.91 4.20 -0.08
Brandon Roy 4.23 1.63 3.96 -0.27
Deron Williams 4.20 2.39 4.06 -0.14
Josh Smith 4.10 6.72 4.33 0.23
Dirk Nowitzki 3.93 5.44 4.02 0.09
Pau Gasol 3.72 5.64 3.92 0.19
Chris Bosh 3.65 1.92 3.39 -0.26
Kevin Garnett 3.62 2.70 3.53 -0.09
Gerald Wallace 3.57 -0.67 3.20 -0.36
Al Horford 3.39 5.44 3.54 0.16
Carmelo Anthony 3.35 4.82 3.47 0.12
Nene Hilario 3.19 1.48 3.11 -0.07
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Crow
Joined: 20 Jan 2009
Posts: 821
PostPosted: Sat Nov 13, 2010 4:16 pm Post subject: Reply with quote
It is probably too early to already actually have deep concerns about the 5-4 Heat; but it might be an alright time to start poking around and thinking about performance to date or thinking about how to think about it, so I'll briefly note a few things that could be considered.
Their strength of schedule so far is rated by Sagarin as just a bit above average but not especially tough overall. They do though have the worst record against the top 16 on his strength rating among teams over .500, going 1-4 in those contests. A tough set of top opponents and some close losses, but, those results provide basis for some concern. Orlando game was the only win so far against a strong opponent. It is very early and small sample, but on the range of what they could have done in this segment, it is pretty low.
Because of this, their average point differential against all teams does not comfort me much, given what the real objective is. The results against top teams are likely to improve but will they get to top 2-4 in the league on this or not? Is ending up 3rd, 4th or 6th best here acceptable? Clearly they won't accept staying anywhere near where they are right now.
Add in that Miami under Spoelstra the previous 2 seasons performed expected - weak and weaker than expected against strong opponents respectively. His playoff record of 4-8 is probably about what you'd expect considering, but it is not a positive to lessen the concern. In the playoffs as well as the regular season against strong teams the performance in year 2 was weaker than year 1, in contrast to the overall regular season record change. 2 years isn't much basis for calling a trend but that is what they have to look at so far, if they get anxious.
If weakness against strong opponents continues with the big 3, it might be enough in time, in itself, to reach the tipping point. Sooner or later, depending on GM or owner direction.
Using some of the same overall vs against top team perspective:
Last season among Orlando, Cleveland and Boston, Boston was the clear 3rd if you looked at overall record or overall point differential. But if you looked at record against top 10 teams (relevant for later playoff round match-ups) it was pretty close to a dead heat http://www.usatoday.com/sports/sagarin/nba0910.htm and if you looked at efficiency differential against playoff level teams in the regular season- well, at least using the easily available NBA efficiency data at hoopstat.com- Boston had a slight lead on Cleveland and Orlando http://www.hoopsstats.com/basketball/fa ... iffeff/9-1.
Head to head among the three Orlando had a lead but I think Boston was second. Looking at just April it was the same way. Orlando probably had the edge overall using last regular season data going into last playoffs and actually had the best average point differential in the playoffs but Boston pulled it out, perhaps using their greater experience, something that I would tend to weight favorably in a playoff prediction. Looking at the 2 previous playoffs would have also favored Boston in some eyes. For that 2 year stretch it was Boston 7 conference finals wins, Cleveland 5, Orlando 4. Boston 4 wins in the NBA finals, Orlando 1. Cleveland 0. It might have a fairly tough call to pick Boston first among them to get thru the east last playoffs all things considered, but there were things notably relevant to the playoffs that made them look better than the overall season data.
Performance against the best teams seems to me worth separate and heightened consideration in general for playoff projections.
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Tue Nov 23, 2010 4:38 pm Post subject: Reply with quote
Code:
SanAntonio Spurs 58.26
LosAngeles Lakers 57
NewOrleans Hornets 55.84
Boston Celtics 55.05
Miami Heat 53.91
Dallas Mavericks 51.17
Orlando Magic 51.05
Chicago Bulls 49.33
Denver Nuggets 49.13
Indiana Pacers 46.88
PortlandTrail Blazers 45.71
Utah Jazz 45.13
OklahomaCity Thunder 44.74
Milwaukee Bucks 42.06
Phoenix Suns 41.64
Atlanta Hawks 40.03
Charlotte Bobcats 37.93
Memphis Grizzlies 36.41
Houston Rockets 36.22
Toronto Raptors 36.18
NewYork Knickerbockers 35.71
Philadelphia 76ers 34.57
GoldenState Warriors 32.84
NewJersey Nets 31.41
Cleveland Cavaliers 30.97
Detroit Pistons 28.74
Washington Wizards 26.95
Minnesota Timberwolves 25.17
Sacramento Kings 25.15
LosAngeles Clippers 24.8
Code:
b2n 5.5
Vegas 4.87
JH 5.14
KP 5.85
KD 6.78
Dsmok1 4.53
Crow 7.12
schtevie 6.79
Indiana, Miami and the Hornets differ the most from the initial projections.
I would imagine Erik Spoelstra is in the hot seat right now.
Portland already with more injury-tough-luck than even the more pessimistic people believed they would have.
I'm kind of surprised Orlando is that "bad". They seemed like such a motivated group at the end of last season and this pre-season.
Definitely surprised the Raptors are doing that well
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 821
PostPosted: Wed Nov 24, 2010 3:16 am Post subject: Reply with quote
It is early still, and because it is early, hard to gauge real well by any measure. For a different measure though, using actual win% (as of Sunday) instead of SRS, I was ahead and you were in 2nd. The average error was higher than you show by 2-3.
How SRS or another form of expected win % compares to actual to date win % might be mildly interesting to watch.
SRS probably has the expected edge over actual win % if you are projecting one season off the entire last one but within a season the more accurate measure may trade off between the two at different stages, perhaps several times. At some point near the end (not sure how near) I'd think actual win% to date - on average, if run over many seasons- would pull ahead of expected but I don't know how it looks exactly and it can certainly vary some for any one season and depending on the exact moment in time.
Your good showing on both might bode well, at least it is something that might lead one to guess that way. My widely different rankings are interesting but I could end up either way or in the middle.
In the end it is a contest predicting actual wins not point differential, schedule adjusted. The two are related and in many cases will be close but the actual pattern of variance between the two may play a role in deciding the outcome. It could be that one prediction just does better generally at matching actual win % than it does at matching point differential, schedule adjusted. I was predicting actual wins, not point differential converted to wins. At least some were doing the latter I believe.
At this point data at nbastuffer.com says teams are over or under achieving their point differential by as much as 20% either way and the average absolute difference is probably between 6-8%. Last season for the full season the range was a bit short of +/- 10% and the average absolute difference damped down to about 2-3%. So the differences will become less a factor with time but at this point in the season it could potentially have a fairly significant impact on the rank order of the predictions on both measures.
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Wed Nov 24, 2010 5:43 am Post subject: Reply with quote
The projected errors for now seem super low, but I don't think they will stay that way.
I wouldn't give much thought to these error projections, it only matters who's best in the end. In the meantime they're fun to look at
Torontos' improvement is probably caused by Amir Johnson being awesome (136 ORTG). He's fouling more (per minute) than last year, but still plays a bit more
Al Horford is also playing quite good but Joe Johnson is shooting 41% from the field, 24% from 3-point range
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3613
Location: Hendersonville, NC
PostPosted: Fri Nov 26, 2010 7:03 am Post subject: Reply with quote
If only we could write as good as some guys play.
b2nb, are your projections based on current W-L, plus expected% over the remaining games in the season?
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
bbstats
Joined: 25 Apr 2010
Posts: 46
PostPosted: Thu Dec 02, 2010 10:37 am Post subject: Reply with quote
Now that my numbers are adjusted for recency, and I plugged in the entire NBA schedule instead of just an average schedule, I suppose I could share my results. Mark my words - December 2, 2010 - Miami wins 50 games!
EDIT: Unadjusted for recency, the main difference is a worse Magic and a better Lakers. Knocking the Lakers down this low makes me hesitant...
Code:
rk team wins
1 San Antonio 60.3
2 Dallas 58.9
3 Boston 57.8
4 Utah 55.7
5 New Orleans 55.2
6 Orlando 54.5
7 Denver 51.4
8 Miami 50.2
9 Oklahoma City 49.8
10 LA Lakers 49.6
11 Chicago 48.9
12 Atlanta 46.1
13 Indiana 44.2
14 Phoenix 40.3
15 Portland 40.1
16 New York 39.5
17 Memphis 36.2
18 Toronto 36.1
19 Charlotte 35.9
20 Houston 35.7
21 Milwaukee 33.8
22 Golden State 33.8
23 Philadelphia 33.1
24 Cleveland 30.6
25 New Jersey 30.3
26 Detroit 29.0
27 LA Clippers 27.1
28 Washington 25.4
29 Minnesota 23.8
30 Sacramento 15.6
Game weight = .5*(days into season/most recent day) + .5*most recent day
where n = sum of 'weighted games'
and NBAsdev=SDEV of (residuals of actual and expected point margins)
(\/ do this * vice versa for away yeam)
'capped'.home.adj.win%=1/n * sigma * norm.dist(Home score - Away Score - hca,0,nbaSDEV,1)
reverse log5 = given a win probability and an opponent's home.adj.win%, what their actual adj. win%?
sos&home.adj.win%=rev.log5 (from opponents down to- > opp's opp's opp's opp's*)
rating=norminv(sos&home.adj.win%,0,nbasdev)
I then take the ratings and average team standard deviations and plug them into probabilities against all other 29 teams, home and away, and average this value -- like so
=Normdist(Team1rtg + hca - Team2rtg, 0, sqrt(team1sdev^2+team2sdev^2),1)
^This is the formula I use to predict season-wide win probabilities.^
*I estimate the fourth level to be 50%.
-> and yeah, 1/n * sigma just means the average value
_________________
http://thebasketballdistribution.blogspot.com
http://twitter.com/bbstats
Back to top
View user's profile Send private message Visit poster's website
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Mon Dec 06, 2010 6:12 am Post subject: Reply with quote
The Hornets were 10-1 before the Stojakovic+Bayless for Jarret Jack trade. Since then they're 4-6. The Raptors were at 4-9, then went 4-3.
You just know Jarret Jack won't look this year in (regularized) adj. +/-
Back to top
View user's profile Send private message
BobboFitos
Joined: 21 Feb 2009
Posts: 201
Location: Cambridge, MA
PostPosted: Tue Dec 07, 2010 12:30 am Post subject: Reply with quote
back2newbelf wrote:
The Hornets were 10-1 before the Stojakovic+Bayless for Jarret Jack trade. Since then they're 4-6. The Raptors were at 4-9, then went 4-3.
You just know Jarret Jack won't look this year in (regularized) adj. +/-
Peja has always crushed APM. Maybe he's literally making the Raptors better... from the bench!
_________________
http://pointsperpossession.com/
@PPPBasketball
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Sat Dec 11, 2010 9:19 am Post subject: Reply with quote
Code:
Boston Celtics 58.8
SanAntonio Spurs 58.57
Miami Heat 57
Dallas Mavericks 55.82
LosAngeles Lakers 51.68
Utah Jazz 51.14
Orlando Magic 50
NewOrleans Hornets 48.93
Denver Nuggets 47.92
Chicago Bulls 47.33
OklahomaCity Thunder 46.7
Atlanta Hawks 46.5
Indiana Pacers 45.52
PortlandTrail Blazers 42.9
NewYork Knickerbockers 42.24
Phoenix Suns 40.75
Houston Rockets 38.59
Milwaukee Bucks 38.16
Philadelphia 76ers 36.92
Memphis Grizzlies 36.87
Charlotte Bobcats 35.24
Toronto Raptors 34.32
GoldenState Warriors 31.13
NewJersey Nets 29.23
LosAngeles Clippers 27.76
Minnesota Timberwolves 27.75
Detroit Pistons 26.81
Cleveland Cavaliers 26.35
Washington Wizards 26
Sacramento Kings 23.06
Lakers had the easiest schedule so far, followed by the Kings and Knicks. Golden State had the hardest. Kings have the hardest remaining schedule, Chicago has the easiest.
Roy seems to be playing well again. Considering the news about his knees a couple of of weeks ago that's quite impressive
Code:
b2n 5.4
Vegas 5.1
JH 5.09
KP 6.22
KD 6.29
Dsmok1 4.71
Crow 6.76
schtevie 6.11
WoW 6.99
WS 5.94
SPM(bbr) 6.28
SRS 6.44
“41” 9.07
(lastyear+41)/2 6.29
I have included some more predictions I found on the web, so that next year we have a broader view on what did well and why
This includes the average predictions from the Wages of Wins super stat geek smackdown. They had eight people making predictions. I don't know why their predictions differ, aren't they all using Wins Produced?
Included from bb-r.com are now WS, SPM and SRS.
For reference I also included the error we get when we predict everybody to win 41, and when we predict everyone to win (last years wins+41)/2, which did pretty well last year, and OK the year before
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 821
PostPosted: Sat Dec 11, 2010 6:01 pm Post subject: Reply with quote
back2newbelf could you check your data-entry for my projections? I double-checked my data entry and formulas and I got an estimated error of about 5.1 for my projections using your expected win projections rounded off to the nearest win. I got similar values for the other projections (within a tenth or two, presumably because of the rounding choice) so I think I am doing it properly.
Last edited by Crow on Tue Jan 04, 2011 2:47 pm; edited 5 times in total
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Sat Dec 11, 2010 7:34 pm Post subject: Reply with quote
you're right. I had the Lakers and Clippers switched in your predictions. Sorry for the mix-up. I now have you at 4.99
Does anyone the difference between bbr.com statistical +/- projections and Dsmok1s'?
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 821
PostPosted: Sat Dec 11, 2010 8:14 pm Post subject: Reply with quote
Thanks for the correction. I thought the LA issue was the most likely place for data entry error as I have done that before too.
Neil's and DSMok1's SPMs are based on different regressions with different variable sets and other differences. DSMok1 has threads here explaining his. Neil started his model and explanations here but it is now probably best reviewed in threads at the basketball-reference.com's blog.
The WOW network projections vary from each other because of varying minute projections, whether and how they dealt with player interactions, aging, the team adjustment (and defense beyond that) and perhaps subjective changes beyond their metric base.
Last edited by Crow on Tue Jan 04, 2011 2:48 pm; edited 1 time in total
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Fri Dec 24, 2010 5:40 am Post subject: Reply with quote
Code:
Boston Celtics 61.73
SanAntonio Spurs 58.02
Miami Heat 57.72
Dallas Mavericks 57.47
LosAngeles Lakers 53.71
Chicago Bulls 53.11
Utah Jazz 50.4
OklahomaCity Thunder 48.34
Denver Nuggets 47.52
Orlando Magic 47.5
NewOrleans Hornets 47.01
Atlanta Hawks 45.41
PortlandTrail Blazers 42.81
NewYork Knickerbockers 41.9
Houston Rockets 41.75
Indiana Pacers 41.27
Phoenix Suns 40.35
Milwaukee Bucks 39.28
Memphis Grizzlies 38.19
Philadelphia 76ers 37.88
GoldenState Warriors 31.84
Toronto Raptors 31.23
NewJersey Nets 29.99
Detroit Pistons 29.69
LosAngeles Clippers 29.18
Charlotte Bobcats 28.4
Washington Wizards 27.02
Cleveland Cavaliers 25.45
Minnesota Timberwolves 24.71
Sacramento Kings 21.15
SRS has Boston and Miami head-and-shoulders above everyone else
Code:
b2n 5.81
Vegas 4.93
JH 4.69
KP 6.75
KD 6.37
Dsmok1 4.97
Crow 5.29
schtevie 6.59
WoW 6.91
WS 6.2
SPM(bbr) 6.58
SRS 6.85
“41” 9.31
(lastyear+41)/2 6.85
Joined: 31 Dec 2004
Posts: 1313
Location: Durham, NC
PostPosted: Tue Oct 12, 2010 8:06 pm Post subject: Reply with quote
mtamada wrote:
Sorry, late reply here. My expansion starts with an example: one of the major ways of utilizing regression to the mean is with Stein estimators. The classic example is if it's two weeks into the baseball season and you see a player who's currently batting .412, and you want to predict what his batting average for the rest of the season will be, you do not (if you're smart) use the unbiased estimate of .412. You want to purposely use a biased estimate, and regress to the mean, because extreme performances are unlikely to be sustained.
The key word is extreme. Stein estimators are remarkable because "extreme" and "mean" can be used in a variety of situations; the baseball example is intuitively obvious, not so obvious is that the mean that you regress to doesn't even have to be the mean of baseball players' batting averages (although obviously, the more similar the variables are to each other, the better the Stein estimator will perform, e.g. if you have data on that same player from other seasons, that can be better than using the mean of all other baseball players).
Are you referring to the James-Stein estimator, or Stein's unbiased risk estimate?
Back to top
View user's profile Send private message Send e-mail AIM Address
mtamada
Joined: 28 Jan 2005
Posts: 377
PostPosted: Thu Oct 14, 2010 3:24 am Post subject: Reply with quote
schtevie wrote:
[...]
But invoking "regression to the mean" arguments won't help much in justifying low-ball estimates for the Heat
The lowest of the lowball, yeah sure there are always people who make goofy forecasts, they have their own obscure reasons. Regression to the mean is a moderate correction to apply, it doesn't create weird forecasts -- but it applies to everybody (everybody who's extreme that is; people near the mean can be expected to have minimal regression).
schtevie wrote:
primarily because what appears to be an extreme estimate, isn't really.
Well that is indeed part of the art of forecasting: recognizing what is extreme and what is not. .412 is extremely high for a batter but quite low for an NBA field goal shooter. LBJ is if not the top player in the league, one of the top players. Therefore his forecasts ought to be extreme, yes?
Yes. But they still should be regressed. Every extreme forecast should be. (Though I'd guess that it's easier to wait until the very end and regress Miami's forecasts overall, rather than for each individual player.)
schtevie wrote:
If one is looking for such a reason, it can be found by invoking the assumed concavity of the basketball production function.
In other words, diminishing marginal returns, or in basketball terms, "there's only one ball on the court", i.e. too many superstars start reducing each other's production. The harder question is do they reduce each other's efficiency. I do not know what the models or evidence show. But your original point about the unrelatedness of regression to the mean still applies here: regardless of whether we assume concavity or linearity, the bottom line still holds: if your forecast is an extreme one, and you haven't applied any regression to the mean yet, do so now. The fact that they Heat are loaded with superstars means that they should be forecast for a lot of wins. But beware of a doctrine of exceptionalism, which claims that they are immune from the laws of statistics.
The 2008 Celtics rolled to 66 wins and a championship. But what happened in 2009 and 2010? The 1969 Lakes had a Big 3 that makes the Heat's look puny -- and won 55 games. The 1977 Sixers had perhaps the two best forwards in the league plus 4 former or future All-Stars, and won 50 games. The 1977 Knicks had 4 All-Pros, and won 42 games. The 2004 Lakers had 4 future Hall of Famers, and won 56.
Last edited by mtamada on Thu Oct 14, 2010 3:30 am; edited 1 time in total
Back to top
View user's profile Send private message
mtamada
Joined: 28 Jan 2005
Posts: 377
PostPosted: Thu Oct 14, 2010 3:26 am Post subject: Reply with quote
gabefarkas wrote:
Are you referring to the James-Stein estimator, or Stein's unbiased risk estimate?
The former, I was not familiar with the latter, although apparently it can be used to derive the former.
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 275
PostPosted: Thu Oct 14, 2010 4:53 am Post subject: Reply with quote
http://www.sportsbook.com/livesports/in ... basketball
vegas now with predictions for all the teams.
The Suns and Nuggets are predicted to win substantially less. Miami at 64.5 right now. T-Wolves lowest at 23.5
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 414
PostPosted: Thu Oct 14, 2010 11:59 am Post subject: Reply with quote
Dang. I should have gotten in on the action at 63.5...
Mike, I think that there is value in being more precise when employing the term "regression to the mean" (RTTM). There are two distinct types of phenomena generating the results we see with NBA, league-wide win estimates across seasons, but it is only the first phenomenon that should be meaningfully associated with the term.
The first is statistical, the second for the lack of a better term, is fundamental or model-based (and with this, the zero-sum constraint on the estimation exercise is an important factor).
What I am referring to as "statistical" is the RTTM you observe when, on average, the kids of tall parents don't grow up to be quite as tall. The counterpart in the NBA is when a team because of truly transient good fortune generates wins in excess of what its "genes" would predict, and the following year (all else equal, which it never is) tends to regress to its expected ability.
What I am calling fundamental or model-based phenomena are the effects of all factors that would fit in the "true" NBA production function (that what G*d would reveal, were she so inclined).
One thorny unknown of particular importance in this regard is the precise effect of aging and injuries. But the general story is known and agreed upon. Younger players improve as their skills and knowledge of the game improve. Past their peak, athleticism deteriorates and the propensity for injury increases.
And how does this class of variables (and their variability) influence apparent RTTM? Well, in the ways we are used to expecting, but this result has nothing to do with the "laws of statistics". Apparent RTTM obtains because of the additional fact that better teams tend to have more minutes played by players at their peak, hence more prone to aging and injury effects, and also because the NBA is a zero-sum league (i.e. the games "unexpectedly" lost due to injury and aging, disproportionately suffered by the better teams, pad the win totals of the below-average younger, healthier teams)
As a thought experiment, imagine we lived in an alternate universe where human physiology was different, such that the young rather than the old tended to be injury-prone, and the league had a really hard salary cap instead of a soft one. Much of the apparent RTTM would vanish.
In light of these arguments, we can (kind of) interpret the recent Celtics' history that was mentioned. The last three years saw win totals of 66, 62, and 50 games.
The first issue is: to what win total should the Celtics have rolled in the first year of the Big 3? (The same question as for the Heat this year.) And the answer is, of course, that I don't really know the precise answer. Nobody does, but 66 is probably pretty close to what the "one true model" (approximated best by APM) would have predicted. KG was out for a stretch and played relatively poorly, just prior and after, but, then again, maybe Paul Pierce was playing "above his age" during the season. Etc.
But that emphasizes the main point: the expected value of this production function for the year in question is what matters for a prediction, and no adjustment for RTTM should be made ex post. (If one ends up requiring ad hoc adjustments, it necessarily implies that the model being used to generate predictions is incomplete and giving systematically biased results.)
What then of the following two years? All that can be said is that the trajectory is entirely consistent with fundamentals, but beyond that I have no firm idea as to how they should have been expected to perform. The 62 games seems about right, and last year, my sense is that they endured quite a few "excess" injuries. But I don't have any great confidence about that statement. And what to expect this year? I wish I had a clue. My guess is that they will "regress" to something above 50 wins. And that is the point.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 825
PostPosted: Thu Oct 14, 2010 6:37 pm Post subject: Reply with quote
The Heat line might move again depending on whether Wade & James are considered ready & 100% closer to or on opening day or are still that way 2-4-8 weeks in, if they adjust the line and still take bets. At least in the abstract it should.
Back to top
View user's profile Send private message
Rhuidean
Joined: 11 Mar 2010
Posts: 40
Location: East Bay, CA
PostPosted: Thu Oct 14, 2010 7:17 pm Post subject: . Reply with quote
Schtevie: You've build a model that tells you X. You also have prior knowledge Y about the situation. Unless you have the utmost confidence in your model X and its results, then it makes sense to consider Y also.
From what I've seen (certain ad-hoc down-weightings of APM/SPM values that people used for LBJ, Wade, and Bosh, weighted APM/SPM values that lead to something close to 100% wins), it doesn't seem as if we should have utmost confidence in most of the models just yet. So incorporating prior knowledge makes some sense.
Back to top
View user's profile Send private message Send e-mail
back2newbelf
Joined: 21 Jun 2005
Posts: 275
PostPosted: Sat Oct 23, 2010 8:22 am Post subject: Reply with quote
Can anyone explain me what this vegas line means?
Atlanta Hawks Over 46.5 (-115) Under 46.5 (-105)
specifically the "-115" and "-105".
Obviously they didn't like their initial projection anymore and changed it a bit, but in what direction?
Also, I want to take back my earlier comments about the Knicks. They added more players than I remembered and Stoudemire looks good (at least in pre season)
Back to top
View user's profile Send private message
haralabob
Joined: 11 Apr 2007
Posts: 27
PostPosted: Sat Oct 23, 2010 12:35 pm Post subject: Reply with quote
back2newbelf wrote:
Can anyone explain me what this vegas line means?
Atlanta Hawks Over 46.5 (-115) Under 46.5 (-105)
specifically the "-115" and "-105".
Obviously they didn't like their initial projection anymore and changed it a bit, but in what direction?
Also, I want to take back my earlier comments about the Knicks. They added more players than I remembered and Stoudemire looks good (at least in pre season)
Assume these lines started out as;
Over 46.5 -110
Under 46.5 -110
You'd have to bet $110 to win 100 in profit on either the over or the under.
You know have to bet $115 to win 100 on the over and only $105 to win $100 on the under.
Rather than change their forecast to 47 wins, they move the "juice" to discourage over bettors and encourage people to bet the under.
Back to top
View user's profile Send private message
HoopStudies
Joined: 30 Dec 2004
Posts: 706
Location: Near Philadelphia, PA
PostPosted: Sun Oct 24, 2010 3:40 pm Post subject: Reply with quote
Am I mistaken or did this thread not actually have anyone's actual predictions in it? Wink
_________________
Dean Oliver
Author, Basketball on Paper
The postings are my own & don't necess represent positions, strategies or opinions of employers.
Back to top
View user's profile Send private message Visit poster's website
back2newbelf
Joined: 21 Jun 2005
Posts: 275
PostPosted: Mon Oct 25, 2010 10:55 am Post subject: Reply with quote
Code:
Boston 53
NewJersey 28
Toronto 25
NewYork 37
Philly 27
Cleveland 35
Milwaukee 42
Chicago 45
Detroit 26
Indiana 31
Orlando 59
Miami 63
Washington 28
Atlanta 48
Charlotte 41
Utah 49
Minnesota 27
Denver 46
Oklahoma 51
Portland 52
GoldenState 36
Sacramento 36
Lakers 54
Phoenix 48
Clippers 27
Memphis 39
Houston 40
SA 49
Dallas 52
NO 36
I'll add more comments later today
If I write "player X is good", it means my rating said he was good. Obviously the rating system could have been wrong. I'm just too lazy to write "my rating system blabla" everytime.
"+" means I think it's good for the team (even if they lost a player)
Boston: Lost Rasheed (+)
Cleveland: Vegas has them at 29.5. Absolutely do not agree. Better Coach(+), still a solid starting 5.
Dallas: More Haywood, Chandler (+)
Denver: Vegas has them at 43.5. No idea why
Detroit: Lost their second best player from last year (Jerebko) (-)
Indiana: Watson was strong for them last year (-)
Clippers: Lost Camby (-). I think they would win <20 without Griffin
Lakers: Barnes is good (+). Kobe doesn't look very healthy and Bynum doesn't start the season (-)
Bucks: Lost Ridnour (-). Do not like Chris Douglas Roberts (-)
Wolves: Replaced players in their (by far) two worst positions, PG and SG (Ridnour for Flynn, Johnson for Brewer) (+)
Philly: Lost their best player in Dalembert and brought in a bad players in Hawes and Nocioni(!) (-)
Phoenix: Vegas has them at 41.5. I don't know. Last time everybody thought the Suns would take a huge hit in the standings, they were just fine (back when Stoudemire missed an entire season)
Portland: (Maybe?) More minutes for Batum/Przybilla/Oden/Camby (+)
Sacramento: No Nocioni, more Dalembert, Cousins, Landry (+)
Washington: No Brandan Haywood (-)
The overall method looked like this:
1. Compare last season depth chart with projected depth chart. Look for significant changes in projected production using https://docs.google.com/leaf?id=0An1RF2 ... 1&cindex=8
2. Drift everybody a bit to the middle
3. I had the west winning way more than last year. Adjust for that
4. Adjust for significant changes in division strength
Last edited by back2newbelf on Tue Oct 26, 2010 9:48 am; edited 3 times in total
Back to top
View user's profile Send private message
Ed Küpfer
Joined: 30 Dec 2004
Posts: 787
Location: Toronto
PostPosted: Mon Oct 25, 2010 11:14 am Post subject: Reply with quote
The Painted Area compiled forecasts from Vegas, John Hollinger, Kevin Pelton, and Kelly Dwyer.
Code:
EAST Vegas JH KP KD
MIA 64 66 60 70
ORL 55 55 52 60
BOS 53 51 42 50
CHI 46 48 51 49
ATL 46 46 35 48
MIL 45 49 42 48
CHA 38 30 32 38
NYK 35 37 44 41
PHI 35 43 33 35
IND 34 31 34 28
WAS 33 30 22 31
DET 31 32 27 19
CLE 29 29 39 12
TOR 26 22 35 31
NJN 25 26 40 27
WEST Vegas JH KP KD
LAL 56 56 46 57
OKC 51 49 48 47
POR 51 55 55 50
SAS 50 54 49 55
DAL 50 48 48 52
UTH 48 47 41 52
HOU 47 46 36 46
DEN 43 46 49 50
PHX 41 38 37 52
NOH 40 45 49 44
MEM 38 36 39 40
LAC 36 27 27 38
GSW 32 32 49 21
SAC 28 28 43 26
MIN 23 26 30 13
KevinP hates me and loves David Lewin. That's what I'm getting from this.
_________________
ed
Back to top
View user's profile Send private message Send e-mail
Kevin Pelton
Site Admin
Joined: 30 Dec 2004
Posts: 979
Location: Seattle
PostPosted: Mon Oct 25, 2010 11:53 am Post subject: Reply with quote
Ed Küpfer wrote:
KevinP hates me and loves David Lewin. That's what I'm getting from this.
Yes, the "do I like their analyst?" factor is heavy in my projections. There are few differences, but if we track this again, my final offering is here:
http://www.basketballprospectus.com/art ... cleid=1219
Back to top
View user's profile Send private message Send e-mail Visit poster's website
back2newbelf
Joined: 21 Jun 2005
Posts: 275
PostPosted: Mon Oct 25, 2010 3:43 pm Post subject: Reply with quote
haralabob wrote:
Assume these lines started out as;
Over 46.5 -110
Under 46.5 -110
You'd have to bet $110 to win 100 in profit on either the over or the under.
You know have to bet $115 to win 100 on the over and only $105 to win $100 on the under.
Rather than change their forecast to 47 wins, they move the "juice" to discourage over bettors and encourage people to bet the under.
Thank you.
Does anyone have a "chance that they'll blow their team up/have a firesale to tank for the lottery"-factor in their predictions?
Cleveland and New Orleans (if Paul doesn't return to form) seem like likely candidates
Back to top
View user's profile Send private message
BobboFitos
Joined: 21 Feb 2009
Posts: 201
Location: Cambridge, MA
PostPosted: Tue Oct 26, 2010 2:37 am Post subject: Reply with quote
back2newbelf wrote:
haralabob wrote:
Assume these lines started out as;
Over 46.5 -110
Under 46.5 -110
You'd have to bet $110 to win 100 in profit on either the over or the under.
You know have to bet $115 to win 100 on the over and only $105 to win $100 on the under.
Rather than change their forecast to 47 wins, they move the "juice" to discourage over bettors and encourage people to bet the under.
Thank you.
Does anyone have a "chance that they'll blow their team up/have a firesale to tank for the lottery"-factor in their predictions?
Cleveland and New Orleans (if Paul doesn't return to form) seem like likely candidates
I'm assuming this is what caused the Denver free-fall (they opened at 49.5 I believe and now stand at 43.5)
_________________
http://pointsperpossession.com/
page 5
Author Message
back2newbelf
Joined: 21 Jun 2005
Posts: 220
PostPosted: Tue Oct 26, 2010 9:52 am Post subject: Reply with quote
edited in some minor changes in wins. Hopefully no one gets injured in the next couple of hours
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 385
PostPosted: Tue Oct 26, 2010 11:13 am Post subject: Reply with quote
What's up with Kevin's socialistic, anti-American sentiments? It's one thing for Kevin to hate Ed (Canada is a fine neighbor and all, but it is foreign, and these are tough times) but why does he hate success? Only three teams at 50 wins or more (four if he is willing to round up Chicago's 49.6)? The last three years have seen 12, 9, and 11 such standouts. Gimme a break!
(He's probably in favor of a harder salary cap too.)
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 527
Location: Where the wind comes sweeping down the plains
PostPosted: Tue Oct 26, 2010 11:42 am Post subject: Reply with quote
Advanced SPM projections for the league win totals, adjusting for strength of schedule and rest days:
Code:
TM Margin Rest SoS Total Team Wins
MIA 11.84 -0.11 -0.71 12.45 Miami 68.5
POR 5.57 -0.05 0.10 5.43 Portland 54.9
SAS 4.74 0.10 0.06 4.78 San Antonio 53.4
ORL 3.92 -0.18 -0.29 4.04 Orlando 51.5
ATL 3.57 -0.07 -0.36 3.85 Atlanta 51.1
BOS 3.00 0.02 -0.42 3.44 Boston 50.0
LAL 3.49 -0.18 -0.01 3.32 LA Lakers 49.7
DEN 3.27 -0.11 0.01 3.14 Denver 49.3
CHI 2.31 0.23 -0.47 3.01 Chicago 48.9
UTA 2.78 -0.01 0.11 2.66 Utah 48.0
OKC 2.36 0.12 0.00 2.47 Oklahoma City 47.5
NOH 2.36 0.01 0.21 2.15 New Orleans 46.7
DAL 1.06 0.08 0.24 0.91 Dallas 43.4
PHO 0.45 0.01 0.27 0.19 Phoenix 41.5
GSW 0.28 -0.17 0.14 -0.03 Golden State 40.9
MIL -0.78 0.21 -0.10 -0.47 Milwaukee 39.8
HOU -0.64 -0.04 0.12 -0.80 Houston 38.9
MEM -0.88 0.06 0.16 -0.98 Memphis 38.4
CHA -2.32 0.13 -0.01 -2.19 Charlotte 35.2
PHI -2.46 -0.17 -0.21 -2.41 Philadelphia 34.6
NYK -2.81 -0.19 0.00 -3.01 New York 33.1
CLE -3.51 0.24 -0.05 -3.22 Cleveland 32.5
NJN -3.69 -0.03 -0.08 -3.65 New Jersey 31.4
DET -3.78 -0.04 0.02 -3.84 Detroit 31.0
SAC -4.35 0.11 0.38 -4.62 Sacramento 29.0
IND -4.65 -0.09 -0.09 -4.65 Indiana 29.0
WAS -4.61 0.03 0.12 -4.70 Washington 28.8
LAC -5.09 -0.10 0.30 -5.49 LA Clippers 26.9
TOR -5.69 -0.02 0.05 -5.76 Toronto 26.3
MIN -5.74 0.20 0.52 -6.06 Minnesota 25.6
I personally think there will be significant diminishing returns for Miami; nearly all of the player's contributions above average are on the offensive side of the ball. It would require Miami to maintain a near-historic level of offense (rivaling Dallas in, what was it, 2005?) to actually achieve this level. That's what the projections say, though, so I'll stick to it. I think they may lost a point or 2 of that margin in actuality, but they may be able to compensate with greater effort on defense.
I manually put in Tiago Splitter as a league-average player; all other players without projections come in at replacement level, which I estimate at -3.5 (rather than the commonly accepted -3.0).
Full updated projections here:
Advanced SPM Player and Team Projections
_________________
GodismyJudgeOK.com/DStats
Back to top
View user's profile Send private message Visit poster's website
DSMok1
Joined: 05 Aug 2009
Posts: 527
Location: Where the wind comes sweeping down the plains
PostPosted: Tue Oct 26, 2010 11:46 am Post subject: Reply with quote
schtevie wrote:
What's up with Kevin's socialistic, anti-American sentiments? It's one thing for Kevin to hate Ed (Canada is a fine neighbor and all, but it is foreign, and these are tough times) but why does he hate success? Only three teams at 50 wins or more (four if he is willing to round up Chicago's 49.6)? The last three years have seen 12, 9, and 11 such standouts. Gimme a break!
(He's probably in favor of a harder salary cap too.)
schtevie... it's the maximum likelihood projection, the best estimate for each team.
Which means there will not be nearly as many teams in the 50s, because of regression to the mean. We don't know which teams will break out.
I would be quite happy to try his or my projections against yours, based on the error for each team. I am quite confident our projections, with the fewer outliers will be closer for each team on average (take the average squared error for each team's win total and we'll see what happens).
_________________
GodismyJudgeOK.com/DStats
Back to top
View user's profile Send private message Visit poster's website
schtevie
Joined: 18 Apr 2005
Posts: 385
PostPosted: Tue Oct 26, 2010 1:15 pm Post subject: Reply with quote
DSMok1 wrote:
schtevie wrote:
What's up with Kevin's socialistic, anti-American sentiments? It's one thing for Kevin to hate Ed (Canada is a fine neighbor and all, but it is foreign, and these are tough times) but why does he hate success? Only three teams at 50 wins or more (four if he is willing to round up Chicago's 49.6)? The last three years have seen 12, 9, and 11 such standouts. Gimme a break!
(He's probably in favor of a harder salary cap too.)
schtevie... it's the maximum likelihood projection, the best estimate for each team.
Which means there will not be nearly as many teams in the 50s, because of regression to the mean. We don't know which teams will break out.
You would have to go all the way back to never to find such equality in the NBA. And not only in terms of the number of 50 win teams, but also in terms of the strength of the top 3 (his average only 55.6 wins, I think). Now, of course, anything can happen and Kevin's estimates may indeed dominate others, but I am betting that the formula that generates this much regression to the mean is not a winning one.
DSMok1 wrote:
I would be quite happy to try his or my projections against yours, based on the error for each team. I am quite confident our projections, with the fewer outliers will be closer for each team on average (take the average squared error for each team's win total and we'll see what happens).
How can I resist this challenge! I climb out of my bleacher seat, stand on home plate and point back to where I came from, and confidently say that I will not be defeated*
My formula is the following. Today being October 23, and 23 being my birth day, I predict that the records of this year's teams will equal last year's, regressed to the mean by the factor 0.23.
To this, I will offer some idiosyncratic adjustments, based upon homerism, random beliefs, and distaste for certain players. To anticipate, I will probably put the Cs down at 55 games, the Heat at 64, I think I like Minnesota to exceed expectations, and I intend, in the interests of international comity to show Houston a little love. Unfortunately, I don't have time at the moment to give my complete predictions, but hope to before game time tonight.
And the changes these cause teams' records, I will apportion according to the simple regression formula, in such a straightforward manner that it improves my chance of victory.
Let the games begin!
*The competition being for average error divided by the time spent in generating the estimates.
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 527
Location: Where the wind comes sweeping down the plains
PostPosted: Tue Oct 26, 2010 1:28 pm Post subject: Reply with quote
schtevie: I agree that the actual distribution will have more teams in the 50s. However, we don't know who they are! So, if I have 3 teams that I project to have 48 wins, 1 will probably have 53, 1 will have 47, and another 44. So 48 was about right, but one or two of those 3 will break 50. I just don't know which. If I projected 1 to get 53, 1, to get 47, and 1 to get 44 when my best information has them at about 48, then I will have worse error, because I will likely have the wrong team getting lucky and the wrong team getting unlucky (where luck includes trades, health, and random variation.)
_________________
GodismyJudgeOK.com/DStats
Back to top
View user's profile Send private message Visit poster's website
Crow
Joined: 20 Jan 2009
Posts: 712
PostPosted: Tue Oct 26, 2010 3:29 pm Post subject: Reply with quote
Atlanta 46
Boston 56
Charlotte 35
Chicago 47
Cleveland 37
Dallas 47
Denver 46
Detroit 25
Golden State 38
Houston 43
Indiana 33
LA Clippers 29
LA Lakers 55
Memphis 42
Miami 63
Milwaukee 42
Minnesota 27
New Jersey 25
New Orleans 39
New York 33
Oklahoma City 47
Orlando 61
Philadelphia 29
Phoenix 44
Portland 53
Sacramento 28
San Antonio 50
Toronto 29
Utah 53
Washington 28
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 220
PostPosted: Tue Oct 26, 2010 5:17 pm Post subject: Reply with quote
Crow wrote:
Could you please post a quick summary of the method(s) used.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 712
PostPosted: Tue Oct 26, 2010 5:36 pm Post subject: Reply with quote
I put together a database of 12 predictions (including yours), applied subjective and varying weights based on my sense of their strength and in some cases past performance, found the weighted average, then made a quick and high level review of last season, pre-season, changes and coaching and tweaked the numbers up to 6 wins either way according to on-balance gut feelings.
Last edited by Crow on Wed Oct 27, 2010 12:43 pm; edited 1 time in total
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 385
PostPosted: Wed Oct 27, 2010 10:48 am Post subject: Reply with quote
OK, a day late. These are my picks starting with a baseline of regression to the mean of 0.23 followed by ad hoc adjustments to make average wins come to 41. (Ordered by division and last year's finish within.)
Boston 57
Toronto 30
New York 32
Philadelphia 28
New Jersey 30
Cleveland 41
Milwaukee 45
Chicago 46
Indiana 25
Detroit 20
Orlando 61
Atlanta 50
Miami 64
Charlotte 35
Washington 19
Utah 54
Denver 50
Portland 50
Oklahoma 50
Minnesota 33
Los Angeles 52
Phoenix 42
Clippers 36
Golden State 30
Sacramento 29
Dallas 46
San Antonio 46
Houston 42
Memphis 45
New Orleans 42
P.S. Ack! Epic modeling fail. I looked at the November calendar to determine my regression coefficient. It should have been 0.26. First excuse for failure lined up.
Back to top
View user's profile Send private message
kief
Joined: 11 May 2009
Posts: 5
PostPosted: Wed Oct 27, 2010 2:27 pm Post subject: Reply with quote
schtevie wrote:
OK, a day late. These are my picks starting with a baseline of regression to the mean of 0.23 followed by ad hoc adjustments to make average wins come to 41. (Ordered by division and last year's finish within.)
Boston 57
Toronto 30
New York 32
Philadelphia 28
New Jersey 30
Cleveland 41
Milwaukee 45
Chicago 46
Indiana 25
Detroit 20
Orlando 61
Atlanta 50
Miami 64
Charlotte 35
Washington 19
Utah 54
Denver 50
Portland 50
Oklahoma 50
Minnesota 33
Los Angeles 52
Phoenix 42
Clippers 36
Golden State 30
Sacramento 29
Dallas 46
San Antonio 46
Houston 42
Memphis 45
New Orleans 42
P.S. Ack! Epic modeling fail. I looked at the November calendar to determine my regression coefficient. It should have been 0.26. First excuse for failure lined up.
Is Miami 64, or 54? If the T-Wolves somehow win more games than Golden State, I'll be amazed.
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 385
PostPosted: Wed Oct 27, 2010 4:10 pm Post subject: Reply with quote
No clue about the West, just a few irrational prejudices and the zero sum constraint. But I'm in it to win it.
If only there were recent 6-year APMs available.......oh well.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 712
PostPosted: Wed Oct 27, 2010 4:26 pm Post subject: Reply with quote
I was just checking on what Adjusted +/- might say about the Heat to follow up on that issue.
Using 4 year RAPM (that went thru 2008-9) and basketball reference's minute projection slightly modified for late roster changes, I got a projected 6.0 point team differential. Using the RAPM from the first two-thirds of 2009-10 I got a projected 5.2 point team differential.
If you looked at RAPM trends at player level and applied age adjustments it might vary some from this.
But, first cut, without putting too fine a point on it, these initial estimates would put team wins somewhere between about 52-60. The average of the two (to give more weight to the most recent) would be 56 wins.
APM gets knocked mildly or heavily for fluctuation (and that may be more accurate for comparison of 1 year to 1 year traditional APM) but I'll note that only 2 of 12 players with usable data varied by more than 1 point on RAPM from their previous 4 year average to last season and it was by less than 2 points in each case.
But, despite best efforts, RAPM is still somewhat context influenced and how players perform in the new team offensive and defensive contexts could vary. RAPM (and other metrics and subjective predictions too) probably would do a better job of predicting Miami's 2011-12 team wins off the 2010-11 player data produce mostly in that team context than it will predict 2010-11 team wins off the 2009-10 player data from several contexts.
The most optimistic predictions based off player boxscore stats top out at 68-70 team wins. Take the blended RAPM estimate of 56 and average it with the most optimistic boxscore predictions and you'd get 60-61. Coincidentally in line with the ESPN panel's average. Which pole will it end up closer to?
A prediction of 64 wins for a team would put that team behind 11 others since 1980. Regression back some from the extreme...
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 712
PostPosted: Wed Oct 27, 2010 5:43 pm Post subject: Reply with quote
There are many tools, many answers.
Use Steve Ilardi's old-version of 6 year APM and you'd get a win projection of 65 or a bit more.
The regression built into RAPM for player evaluation may or may not be a good thing for using it to predict at team level.
The 4 year RAPM and the 6 year APM values have a .85 correlation across the rated Heat players, but at expected they vary most at top and bottom and the variance at the top ends up mattering more as they play much more than the bottom guys.
4 year RAPM combined for Wade-James-Bosh is +10.3 per 48 minutes. For 6 year APM, +18.0. How much will diminishing returns impact star player Adjusted +/-? How will it affect the role players? Just one case but an interesting one. Blend these two estimates equally and again you are just a bit above 60 wins.
6 year APM with ridge regression does sound like a good thing to check for veterans. Using it and say 2 year data might help with building decent league level curves for age / experience and by position or role.
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 220
PostPosted: Thu Nov 04, 2010 5:57 pm Post subject: Reply with quote
Code:
Miami Heat 69.92
Orlando Magic 63.39
LosAngeles Lakers 62.5
Denver Nuggets 61.01
NewOrleans Hornets 60.61
PortlandTrail Blazers 54.57
Dallas Mavericks 53.73
Atlanta Hawks 52.29
Boston Celtics 50.52
SanAntonio Spurs 46.53
NewYork Knickerbockers 45.97
Phoenix Suns 45.64
Philadelphia 76ers 43.11
Houston Rockets 42.9
Memphis Grizzlies 41.49
GoldenState Warriors 40.28
Utah Jazz 37.72
Chicago Bulls 36.2
Milwaukee Bucks 34.61
Washington Wizards 34.23
Indiana Pacers 34.18
Toronto Raptors 34.01
Sacramento Kings 29.87
Charlotte Bobcats 26.68
Cleveland Cavaliers 24.03
LosAngeles Clippers 22.56
NewJersey Nets 22.27
Minnesota Timberwolves 21.59
OklahomaCity Thunder 19.37
Detroit Pistons 18.23
interestingly, the majority of teams is already projected by SRS to win roughly as many games as most people thought, with the obvious exceptions being NOH, NYK, Philly, Utah and most notably Oklahoma
Code:
b2n 8.68
Vegas 8.62
JH 8.21
KP 9.43
KD 8.83
DSMOK 7.68
Crow 9.29
schtevie 9.25
page 6
Crow
Joined: 20 Jan 2009
Posts: 821
PostPosted: Sat Nov 06, 2010 12:19 am Post subject: Reply with quote
Before tonight's game:
Wade- Shots down, eFG% close to unchanged, TS% close to unchanged, FTAs down about 10%, usage close to unchanged, TR% down by about a third, assists per game down by a third.
James- Shots down, eFG% down, TS% down, FTAs down about 20%, usage down slightly, TR% down by a third, assists per game down by 1.
Bosh- Shots down, eFG% down, TS% down, FTAs down about 45%, usage down by almost 30%, TR% down by a third, assists per game down by a third.
So far no jumps up for the stars on these stats from playing together. Plenty of givebacks from sharing the ball / court / shot clock / best looks and passing and rebounding opportunities.
At least 5 supporting players are shooting better early though and the team as a whole is second best on eFG%. But it is not other-worldly.
And they are 10th on turnover rate, 27th on OR% and 17th on FT/FGA.
from one of my earlier posts
"... can the Heat offense get to 115- 117.5 on offense? ... In the last 30 years no team has gotten above 116 on offensive efficiency and only 6 of probably 160-170 team seasons (or less than 4%) broke 115. "
After tonight they will move closer toward 109. Denver probably moves past them so the Heat may tick down from 7th best offensive efficiency to 8th.
Very good team defense though. Coaching matters and contributed significantly to that. Didn't expect as much there or as quick. Will be interesting to see what level they post for the season.
Last edited by Crow on Sat Nov 06, 2010 1:05 pm; edited 1 time in total
Back to top
View user's profile Send private message
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
PostPosted: Sat Nov 06, 2010 8:50 am Post subject: Reply with quote
Here's what's happened so far for the Heat:
Code:
Player %Min Proj. ASPM ASPM (regr) Updated Change
Dwyane Wade 68.7% 9.04 6.95 8.86 -0.18
LeBron James 73.9% 11.72 6.04 10.90 -0.82
James Jones 56.9% -3.39 2.77 -2.40 0.99
Joel Anthony 40.2% -1.70 2.44 -1.38 0.33
Chris Bosh 66.6% 3.65 1.92 3.39 -0.26
Zydrunas Ilgauskas 33.3% -2.65 1.74 -2.04 0.60
Eddie House 45.8% -2.73 1.26 -1.90 0.83
Mario Chalmers 8.7% -0.51 0.94 -0.48 0.03
Carlos Arroyo 46.5% -2.15 -0.21 -1.96 0.19
Udonis Haslem 48.5% -1.64 -1.32 -1.62 0.02
Juwan Howard 3.5% -3.46 -3.56 -3.47 0.00
Jamaal Magloire 2.8% -4.41 -5.65 -4.45 -0.04
Jerry Stackhouse 4.9% -3.26 -6.75 -3.35 -0.09
Wade, James, and Bosh are all producing below their projections, but all of the secondary players are performing above their expectations. Lebron James, in particular, is performing below expectations. This is not consistent with other superstars of the top 20 in projected ASPM; it is pretty even between those improving on projected and those not.
Since the rest of Miami players are improving their ASPM's and James, Wade, and Bosh are decreasing, I would say that this strongly shows the handicap of ASPM--true APM would pick up on that change for the other players and trace it back to the 3 stars.
Code:
Player Proj. SPM RegrSPM Upd. SPM Change
LeBron James 11.72 6.04 10.90 -0.82
Chris Paul 10.17 12.49 10.36 0.19
Dwyane Wade 9.04 6.95 8.86 -0.18
Kevin Durant 5.86 -1.29 5.22 -0.65
Dwight Howard 5.59 10.39 5.84 0.25
Rajon Rondo 4.78 6.73 4.99 0.21
Kobe Bryant 4.62 6.36 4.77 0.15
Tim Duncan 4.30 4.74 4.33 0.02
Manu Ginobili 4.28 2.91 4.20 -0.08
Brandon Roy 4.23 1.63 3.96 -0.27
Deron Williams 4.20 2.39 4.06 -0.14
Josh Smith 4.10 6.72 4.33 0.23
Dirk Nowitzki 3.93 5.44 4.02 0.09
Pau Gasol 3.72 5.64 3.92 0.19
Chris Bosh 3.65 1.92 3.39 -0.26
Kevin Garnett 3.62 2.70 3.53 -0.09
Gerald Wallace 3.57 -0.67 3.20 -0.36
Al Horford 3.39 5.44 3.54 0.16
Carmelo Anthony 3.35 4.82 3.47 0.12
Nene Hilario 3.19 1.48 3.11 -0.07
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Crow
Joined: 20 Jan 2009
Posts: 821
PostPosted: Sat Nov 13, 2010 4:16 pm Post subject: Reply with quote
It is probably too early to already actually have deep concerns about the 5-4 Heat; but it might be an alright time to start poking around and thinking about performance to date or thinking about how to think about it, so I'll briefly note a few things that could be considered.
Their strength of schedule so far is rated by Sagarin as just a bit above average but not especially tough overall. They do though have the worst record against the top 16 on his strength rating among teams over .500, going 1-4 in those contests. A tough set of top opponents and some close losses, but, those results provide basis for some concern. Orlando game was the only win so far against a strong opponent. It is very early and small sample, but on the range of what they could have done in this segment, it is pretty low.
Because of this, their average point differential against all teams does not comfort me much, given what the real objective is. The results against top teams are likely to improve but will they get to top 2-4 in the league on this or not? Is ending up 3rd, 4th or 6th best here acceptable? Clearly they won't accept staying anywhere near where they are right now.
Add in that Miami under Spoelstra the previous 2 seasons performed expected - weak and weaker than expected against strong opponents respectively. His playoff record of 4-8 is probably about what you'd expect considering, but it is not a positive to lessen the concern. In the playoffs as well as the regular season against strong teams the performance in year 2 was weaker than year 1, in contrast to the overall regular season record change. 2 years isn't much basis for calling a trend but that is what they have to look at so far, if they get anxious.
If weakness against strong opponents continues with the big 3, it might be enough in time, in itself, to reach the tipping point. Sooner or later, depending on GM or owner direction.
Using some of the same overall vs against top team perspective:
Last season among Orlando, Cleveland and Boston, Boston was the clear 3rd if you looked at overall record or overall point differential. But if you looked at record against top 10 teams (relevant for later playoff round match-ups) it was pretty close to a dead heat http://www.usatoday.com/sports/sagarin/nba0910.htm and if you looked at efficiency differential against playoff level teams in the regular season- well, at least using the easily available NBA efficiency data at hoopstat.com- Boston had a slight lead on Cleveland and Orlando http://www.hoopsstats.com/basketball/fa ... iffeff/9-1.
Head to head among the three Orlando had a lead but I think Boston was second. Looking at just April it was the same way. Orlando probably had the edge overall using last regular season data going into last playoffs and actually had the best average point differential in the playoffs but Boston pulled it out, perhaps using their greater experience, something that I would tend to weight favorably in a playoff prediction. Looking at the 2 previous playoffs would have also favored Boston in some eyes. For that 2 year stretch it was Boston 7 conference finals wins, Cleveland 5, Orlando 4. Boston 4 wins in the NBA finals, Orlando 1. Cleveland 0. It might have a fairly tough call to pick Boston first among them to get thru the east last playoffs all things considered, but there were things notably relevant to the playoffs that made them look better than the overall season data.
Performance against the best teams seems to me worth separate and heightened consideration in general for playoff projections.
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Tue Nov 23, 2010 4:38 pm Post subject: Reply with quote
Code:
SanAntonio Spurs 58.26
LosAngeles Lakers 57
NewOrleans Hornets 55.84
Boston Celtics 55.05
Miami Heat 53.91
Dallas Mavericks 51.17
Orlando Magic 51.05
Chicago Bulls 49.33
Denver Nuggets 49.13
Indiana Pacers 46.88
PortlandTrail Blazers 45.71
Utah Jazz 45.13
OklahomaCity Thunder 44.74
Milwaukee Bucks 42.06
Phoenix Suns 41.64
Atlanta Hawks 40.03
Charlotte Bobcats 37.93
Memphis Grizzlies 36.41
Houston Rockets 36.22
Toronto Raptors 36.18
NewYork Knickerbockers 35.71
Philadelphia 76ers 34.57
GoldenState Warriors 32.84
NewJersey Nets 31.41
Cleveland Cavaliers 30.97
Detroit Pistons 28.74
Washington Wizards 26.95
Minnesota Timberwolves 25.17
Sacramento Kings 25.15
LosAngeles Clippers 24.8
Code:
b2n 5.5
Vegas 4.87
JH 5.14
KP 5.85
KD 6.78
Dsmok1 4.53
Crow 7.12
schtevie 6.79
Indiana, Miami and the Hornets differ the most from the initial projections.
I would imagine Erik Spoelstra is in the hot seat right now.
Portland already with more injury-tough-luck than even the more pessimistic people believed they would have.
I'm kind of surprised Orlando is that "bad". They seemed like such a motivated group at the end of last season and this pre-season.
Definitely surprised the Raptors are doing that well
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 821
PostPosted: Wed Nov 24, 2010 3:16 am Post subject: Reply with quote
It is early still, and because it is early, hard to gauge real well by any measure. For a different measure though, using actual win% (as of Sunday) instead of SRS, I was ahead and you were in 2nd. The average error was higher than you show by 2-3.
How SRS or another form of expected win % compares to actual to date win % might be mildly interesting to watch.
SRS probably has the expected edge over actual win % if you are projecting one season off the entire last one but within a season the more accurate measure may trade off between the two at different stages, perhaps several times. At some point near the end (not sure how near) I'd think actual win% to date - on average, if run over many seasons- would pull ahead of expected but I don't know how it looks exactly and it can certainly vary some for any one season and depending on the exact moment in time.
Your good showing on both might bode well, at least it is something that might lead one to guess that way. My widely different rankings are interesting but I could end up either way or in the middle.
In the end it is a contest predicting actual wins not point differential, schedule adjusted. The two are related and in many cases will be close but the actual pattern of variance between the two may play a role in deciding the outcome. It could be that one prediction just does better generally at matching actual win % than it does at matching point differential, schedule adjusted. I was predicting actual wins, not point differential converted to wins. At least some were doing the latter I believe.
At this point data at nbastuffer.com says teams are over or under achieving their point differential by as much as 20% either way and the average absolute difference is probably between 6-8%. Last season for the full season the range was a bit short of +/- 10% and the average absolute difference damped down to about 2-3%. So the differences will become less a factor with time but at this point in the season it could potentially have a fairly significant impact on the rank order of the predictions on both measures.
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Wed Nov 24, 2010 5:43 am Post subject: Reply with quote
The projected errors for now seem super low, but I don't think they will stay that way.
I wouldn't give much thought to these error projections, it only matters who's best in the end. In the meantime they're fun to look at
Torontos' improvement is probably caused by Amir Johnson being awesome (136 ORTG). He's fouling more (per minute) than last year, but still plays a bit more
Al Horford is also playing quite good but Joe Johnson is shooting 41% from the field, 24% from 3-point range
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3613
Location: Hendersonville, NC
PostPosted: Fri Nov 26, 2010 7:03 am Post subject: Reply with quote
If only we could write as good as some guys play.
b2nb, are your projections based on current W-L, plus expected% over the remaining games in the season?
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
bbstats
Joined: 25 Apr 2010
Posts: 46
PostPosted: Thu Dec 02, 2010 10:37 am Post subject: Reply with quote
Now that my numbers are adjusted for recency, and I plugged in the entire NBA schedule instead of just an average schedule, I suppose I could share my results. Mark my words - December 2, 2010 - Miami wins 50 games!
EDIT: Unadjusted for recency, the main difference is a worse Magic and a better Lakers. Knocking the Lakers down this low makes me hesitant...
Code:
rk team wins
1 San Antonio 60.3
2 Dallas 58.9
3 Boston 57.8
4 Utah 55.7
5 New Orleans 55.2
6 Orlando 54.5
7 Denver 51.4
8 Miami 50.2
9 Oklahoma City 49.8
10 LA Lakers 49.6
11 Chicago 48.9
12 Atlanta 46.1
13 Indiana 44.2
14 Phoenix 40.3
15 Portland 40.1
16 New York 39.5
17 Memphis 36.2
18 Toronto 36.1
19 Charlotte 35.9
20 Houston 35.7
21 Milwaukee 33.8
22 Golden State 33.8
23 Philadelphia 33.1
24 Cleveland 30.6
25 New Jersey 30.3
26 Detroit 29.0
27 LA Clippers 27.1
28 Washington 25.4
29 Minnesota 23.8
30 Sacramento 15.6
Game weight = .5*(days into season/most recent day) + .5*most recent day
where n = sum of 'weighted games'
and NBAsdev=SDEV of (residuals of actual and expected point margins)
(\/ do this * vice versa for away yeam)
'capped'.home.adj.win%=1/n * sigma * norm.dist(Home score - Away Score - hca,0,nbaSDEV,1)
reverse log5 = given a win probability and an opponent's home.adj.win%, what their actual adj. win%?
sos&home.adj.win%=rev.log5 (from opponents down to- > opp's opp's opp's opp's*)
rating=norminv(sos&home.adj.win%,0,nbasdev)
I then take the ratings and average team standard deviations and plug them into probabilities against all other 29 teams, home and away, and average this value -- like so
=Normdist(Team1rtg + hca - Team2rtg, 0, sqrt(team1sdev^2+team2sdev^2),1)
^This is the formula I use to predict season-wide win probabilities.^
*I estimate the fourth level to be 50%.
-> and yeah, 1/n * sigma just means the average value
_________________
http://thebasketballdistribution.blogspot.com
http://twitter.com/bbstats
Back to top
View user's profile Send private message Visit poster's website
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Mon Dec 06, 2010 6:12 am Post subject: Reply with quote
The Hornets were 10-1 before the Stojakovic+Bayless for Jarret Jack trade. Since then they're 4-6. The Raptors were at 4-9, then went 4-3.
You just know Jarret Jack won't look this year in (regularized) adj. +/-
Back to top
View user's profile Send private message
BobboFitos
Joined: 21 Feb 2009
Posts: 201
Location: Cambridge, MA
PostPosted: Tue Dec 07, 2010 12:30 am Post subject: Reply with quote
back2newbelf wrote:
The Hornets were 10-1 before the Stojakovic+Bayless for Jarret Jack trade. Since then they're 4-6. The Raptors were at 4-9, then went 4-3.
You just know Jarret Jack won't look this year in (regularized) adj. +/-
Peja has always crushed APM. Maybe he's literally making the Raptors better... from the bench!
_________________
http://pointsperpossession.com/
@PPPBasketball
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Sat Dec 11, 2010 9:19 am Post subject: Reply with quote
Code:
Boston Celtics 58.8
SanAntonio Spurs 58.57
Miami Heat 57
Dallas Mavericks 55.82
LosAngeles Lakers 51.68
Utah Jazz 51.14
Orlando Magic 50
NewOrleans Hornets 48.93
Denver Nuggets 47.92
Chicago Bulls 47.33
OklahomaCity Thunder 46.7
Atlanta Hawks 46.5
Indiana Pacers 45.52
PortlandTrail Blazers 42.9
NewYork Knickerbockers 42.24
Phoenix Suns 40.75
Houston Rockets 38.59
Milwaukee Bucks 38.16
Philadelphia 76ers 36.92
Memphis Grizzlies 36.87
Charlotte Bobcats 35.24
Toronto Raptors 34.32
GoldenState Warriors 31.13
NewJersey Nets 29.23
LosAngeles Clippers 27.76
Minnesota Timberwolves 27.75
Detroit Pistons 26.81
Cleveland Cavaliers 26.35
Washington Wizards 26
Sacramento Kings 23.06
Lakers had the easiest schedule so far, followed by the Kings and Knicks. Golden State had the hardest. Kings have the hardest remaining schedule, Chicago has the easiest.
Roy seems to be playing well again. Considering the news about his knees a couple of of weeks ago that's quite impressive
Code:
b2n 5.4
Vegas 5.1
JH 5.09
KP 6.22
KD 6.29
Dsmok1 4.71
Crow 6.76
schtevie 6.11
WoW 6.99
WS 5.94
SPM(bbr) 6.28
SRS 6.44
“41” 9.07
(lastyear+41)/2 6.29
I have included some more predictions I found on the web, so that next year we have a broader view on what did well and why
This includes the average predictions from the Wages of Wins super stat geek smackdown. They had eight people making predictions. I don't know why their predictions differ, aren't they all using Wins Produced?
Included from bb-r.com are now WS, SPM and SRS.
For reference I also included the error we get when we predict everybody to win 41, and when we predict everyone to win (last years wins+41)/2, which did pretty well last year, and OK the year before
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 821
PostPosted: Sat Dec 11, 2010 6:01 pm Post subject: Reply with quote
back2newbelf could you check your data-entry for my projections? I double-checked my data entry and formulas and I got an estimated error of about 5.1 for my projections using your expected win projections rounded off to the nearest win. I got similar values for the other projections (within a tenth or two, presumably because of the rounding choice) so I think I am doing it properly.
Last edited by Crow on Tue Jan 04, 2011 2:47 pm; edited 5 times in total
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Sat Dec 11, 2010 7:34 pm Post subject: Reply with quote
you're right. I had the Lakers and Clippers switched in your predictions. Sorry for the mix-up. I now have you at 4.99
Does anyone the difference between bbr.com statistical +/- projections and Dsmok1s'?
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 821
PostPosted: Sat Dec 11, 2010 8:14 pm Post subject: Reply with quote
Thanks for the correction. I thought the LA issue was the most likely place for data entry error as I have done that before too.
Neil's and DSMok1's SPMs are based on different regressions with different variable sets and other differences. DSMok1 has threads here explaining his. Neil started his model and explanations here but it is now probably best reviewed in threads at the basketball-reference.com's blog.
The WOW network projections vary from each other because of varying minute projections, whether and how they dealt with player interactions, aging, the team adjustment (and defense beyond that) and perhaps subjective changes beyond their metric base.
Last edited by Crow on Tue Jan 04, 2011 2:48 pm; edited 1 time in total
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 274
PostPosted: Fri Dec 24, 2010 5:40 am Post subject: Reply with quote
Code:
Boston Celtics 61.73
SanAntonio Spurs 58.02
Miami Heat 57.72
Dallas Mavericks 57.47
LosAngeles Lakers 53.71
Chicago Bulls 53.11
Utah Jazz 50.4
OklahomaCity Thunder 48.34
Denver Nuggets 47.52
Orlando Magic 47.5
NewOrleans Hornets 47.01
Atlanta Hawks 45.41
PortlandTrail Blazers 42.81
NewYork Knickerbockers 41.9
Houston Rockets 41.75
Indiana Pacers 41.27
Phoenix Suns 40.35
Milwaukee Bucks 39.28
Memphis Grizzlies 38.19
Philadelphia 76ers 37.88
GoldenState Warriors 31.84
Toronto Raptors 31.23
NewJersey Nets 29.99
Detroit Pistons 29.69
LosAngeles Clippers 29.18
Charlotte Bobcats 28.4
Washington Wizards 27.02
Cleveland Cavaliers 25.45
Minnesota Timberwolves 24.71
Sacramento Kings 21.15
SRS has Boston and Miami head-and-shoulders above everyone else
Code:
b2n 5.81
Vegas 4.93
JH 4.69
KP 6.75
KD 6.37
Dsmok1 4.97
Crow 5.29
schtevie 6.59
WoW 6.91
WS 6.2
SPM(bbr) 6.58
SRS 6.85
“41” 9.31
(lastyear+41)/2 6.85
Last edited by Crow on Thu May 12, 2011 5:53 pm, edited 1 time in total.
Re: 2010-11 NBA Win Predictions
page 7
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Sun Jan 02, 2011 11:26 am Post subject: Reply with quote
Happy new year everyone
Code:
SanAntonio Spurs 60.97
Miami Heat 59.04
Boston Celtics 58.7
Dallas Mavericks 54.67
LosAngeles Lakers 53.01
Chicago Bulls 52.76
Orlando Magic 50.25
Utah Jazz 49.6
OklahomaCity Thunder 48.02
Denver Nuggets 47.39
NewOrleans Hornets 46.73
Atlanta Hawks 45.98
PortlandTrail Blazers 42.84
NewYork Knickerbockers 42.72
Houston Rockets 41.83
Phoenix Suns 39.48
Philadelphia 76ers 38.7
Memphis Grizzlies 38.68
Milwaukee Bucks 38.42
Indiana Pacers 38.28
GoldenState Warriors 34.45
Toronto Raptors 32.07
Detroit Pistons 30.07
LosAngeles Clippers 29.85
Charlotte Bobcats 28.85
NewJersey Nets 27.64
Minnesota Timberwolves 27.17
Washington Wizards 26.97
Cleveland Cavaliers 23.81
Sacramento Kings 21.04
The Celtics are having a hard time without Garnett. Same with Dallas and Nowitzki. Indiana has cooled off quite a bit. Warriors, Suns and Bucks had the hardest schedule, Kings, Spurs and Heat the easiest. Kings, Bobcats and Clippers with the hardest remaining schedule, 76ers, Bucks and Celtics with the easiest
Code:
b2n 5.43
Vegas 4.8
JH 4.41
KP 6.33
KD 6.18
Dsmok1 4.75
Crow 4.92
schtevie 6.47
WoW 6.86
WS 6.11
SPM(bbr) 6.5
SRS 6.6
“41” 9.3
(lastyear+41)/2 6.42
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 246
PostPosted: Sun Jan 02, 2011 11:42 am Post subject: Reply with quote
Vegas, huh? Are those based on pre-season odds?
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Sun Jan 02, 2011 11:45 am Post subject: Reply with quote
pre-season over-under
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 246
PostPosted: Sun Jan 02, 2011 12:14 pm Post subject: Reply with quote
back2newbelf wrote:
pre-season over-under
That's impressive. Do they have their own stat guys calculate this? Or is it based on how people bet?
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Mon Jan 03, 2011 5:51 am Post subject: Reply with quote
EvanZ wrote:
That's impressive. Do they have their own stat guys calculate this? Or is it based on how people bet?
If it was just the latter I would imagine more of us would be sitting at some Hawaian beach with a cocktail in their hand, instead of posting on internet message boards from somewhere cold.
Sacramentos' schedule from Jan 28th until Feb 23rd : @LAL, NOH, BOS, SA, UTA, DAL, OKC, @PHX, @OKC, @DAL, @MIA, @ORL
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 246
PostPosted: Mon Jan 03, 2011 7:49 am Post subject: Reply with quote
back2newbelf wrote:
If it was just the latter I would imagine more of us would be sitting at some Hawaian beach with a cocktail in their hand, instead of posting on internet message boards from somewhere cold.
If that were an option, I have to think I wouldn't be here either. Laughing
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Sun Jan 16, 2011 5:33 am Post subject: Reply with quote
Code:
SanAntonio Spurs 61.2
Boston Celtics 57.87
Miami Heat 57.35
LosAngeles Lakers 55.3
Chicago Bulls 52.85
Orlando Magic 51.81
Dallas Mavericks 50.65
OklahomaCity Thunder 49.9
Utah Jazz 49.05
Atlanta Hawks 48.36
NewOrleans Hornets 48.18
Denver Nuggets 47.06
NewYork Knickerbockers 44.3
PortlandTrail Blazers 42.77
Memphis Grizzlies 41.21
Houston Rockets 39.57
Indiana Pacers 38.01
Milwaukee Bucks 37.93
Philadelphia 76ers 37.85
Phoenix Suns 36.82
GoldenState Warriors 35.22
Charlotte Bobcats 31.57
LosAngeles Clippers 31.51
Detroit Pistons 30.7
Toronto Raptors 30.01
Washington Wizards 27.16
NewJersey Nets 26.38
Minnesota Timberwolves 26.05
Sacramento Kings 23.13
Cleveland Cavaliers 20.23
Code:
b2n 5.43
Vegas 4.65
JH 4.49
KP 6.37
KD 6
Dsmok1 4.76
Crow 4.94
schtevie 6.42
WoW 6.86
WS 6.44
SPM(bbr) 6.51
SRS 6.69
“41” 9.52
(lastyear+41)/2 6.54
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 246
PostPosted: Sun Jan 16, 2011 9:25 am Post subject: Reply with quote
Are those numbers the sqrt of the residuals?
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Sun Jan 16, 2011 1:00 pm Post subject: Reply with quote
It's just average of absolute difference between expected wins(by in-season SRS) and the forecast
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 408
PostPosted: Sun Jan 16, 2011 1:08 pm Post subject: Reply with quote
Please reformulate so that I may be in the lead.
Back to top
View user's profile Send private message
Mogilny
Joined: 05 Aug 2010
Posts: 18
PostPosted: Sun Jan 23, 2011 6:55 am Post subject: Reply with quote
EvanZ wrote:
back2newbelf wrote:
pre-season over-under
That's impressive. Do they have their own stat guys calculate this? Or is it based on how people bet?
The accuracy of prediction markets is often amazing. Alot of interesting stuff has been written about it, Robin Hanson over att overcomingbias.com sometimes posts intriguing sectors of application on his blog if you are interested.
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Sun Jan 23, 2011 7:13 am Post subject: Reply with quote
Code:
SanAntonio Spurs 60.94
Boston Celtics 57.83
Miami Heat 56.73
LosAngeles Lakers 54.9
Orlando Magic 53.05
Chicago Bulls 52.99
NewOrleans Hornets 51.13
Dallas Mavericks 50.2
OklahomaCity Thunder 49.47
Atlanta Hawks 48.48
Denver Nuggets 46.39
Utah Jazz 46.19
PortlandTrail Blazers 44.31
NewYork Knickerbockers 42.15
Memphis Grizzlies 41.39
Houston Rockets 39.54
Phoenix Suns 38.49
Philadelphia 76ers 37.98
Milwaukee Bucks 37.4
Indiana Pacers 36.53
GoldenState Warriors 35.82
LosAngeles Clippers 33.37
Charlotte Bobcats 32.41
Detroit Pistons 31.62
Toronto Raptors 28.73
Washington Wizards 27.73
NewJersey Nets 27.27
Minnesota Timberwolves 25.18
Sacramento Kings 22.6
Cleveland Cavaliers 19.16
Hornets and Clippers are hot, Cavaliers are ice cold
Code:
b2n 5.43
Vegas 4.48
JH 4.44
KP 6.55
KD 5.93
Dsmok1 4.7
Crow 4.96
schtevie 6.45
WoW 6.61
WS 6.37
SPM(bbr) 6.39
SRS 6.7
“41” 9.41
(lastyear+41)/2 6.42
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 795
PostPosted: Sun Jan 23, 2011 1:12 pm Post subject: Reply with quote
Vegas and Hollinger vary in team win projections by an average (absolute value) of 2.6. They were within 2 of each other in 16 cases. Within 4 in 26 cases. The biggest differences are Charlotte, the Clippers and Philly, all different by 8 or 9. Vegas is currently projected by SRS to end up slightly closer on these 3.
I weighted Vegas into my preliminary first cut mix but probably not heavily enough. I was between these two projections on Charlotte and the Clippers and look close with that but went too independent, way lower than these 2 and probably way wrong or way way wrong on Philly.
I was different than Vegas by an average of 3.2. Different by 5 or more for Cleveland, Detroit, Golden State, Clippers, Orlando, Philly, Utah and Washington. I might end up closer or equidistant on a couple. I overplayed my subjective view on a few. Maybe I should have weighted Vegas even higher than usual especially in the east where I know less. I doubt I could have talked myself out of varying much on some of these projections though.
DSMok1's projection varied from Vegas by an average of 3.9. Different by 5 or more for 10 teams.
back2newbelf's varied from Vegas by an average of 3.5. Different by 5 or more for 9 teams.
A few other close projections:
Vegas liked the Clippers a lot more than most. Schtevie also had the same 36 win projection and Kelly (KD) had 38. (The ESPN panel had 35.) Schtevie may be closest on Boston. KD probably closest on Cleveland. Kevin (KP) may end up closest on Indiana and Chicago.
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Sun Jan 23, 2011 6:15 pm Post subject: Reply with quote
If you wanted to make good predictions every year, you could just use the vegas lines and tweak them a *tiny* bit and be better or very close to what will probably be the best.
But I don't think that is the point here really. I think we should be out to find the (combination of) metric(s) that will ultimately be best in predicting team wins. Thinking we could consistently beat Vegas with that methodology is probably naive, there are literally billions of $'s at stake for them, none for us. So the quest should probably be to come as close as possible.
That's also why I'm not too hot on predictions where it wasn't clearly explained what produced the predictions. We don't gain any sort of wisdom for the future and the whole thing just turns into some kind of internet-dick-waving
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 246
PostPosted: Sun Jan 23, 2011 7:00 pm Post subject: Reply with quote
I'll definitely take part in this next year.
(Or maybe in the playoffs, if you guys do that?)
Allocation of minutes probably plays as big a role in predictions as the metrics themselves. I think a more objective measure of each metric is to retrodict the previous season using the minutes that were actually played. Sure, that doesn't help prediction, because you still have to actually get the minutes right. But it can help us compare the relative merits of the various metrics. Of course, it won't help us compare with Vegas or Hollinger, which are black boxes, as far as we're concerned.
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
EvanZ
Joined: 22 Nov 2010
Posts: 217
PostPosted: Sun Jan 23, 2011 7:02 pm Post subject: Reply with quote
Mogilny wrote:
EvanZ wrote:
back2newbelf wrote:
pre-season over-under
That's impressive. Do they have their own stat guys calculate this? Or is it based on how people bet?
The accuracy of prediction markets is often amazing. Alot of interesting stuff has been written about it, Robin Hanson over att overcomingbias.com sometimes posts intriguing sectors of application on his blog if you are interested.
Thanks, I'll check it out.
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 776
PostPosted: Sun Jan 23, 2011 7:22 pm Post subject: Reply with quote
Not sure b2nb if you are referring to my predictions mainly, partially or not... but I'll just note that I explained my method in response to your question on page 5 of this thread. 11 of the 12 sources used in the first cut were metric-based, 9 known, 2 unknown and the ESPN panel. While the contest is a showcase for pure metrics against each other, mine is heavily a blended metric with a little extra. I think there ought to be blends under study too and one of your comments agrees with that.
I haven't backward mapped what the blended stats weights produced are for this grand mix as it would be quite tricky. If it did really well. maybe that would be worthwhile. I have the data on the blend before the adjustment too so I can see what the blended metric said equally weighted, with my weights and after what I added on top of it. So I am learning a few things and that really is my main focus.
Simpler mixes of fewer metrics would be easier to study. That can still be done and I might look at trying to optimize such a mix on this season and last season and then see what it can do in the future. I don't know if I'll join the contest in the future: but, if I do, I'll probably use EZPM100 in the mix of a blended metric and probably draw from a more limited set of other resources based on how they did this season. I think there is wisdom in the masses but more in some places than others.
Just as a mini-test I shifted to a 5 resource metric, quickly changed the weights and got results slightly better than Vegas. Whether it would last I don't know- yet; but it is suggestive of the possible promise of weighted metrics at team level performance projection.
Good team level performance projection can have value to teams with season ahead of them. An indication a bit ahead of game results to go get the piece you need to get higher, or try for the deal to get back in contention or be more willing to maybe scrape it and start over.
Comparing metrics against Vegas is a simple way to see differences in the metrics. It is a valuable comparison I think because in the last 4 years it has always been 1st or 2nd.
While they are helpful even without knowing exactly what they do, you don't have to stop and treat them as a brick wall. Trying to backward map exactly what the Hollinger and Vegas predictions thought of specific players and assumed on minutes and how they weighted stat contributions would be difficult but you might be able to do fairly well with enough technique and effort. It might be easier to simply optimize a metric of your own construction that tracked well with them over several years, if that is possible. I think you could get pretty useful insight out of it at the level of players and stat weights. I am sure there are a good number of people that try to do this for betting purposes but I think teams could probably benefit from doing this too as another source of analysis at season level (and probably at game level too).
If you weren't betting or just trying to beat Vegas in a little contest, you might think just focus on matching up a metric with the actual results. But there might be value in both matching up with the best projection and the actual results and the former might be more valuable in the long-run.
Last edited by Crow on Mon Jan 24, 2011 12:59 pm; edited 1 time in total
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 42
PostPosted: Mon Jan 24, 2011 12:12 am Post subject: Reply with quote
Wanted to pop in and mention that the "wisdom of the crowds" element of Vegas Regular Season Win Totals is particularly sharp because the general public typically doesn't bet these (NBA isn't on their yet radar in the heart of football, and the public isn't fond of bets where they have to let their money sit in sportsbook coffers for several months to see if they win).
So, the Vegas numbers in this instance represent what the oddsmakers think, shaped by what the most avid professional wagerers think...
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 776
PostPosted: Thu Jan 27, 2011 1:02 pm Post subject: Reply with quote
Was the Vegas line for Miami, one of which was at 63.5 at the start of the season, the true estimate or the true estimate adjusted by expected bettor response to it and set to make money for Vegas?
It looks like under 63.5 is going to win. Not sure which way the sharpies or self-considered sharpies broke more than the other.
The discussed ESPN panel average of 61 wins will probably be too high. They have to win 76% just to get to 60. Hollinger is currently projecting 57.
My 63 probably was too high, but it was under the Vegas line, as were 2 other projections here.
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 42
PostPosted: Thu Jan 27, 2011 2:50 pm Post subject: Reply with quote
Depends on if you're referencing an opening line crow, or where it closed after the "wisdom of the crowds" expressed itself with bets.
Generally, the opening line from oddsmakers would reflect anticipating "expected bettor response...set to make money for Vegas?" theme...as oddsmakers take what they think themselves about the proposition, how they expect bettors to react, and see if they can put their sportsbooks in position to win money by either splitting the action and collecting a vigorish...or by taking a position against public sentiment that would yield a free point or two in a game, or a free half-win or a win in this kind of prop, in addition to that 11/10 vigorish in their favor.
(I've freelanced as a ghostwriter in that world to some degree, and that's what oddsmakers say).
The closing line would better reflect a "true estimate" in a wisdom of the crowds sense, because it reflects input from the market. There are some who believe that closing lines represent a "perfected market." Professional wagerers aim to "beat the closer" figuring that this means they're on the right side...and that the right side will win enough over the long haul to grind out a profit.
I'm not sure what the opening and closing numbers were way back then on the Heat's win total. Did it open at 64 or 65 and drop down to 63.5?
I know that many sharps were looking to bet the Under before it came up because oddsmakers had been quoted in the media about how dominant they expected the Heat to be. But, then the very respected Haralabos Voulgaris had projected 64 wins for "the big 3 and nine scrubs' with his simulation system (fourth box down at)
http://aloneinthecorner.com/
I think more sharps were on Under than Over just figuring that the likelihood of injury(ies) would make it difficult to reach the number....
Back to top
View user's profile Send private message
BobboFitos
Joined: 21 Feb 2009
Posts: 184
Location: Cambridge, MA
PostPosted: Thu Jan 27, 2011 3:20 pm Post subject: Reply with quote
Jeff Fogle wrote:
The closing line would better reflect a "true estimate" in a wisdom of the crowds sense, because it reflects input from the market. There are some who believe that closing lines represent a "perfected market." Professional wagerers aim to "beat the closer" figuring that this means they're on the right side...and that the right side will win enough over the long haul to grind out a profit.
For whatever it's worth, the closing line is the more accurate line, of course - since it reflects current injury statuses that may have been up in the air when the lines come out / and other miscellaneous information etc. That said, beating the opening line is what's important to a sports bettor.
I just reread what you wrote, and I'm pretty exhausted so maybe this is actually what you're saying. If that's the case, my apologies. The opening line -> closing line is used as a gauge to see how accurate you model is. If your initial line is closer to the closing line then the actual opening line, boom, you have the more accurate predictor model. (And then you can start printing $)
_________________
-Rob
Back to top
View user's profile Send private message AIM Address
Jeff Fogle
Joined: 11 Jan 2011
Posts: 42
PostPosted: Thu Jan 27, 2011 4:21 pm Post subject: Reply with quote
Think we're saying the same thing BF, but "beat" can be used in different ways...and I should have been clearer in expressing that.
If an analyst's methodology is superior to the opening number, it's beating the opener.
If a bettor is trying to get in before the line has moved against him and taken away value, then winning that race is beating the closer. At least, that's the term professional wagerers use when talking amongst themselves.
Say a line opens at favorite -2...and pro bettors like the favorite...because they think the line should be -3.5. They'll bet at -2, -2.5, and -3, then stop at -3.5. Maybe the line even moves to -4 because a bandwagon effect creates stragglers.
The bets at -2.5 and -3 "beat the closer" in terms of the race to get a good number.
If the selection then went on to win, the methodologies used to come up with the pick "beat the opener."
So, you could say pro bettors want to beat both!
Let's change it a bit though, and move the line the other way. A "perfected market" guy bets -2 because he thought the line should be -3.5. If the line went down to -1 or pick-em, he'd be saying to himself "Crap, I've got the wrong side" rather than "hurray, more points for me." The perfected market guys have such respect for the market that they would largely trust line moves over their own methodology. Their goal is to be in synch with the movement, but get in before "perfection" is attained. If they do that (beat the closer), they'll win more than they lose.
I'm always surprised by how little this mindset cares about the actual final score. They figure their work is done when they beat the closer, and everything will come out fine in the wash over the long haul.
You want a predictive formula that beats the openers, and a betting strategy that beats the closers is a way to say it I guess. Though, if a formula exists that is so good against the number that it throws the perfected market theory out the window (and I've had it resting on a ledge myself in many conversations), then all that matters is matching reality better than the market does.
Back to top
View user's profile Send private message
BobboFitos
Joined: 21 Feb 2009
Posts: 184
Location: Cambridge, MA
PostPosted: Thu Jan 27, 2011 5:15 pm Post subject: Reply with quote
Jeff Fogle wrote:
Think we're saying the same thing BF, but "beat" can be used in different ways...and I should have been clearer in expressing that.
.
Yeah, upon re reading your post again, I think we are in fact saying the same thing, I was just tripped up by word usage. All good!
_________________
-Rob
Back to top
View user's profile Send private message AIM Address
Crow
Joined: 20 Jan 2009
Posts: 776
PostPosted: Thu Jan 27, 2011 5:42 pm Post subject: Reply with quote
The line of the site I checked in early summer was 64.5 wins but I am not sure if that was the opening line. It might have been higher. It was 63.5 in late September.
An expectation 64 in the early summer would have suggested a different action then than later. In July you would have, if you followed thru right then, gone under the line barely. By late September / October with the same expectation you'd have gone over barely.
Spurs at 20-1 for the title looked like good odds in the early summer. Assuming you were ok with probably an 80+% chance of losing.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 776
PostPosted: Sat Feb 12, 2011 6:19 pm Post subject: Reply with quote
Hollinger's playoff ratings give the Celtics, Heat, Spurs. Lakers and Mavs about a 68% chance of winning the title. Much of the remaining chance is given to Chicago and Orlando. I'd guess that these 5 have a somewhat higher total chance to win the title and, of the remaining chance, Chicago and Orlando may be getting a bit too much of it, because of average point differential and the chance to get to the title game.
Looking at regular season performance against east and west playoff level teams and not trying to weigh everything right now, I'd say the Celtics have a better chance to get to the finals than the Heat (Hollinger's numbers say the reverse) but then they are probably close to equal on chance to win that series once there.
The Lakers appear to be better than the Spurs and Mavs against the playoff west in general but weaker against the playoff east, though this is for all 8 and not just the top 2 or 4 and their performance against the east so far may be more about concentration than ability.
If the Spurs win the west I'd say they are favored to win the title. If the Mavs do, I'd say toss-up. If the Lakers get to the Finals, I am not sure if I'd say even or lean slightly toward Boston against them but I'd lean slightly towards the Lakers against the Heat.
Back to top
View user's profile Send private message
huevonkiller
Joined: 25 May 2010
Posts: 11
Location: Miami, Fl
PostPosted: Sat Feb 12, 2011 6:49 pm Post subject: Reply with quote
Crow wrote:
Hollinger's playoff ratings give the Celtics, Heat, Spurs. Lakers and Mavs about a 68% chance of winning the title. Much of the remaining chance is given to Chicago and Orlando. I'd guess that these 5 have a somewhat higher total chance to win the title and, of the remaining chance, Chicago and Orlando may be getting a bit too much of it, because of average point differential and the chance to get to the title game.
Looking at regular season performance against east and west playoff level teams and not trying to weigh everything right now, I'd say the Celtics have a better chance to get to the finals than the Heat (Hollinger's numbers say the reverse) but then they are probably close to equal on chance to win that series once there.
The Lakers appear to be better than the Spurs and Mavs against the playoff west in general but weaker against the playoff east, though this is for all 8 and not just the top 2 or 4 and their performance against the east so far may be more about concentration than ability.
If the Spurs win the west I'd say they are favored to win the title. If the Mavs do, I'd say toss-up. If the Lakers get to the Finals, I am not sure if I'd say even or lean slightly toward Boston against them but I'd lean slightly towards the Lakers against the Heat.
I wouldn't say so, any numbers to support those conclusions?
From what I've seen out of the Heat, they're 30-4 in James' last 34 games, and that's with injuries. I understand why Hollinger would want to weigh the latter parts of the season more. They were 9-8 at the start of the year, although basketball-reference still had them as a top tier team even then.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 776
PostPosted: Sat Feb 12, 2011 7:25 pm Post subject: Reply with quote
You are right to emphasize more recent for the Heat to allow adjustment into the new team and I was mainly focused on one dimension of the overall issue, but they are the clear weakest of the big 5 against the top 8 of their own conference on net efficiency for the season and only 8-7 against them. Meanwhile the Celtics are second only to the Lakers against the top 8 of their own conference on net efficiency and are 14-2 against that group.
The Heat might be or probably are better for the last 20-30 games, haven't checked the detail, but I don't think they've entirely closed the distance on these criteria in comparison with the others in the top 5 and especially Boston.
The Heat are the weakest on W-L against the league's best 10 but are probably better against the western teams in this group than the east so I rated the road to the finals as perhaps harder for them than the Finals itself. And I hoped others might share their perspectives on the end game prospects too.
After the '04 Pistons, the champ has usually been at or above 109 on offensive efficiency and always better than 105 on defensive efficiency. This is not an absolute but 3 of the named top 5 this season meet both criteria, Dallas misses by a little on defense, Boston by a bit more on offense.
Orlando misses mildly on offense, Chicago misses on offense by a pretty wide margin but their really strong defense could offset that. It did for 2 title teams since the '04 Pistons and it certainly did for those Pistons.
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 42
PostPosted: Sat Feb 12, 2011 8:22 pm Post subject: Reply with quote
Generally seeing:
Miami +200
Boston +300
Lakers +300
San Antonio +400 or +500
Orlando +1000 (lower in some places)
Dallas +1200
Chicago +1800
Those are moneylines. Risk $100 to win $200 with Miami, to win $300 with Boston, etc..
Miami opens at -1 or -1.5 on the road in Boston Sunday (just saw those). Boston's shorthanded though, so not a perfect proxy. Market saying Miami's 4 or 4.5 points better than Boston tomorrow on a neutral court.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 776
PostPosted: Sat Feb 12, 2011 10:03 pm Post subject: Reply with quote
I'll have to think more about those numbers.
But, the Vegas estimate just prior to the start of the season of the Spurs win total is looking off by probably the largest amount (way low). I guess they still aren't as high on them as their current record and SRS for some reason(s).
SRS has the Spurs ahead of the Heat and the Celtics only about 1 point behind the Heat. B-R's Adjusted Efficiencies has the Heat ahead of the Spurs by about a 1/3rd of a point and the Celtics behind by about 1 1/3rd. Hollinger's power rating is similar to SRS.
The Heat have had a very strong last 5 games and the Celtics have been just neutral but I don't know how much that figures in the title hopes bet. I wouldn't think much.
I can see the 3 teams ahead of them being the more popular bets and affecting the rest.
At first look, that Mavs price looks relatively attractive.
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 42
PostPosted: Sat Feb 12, 2011 11:51 pm Post subject: Reply with quote
Think you're right about popularity influencing those numbers crow. Spurs just don't resonate with bettors the way Kobe, LeBron, or the Celtics do.
Spurs beating the market about 63% of the time this year too...meaning prices continue to be behind the reality curve...
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Sun Feb 13, 2011 12:33 am Post subject: Reply with quote
Looks the Heat have an even split against the spread, Celtics -2 games, Lakers -4, Mavs +5 and Spurs +14. Spurs have the 2nd highest % of ATS covers, Mavs 7th best. Lakers 8th worst.
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 300
PostPosted: Sun Feb 13, 2011 8:04 am Post subject: Reply with quote
If SAS is good enough to win down the stretch with Duncan playing 12 minutes and Ginobili playing 8 minutes coming off the bench like last night, they should be well rested enough to get it done in the post-season. Laughing
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
huevonkiller
Joined: 25 May 2010
Posts: 15
Location: Miami, Fl
PostPosted: Sun Feb 13, 2011 4:04 pm Post subject: Reply with quote
Crow wrote:
You are right to emphasize more recent for the Heat to allow adjustment into the new team and I was mainly focused on one dimension of the overall issue, but they are the clear weakest of the big 5 against the top 8 of their own conference on net efficiency for the season and only 8-7 against them. Meanwhile the Celtics are second only to the Lakers against the top 8 of their own conference on net efficiency and are 14-2 against that group.
The Heat might be or probably are better for the last 20-30 games, haven't checked the detail, but I don't think they've entirely closed the distance on these criteria in comparison with the others in the top 5 and especially Boston.
The Heat are the weakest on W-L against the league's best 10 but are probably better against the western teams in this group than the east so I rated the road to the finals as perhaps harder for them than the Finals itself. And I hoped others might share their perspectives on the end game prospects too.
After the '04 Pistons, the champ has usually been at or above 109 on offensive efficiency and always better than 105 on defensive efficiency. This is not an absolute but 3 of the named top 5 this season meet both criteria, Dallas misses by a little on defense, Boston by a bit more on offense.
Orlando misses mildly on offense, Chicago misses on offense by a pretty wide margin but their really strong defense could offset that. It did for 2 title teams since the '04 Pistons and it certainly did for those Pistons.
Are these new Heat the weakest against the top level teams? That's all that matters.
If you're saying they're not a completely different team, I don't buy that. The Superstars look very comfortable now and the numbers support it. LeBron was at 23 PER, Wade at 21 PER when they were 9-8. So Crow you need to be more specifiic.
In a 7 game series, with lets say 4 games in Miami, would the Heat be favored to lose against Boston? Just today for example the homecourt got the Celtics through a win. Let's see how that plays out later.
If Zydrunas gets benched for Joel Anthony earlier in the third, they win, or if Mike Miller makes 1 out of his four missed three pointers? All that is correctable at home especially.
I've already seen BBR's rankings, seems pretty objective to me.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Sun Feb 13, 2011 8:29 pm Post subject: Reply with quote
You see more of the Heat and know them better than I do. I offered an opinion based on a few stats that I think are important. I don't necessarily expect to sway you.
But as for recent stats for the "new" adjusted Heat, it looks like the Heat are about 7-6 against playoff level teams since mid-December and not much different than their near .500 performance against that level of teams for the early part of the season. Better than the first 2 weeks sure, but not different against playoff teams in the last 2 months vs the first 7-8 weeks. Smashing the also-rans consistently doesn't mean a whole lot to me in projecting the playoffs. It means more to some others.
In a quick and rough check it looks like Boston, San Antonio, the Lakers and Dallas all have had better records against playoff teams in the last 2 months than Miami, maintaining this difference for the whole season. Against the top 10 has also been less favorable for the Heat for the whole season compared to these other 4 top teams.
For an early sketch of playoff prospects of the top 5 I went by the actual regular season record against east and west playoff teams, FWIW to any.
Back to top
View user's profile Send private message
huevonkiller
Joined: 25 May 2010
Posts: 15
Location: Miami, Fl
PostPosted: Mon Feb 14, 2011 1:46 pm Post subject: Reply with quote
Crow wrote:
You see more of the Heat and know them better than I do. I offered an opinion based on a few stats that I think are important. I don't necessarily expect to sway you.
But as for recent stats for the "new" adjusted Heat, it looks like the Heat are about 7-6 against playoff level teams since mid-December and not much different than their near .500 performance against that level of teams for the early part of the season. Better than the first 2 weeks sure, but not different against playoff teams in the last 2 months vs the first 7-8 weeks. Smashing the also-rans consistently doesn't mean a whole lot to me in projecting the playoffs. It means more to some others.
In a quick and rough check it looks like Boston, San Antonio, the Lakers and Dallas all have had better records against playoff teams in the last 2 months than Miami, maintaining this difference for the whole season. Against the top 10 has also been less favorable for the Heat for the whole season compared to these other 4 top teams.
For an early sketch of playoff prospects of the top 5 I went by the actual regular season record against east and west playoff teams, FWIW to any.
It is just coincidence that they lost those games when LeBron/Bosh got hurt? I mean we're talking about the playoffs, Shaq teams used to be like that in the regular season too. I'd actually favor the Heat more in the playoffs with those shorter rotations.
28 point loss first game without LeBron to Denver, Chicago without LeBron, New York , Atlanta OT? The Heat should be favored to win against all those teams in the playoffs.
They've lost two games, one to Dallas and one to Boston yesterday. For the purposes of the East let's discuss Boston. Will they get homecourt in the playoffs and does it even matter? We know they beat the Cavs last year (and Boston was the best team when healthy at the start of the year), but now we see what a terrible team they actually are especially defensively.
Have these New Heat lost at home against the Boston Celtics yet? That is probably the more important question, and Neil has similar sentiments.
http://www.basketball-reference.com/blog/?p=8811
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Mon Feb 14, 2011 3:18 pm Post subject: Reply with quote
Star injuries hurt the Heat a bit against top teams vs not being injured, but some of the other top teams have as much or more missed games by stars.
Without the first round how a team performed vs. elite teams vs. a team's record against all teams is real close to the same for playoffs at home. Overall record appears to be a bit better at determining road strength. But sample size is a difference and may be a factor. A comparison of top 10 to mid 10 to bottom 10 would be sample size neutral. That is what is still needed as Scott suggested at B-R.
It might not be wise to declare vs. elite or vs. overall the right or better choice if they are close on equal sized samples. It might be worthwhile to see how good each does in projecting the performance of particular teams over a 3 year stretch, especially the largely intact over that time teams.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Mon Feb 14, 2011 4:10 pm Post subject: Reply with quote
Averages are one thing, but what about just for the champ?
Rank for champ (last 6 seasons)
Regular season W-L 2.6
Regular season SRS 2.8
Regular season W-L against top16 2.0
Regular season W-L against top10 1.5
Value in predicting who becomes champ is what I based my preference for performance against top 10 over overall record on.
The '07-08 Celtics had the advantage on the Lakers on all these criteria. In '08-09 the Celtics still had the SRS advantage and a better W% against top16 and were only down modestly on overall W-L. Performance against the top 10 however swung dramatically and massively in the Lakers' favor and they went on to win the title. In '09-10 the Lakers had the advantage on all these criteria but especially on performance against the top 10. By performance against the top 10 no one was anywhere close to the Lakers. Same for the 2008 and 2009 champs.
So for the last 6 seasons there was one really good team against the top 10 3 times and all won the title. Once there was two good teams against the top 10 and the better team against the top 10 won. Twice there was three good teams against the top 10. Once the best won and once the 4th place team on this criteria pulled the upset (Miami with Shaq)
Best regular season W-L picked one winner in 6 seasons with one tie / non-decision. Regular season SRS called 3 winners. Regular season W-L against top16 called 4 winners. Regular season W-L against the top 10 called 5 winners. Maybe a longer study would change the results but regular season W-L against the top 10 looks stronger than overall record recently.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Mon Feb 14, 2011 9:57 pm Post subject: Reply with quote
Found Sagarin data for performance against their top 10 for 4 of the 5 years prior to the initial 6 year study. Of these 4 seasons the #1 team on W-L against the top 10 won the title 3 times. So for 10 total seasons it is 8 title winners were #1 on this and 2 were 4th. Looks like a pretty good measure to me.
Is anything better? It isn't overall record or overall SRS. With the added years, best regular season W-L picked 4 total winners in 11 seasons with one tie / non-decision (no missing data gives an extra year). Regular season SRS called 5 winners in 11 seasons.
80% winners vs. about 37-45%. You decide what is "better". If the focus is finding the title winner, it looks pretty strong toward best W-L against the top 10 (though definitions on the top 10 vary and may affect the results some).
(If you really care about separating 2nd and 3rd round winners and losers it is a closer race but one affected by sample size.)
Back to top
View user's profile Send private message
huevonkiller
Joined: 25 May 2010
Posts: 15
Location: Miami, Fl
PostPosted: Tue Feb 15, 2011 2:08 am Post subject: Reply with quote
Crow wrote:
Found Sagarin data for performance against their top 10 for 4 of the 5 years prior to the initial 6 year study. Of these 4 seasons the #1 team on W-L against the top 10 won the title 3 times. So for 10 total seasons it is 8 title winners were #1 on this and 2 were 4th. Looks like a pretty good measure to me.
Is anything better? It isn't overall record or overall SRS. With the added years, best regular season W-L picked 4 total winners in 11 seasons with one tie / non-decision (no missing data gives an extra year). Regular season SRS called 5 winners in 11 seasons.
80% winners vs. about 37-45%. You decide what is "better". If the focus is finding the title winner, it looks pretty strong toward best W-L against the top 10 (though definitions on the top 10 vary and may affect the results some).
(If you really care about separating 2nd and 3rd round winners and losers it is a closer race but one affected by sample size.)
If you really care about playoff basketball, you wouldn't be so nonchalant about LeBron James getting injured he's like his own 40 win team. Lol right Cavs? Dwyane Wade and Chris Bosh are pretty decent too. Wade is still the better than Kevin Durant and Chris Bosh has become a very good defender.
I'll accept that barometer though if you want. First because when the trio is together they've dominated every team, and lost two toss-up games. Those losses to Denver, Chicago, Atlanta, New York mean absolutely nothing.
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3616
Location: Hendersonville, NC
PostPosted: Tue Feb 15, 2011 3:37 am Post subject: Reply with quote
Crow wrote:
.. In '09-10 the Lakers had the advantage on all these criteria but especially on performance against the top 10. By performance against the top 10 no one was anywhere close to the Lakers. ..
Did you subtract out the Lakers' 16-7 postseason record?
Sagarin on USA Today ranks teams with RS and PO records combined.
LA played 105 games, going 32-21 vs (eventual) Top10 teams.
That puts them at 16-14 going into the playoffs.
This still doesn't tell us exactly who is going to end up as a top 10 team. That depends on the playoffs.
http://www.usatoday.com/sports/sagarin/nba0910.htm
Quote:
... record versus teams in these
rating's CURRENT top 10...
Ranked by regular season W% vs eventual Top 10 teams.
Code:
vs top 10 playoff total reg.sea.
tm W% W L W L W L
Cle .591 2 4 15 13 13 9
Den .586 2 4 19 16 17 12
Dal .556 2 4 17 16 15 12
Orl .545 2 4 14 14 12 10
LAL .533 16 7 32 21 16 14
Uta .483 4 6 18 21 14 15
Phx .481 6 4 19 18 13 14
SAS .414 4 6 16 23 12 17
Okl .414 2 4 14 21 12 17
Bos .409 11 8 20 21 9 13
.500 51 51 184 184 133 133
In this particular season, Atlanta may have dropped out of the top 10 by way of their postseason (4-3 vs Mil, 0-4 vs Orl).
If so, even a snapshot of the end-of-season records 'vs Top 10' would not be the same as the W% seen here.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
Mike G
Joined: 14 Jan 2005
Posts: 3616
Location: Hendersonville, NC
PostPosted: Tue Feb 15, 2011 8:04 am Post subject: Reply with quote
Nothing in last season's monthly progress suggests the Lakers were about to peak going into the playoffs.
Code:
LAL W L W% Pts Opp Diff
Nov 12 2 .86 105.9 96.4 9.5
Dec 12 3 .80 104.1 98.1 6.0
Jan 12 5 .71 102.9 95.9 7.0
Feb 8 4 .67 98.3 94.1 4.2
Mar 9 6 .60 101.5 100.7 .8
Apr 3 4 .43 95.0 96.6 -1.6
http://www.basketball-reference.com/tea ... 10/splits/
Kobe sat the last 2 games (1-1).
They wind up the season at SRS of 4.78 .
Then proceed to beat a bunch of teams that were supposedly about as good:
Code:
opp W L Pts Opp Diff OpSRS LAL
Okl 4 2 95.5 93.8 1.70 3.55 5.25
Uta 4 0 109.3 102.0 7.25 5.33 12.58
Phx 4 2 113.5 109.3 4.17 4.68 8.85
Bos 4 3 90.6 87.1 3.43 3.37 6.80
16 7 7.93
The LAL column is just Opponent SRS + PPG Diff.
(I won't call it their SRS in the series, but I don't know what else it would be.)
Per game, the Lakes were 7.93 points above average.
Per series, the average is +8.37
In playoffs, they were 3.15 PPG better than in the season.
To not punish them for sweeping Utah -- per series, their improvement was (8.37-4.78) 3.59 ppg.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Tue Feb 15, 2011 11:58 am Post subject: Reply with quote
Thanks Mike for the reminder that Sagarin on USA Today ranks teams with RS and PO records combined. I had noticed that before but didn't process and remember it while doing this analysis. It could change things considerably.
Whether best against top 10 regular season still does better than overall record or SRS on recent average in predicting the title winner or not is a question that would a revised study to answer.
The '09-10 Lakers were the next closest to the '05-06 Heat among the last 6 champs in terms of low-end regular season accomplishment so they may not not be a typical recent case for the relationship between regular season and playoffs.
Last edited by Crow on Tue Feb 15, 2011 12:49 pm; edited 1 time in total
Back to top
View user's profile Send private message
huevonkiller
Joined: 25 May 2010
Posts: 15
Location: Miami, Fl
PostPosted: Tue Feb 15, 2011 12:47 pm Post subject: Reply with quote
Crow wrote:
Thanks Mike for the reminder that Sagarin on USA Today ranks teams with RS and PO records combined. I had noticed that before but didn't recall it for this analysis. It could change things considerably, but I'll have to check if against top 10 regular season still does better than overall record or SRS on recent average or not.
The '09-10 Lakers were the next closest to the '05-06 Heat among the last 6 champs in terms of low-end regular season accomplishment so they may not not be a typical case for the relationship between regular season and playoffs.
Looks pretty typical to me actually. Two teams that got their superstars healthy made a run. 2006 Shaquille O'Neal played 46% of the minutes at his position in the regular season. Then 68% in the post-season, and the counterpart PER he allowed also improved (defense) .
Similarly the Lakers' core in 2010 was injured, then they righted themselves. Kobe got his knee drained, etc.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Tue Feb 15, 2011 12:51 pm Post subject: Reply with quote
The regular season performance against the top 10 by the typical recent title winner is my focus.
Re-checking whether the very best on top 10 for just the regular season predicts the title winner, the right calls drop from 8 of 10 to 4 of 10. But the title winner is still usually top 2-3 on this.
Given the data it might be better to say that performance against the top 10 regular season and overall are about the same as independent predictors of title winners, not either one clearly better than the other.
The combination may be more powerful. Title winners very rarely fall outside the top 3 on at least one- that is, only once in 10 seasons, the one where Shaq's light minutes skewed the predictive power of regular season data.
Looking at top 3s this season (W%, SRS and against the top 10), you end up with these top contenders: Celtics, Heat, Spurs... and the Bulls and Mavs both +2 against the top 10 and essentially tied for 3rd on that though the Bulls would have a slight lead by W%. The Lakers fall short of the top 3 cutoffs each time. But account for Bynum's injury and being a 2 time returning champ who is perhaps waiting to turn it on again and you get back to the original 5 top contenders and the Bulls, at least at this moment given schedules to date.
All opponents in a top 10 are not of equal strength of course. Maybe the historical survey could be SOS schedule weighted or limited to top 5-6 to see if that matters much. Conference strength affects the raw W%. Being #1 on one criteria but not the other could be compared for each combination vs being #1 on both (or more than 2 if SRS is included). The study could be refined with more time.
Last edited by Crow on Thu Mar 24, 2011 8:11 pm; edited 1 time in total
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 300
PostPosted: Tue Feb 15, 2011 2:24 pm Post subject: Reply with quote
Sorry...getting back on top of the thread...Laughing
I know this is about "win" prediction, but I was thinking it may be more useful, in terms of developing models, to do the comparison with point differential directly. Most of our models are working with that, right?
Sacramento has only won 12 games. They "should" have won 16 according to their current p.d. One of the WoW bloggers (Alex @ Sports Skeptic) has been crunching some numbers using WP and ezPM. He found that ezPM has SAC winning 21 at this point. It looks bad (+9) compared to actual wins, but by wins according to p.d. it's about +5.
I know Vegas won't care. They don't have a p.d. bet (I assume). But in terms of the models...
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Sun Jan 02, 2011 11:26 am Post subject: Reply with quote
Happy new year everyone
Code:
SanAntonio Spurs 60.97
Miami Heat 59.04
Boston Celtics 58.7
Dallas Mavericks 54.67
LosAngeles Lakers 53.01
Chicago Bulls 52.76
Orlando Magic 50.25
Utah Jazz 49.6
OklahomaCity Thunder 48.02
Denver Nuggets 47.39
NewOrleans Hornets 46.73
Atlanta Hawks 45.98
PortlandTrail Blazers 42.84
NewYork Knickerbockers 42.72
Houston Rockets 41.83
Phoenix Suns 39.48
Philadelphia 76ers 38.7
Memphis Grizzlies 38.68
Milwaukee Bucks 38.42
Indiana Pacers 38.28
GoldenState Warriors 34.45
Toronto Raptors 32.07
Detroit Pistons 30.07
LosAngeles Clippers 29.85
Charlotte Bobcats 28.85
NewJersey Nets 27.64
Minnesota Timberwolves 27.17
Washington Wizards 26.97
Cleveland Cavaliers 23.81
Sacramento Kings 21.04
The Celtics are having a hard time without Garnett. Same with Dallas and Nowitzki. Indiana has cooled off quite a bit. Warriors, Suns and Bucks had the hardest schedule, Kings, Spurs and Heat the easiest. Kings, Bobcats and Clippers with the hardest remaining schedule, 76ers, Bucks and Celtics with the easiest
Code:
b2n 5.43
Vegas 4.8
JH 4.41
KP 6.33
KD 6.18
Dsmok1 4.75
Crow 4.92
schtevie 6.47
WoW 6.86
WS 6.11
SPM(bbr) 6.5
SRS 6.6
“41” 9.3
(lastyear+41)/2 6.42
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 246
PostPosted: Sun Jan 02, 2011 11:42 am Post subject: Reply with quote
Vegas, huh? Are those based on pre-season odds?
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Sun Jan 02, 2011 11:45 am Post subject: Reply with quote
pre-season over-under
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 246
PostPosted: Sun Jan 02, 2011 12:14 pm Post subject: Reply with quote
back2newbelf wrote:
pre-season over-under
That's impressive. Do they have their own stat guys calculate this? Or is it based on how people bet?
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Mon Jan 03, 2011 5:51 am Post subject: Reply with quote
EvanZ wrote:
That's impressive. Do they have their own stat guys calculate this? Or is it based on how people bet?
If it was just the latter I would imagine more of us would be sitting at some Hawaian beach with a cocktail in their hand, instead of posting on internet message boards from somewhere cold.
Sacramentos' schedule from Jan 28th until Feb 23rd : @LAL, NOH, BOS, SA, UTA, DAL, OKC, @PHX, @OKC, @DAL, @MIA, @ORL
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 246
PostPosted: Mon Jan 03, 2011 7:49 am Post subject: Reply with quote
back2newbelf wrote:
If it was just the latter I would imagine more of us would be sitting at some Hawaian beach with a cocktail in their hand, instead of posting on internet message boards from somewhere cold.
If that were an option, I have to think I wouldn't be here either. Laughing
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Sun Jan 16, 2011 5:33 am Post subject: Reply with quote
Code:
SanAntonio Spurs 61.2
Boston Celtics 57.87
Miami Heat 57.35
LosAngeles Lakers 55.3
Chicago Bulls 52.85
Orlando Magic 51.81
Dallas Mavericks 50.65
OklahomaCity Thunder 49.9
Utah Jazz 49.05
Atlanta Hawks 48.36
NewOrleans Hornets 48.18
Denver Nuggets 47.06
NewYork Knickerbockers 44.3
PortlandTrail Blazers 42.77
Memphis Grizzlies 41.21
Houston Rockets 39.57
Indiana Pacers 38.01
Milwaukee Bucks 37.93
Philadelphia 76ers 37.85
Phoenix Suns 36.82
GoldenState Warriors 35.22
Charlotte Bobcats 31.57
LosAngeles Clippers 31.51
Detroit Pistons 30.7
Toronto Raptors 30.01
Washington Wizards 27.16
NewJersey Nets 26.38
Minnesota Timberwolves 26.05
Sacramento Kings 23.13
Cleveland Cavaliers 20.23
Code:
b2n 5.43
Vegas 4.65
JH 4.49
KP 6.37
KD 6
Dsmok1 4.76
Crow 4.94
schtevie 6.42
WoW 6.86
WS 6.44
SPM(bbr) 6.51
SRS 6.69
“41” 9.52
(lastyear+41)/2 6.54
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 246
PostPosted: Sun Jan 16, 2011 9:25 am Post subject: Reply with quote
Are those numbers the sqrt of the residuals?
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Sun Jan 16, 2011 1:00 pm Post subject: Reply with quote
It's just average of absolute difference between expected wins(by in-season SRS) and the forecast
Back to top
View user's profile Send private message
schtevie
Joined: 18 Apr 2005
Posts: 408
PostPosted: Sun Jan 16, 2011 1:08 pm Post subject: Reply with quote
Please reformulate so that I may be in the lead.
Back to top
View user's profile Send private message
Mogilny
Joined: 05 Aug 2010
Posts: 18
PostPosted: Sun Jan 23, 2011 6:55 am Post subject: Reply with quote
EvanZ wrote:
back2newbelf wrote:
pre-season over-under
That's impressive. Do they have their own stat guys calculate this? Or is it based on how people bet?
The accuracy of prediction markets is often amazing. Alot of interesting stuff has been written about it, Robin Hanson over att overcomingbias.com sometimes posts intriguing sectors of application on his blog if you are interested.
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Sun Jan 23, 2011 7:13 am Post subject: Reply with quote
Code:
SanAntonio Spurs 60.94
Boston Celtics 57.83
Miami Heat 56.73
LosAngeles Lakers 54.9
Orlando Magic 53.05
Chicago Bulls 52.99
NewOrleans Hornets 51.13
Dallas Mavericks 50.2
OklahomaCity Thunder 49.47
Atlanta Hawks 48.48
Denver Nuggets 46.39
Utah Jazz 46.19
PortlandTrail Blazers 44.31
NewYork Knickerbockers 42.15
Memphis Grizzlies 41.39
Houston Rockets 39.54
Phoenix Suns 38.49
Philadelphia 76ers 37.98
Milwaukee Bucks 37.4
Indiana Pacers 36.53
GoldenState Warriors 35.82
LosAngeles Clippers 33.37
Charlotte Bobcats 32.41
Detroit Pistons 31.62
Toronto Raptors 28.73
Washington Wizards 27.73
NewJersey Nets 27.27
Minnesota Timberwolves 25.18
Sacramento Kings 22.6
Cleveland Cavaliers 19.16
Hornets and Clippers are hot, Cavaliers are ice cold
Code:
b2n 5.43
Vegas 4.48
JH 4.44
KP 6.55
KD 5.93
Dsmok1 4.7
Crow 4.96
schtevie 6.45
WoW 6.61
WS 6.37
SPM(bbr) 6.39
SRS 6.7
“41” 9.41
(lastyear+41)/2 6.42
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 795
PostPosted: Sun Jan 23, 2011 1:12 pm Post subject: Reply with quote
Vegas and Hollinger vary in team win projections by an average (absolute value) of 2.6. They were within 2 of each other in 16 cases. Within 4 in 26 cases. The biggest differences are Charlotte, the Clippers and Philly, all different by 8 or 9. Vegas is currently projected by SRS to end up slightly closer on these 3.
I weighted Vegas into my preliminary first cut mix but probably not heavily enough. I was between these two projections on Charlotte and the Clippers and look close with that but went too independent, way lower than these 2 and probably way wrong or way way wrong on Philly.
I was different than Vegas by an average of 3.2. Different by 5 or more for Cleveland, Detroit, Golden State, Clippers, Orlando, Philly, Utah and Washington. I might end up closer or equidistant on a couple. I overplayed my subjective view on a few. Maybe I should have weighted Vegas even higher than usual especially in the east where I know less. I doubt I could have talked myself out of varying much on some of these projections though.
DSMok1's projection varied from Vegas by an average of 3.9. Different by 5 or more for 10 teams.
back2newbelf's varied from Vegas by an average of 3.5. Different by 5 or more for 9 teams.
A few other close projections:
Vegas liked the Clippers a lot more than most. Schtevie also had the same 36 win projection and Kelly (KD) had 38. (The ESPN panel had 35.) Schtevie may be closest on Boston. KD probably closest on Cleveland. Kevin (KP) may end up closest on Indiana and Chicago.
Back to top
View user's profile Send private message
back2newbelf
Joined: 21 Jun 2005
Posts: 250
PostPosted: Sun Jan 23, 2011 6:15 pm Post subject: Reply with quote
If you wanted to make good predictions every year, you could just use the vegas lines and tweak them a *tiny* bit and be better or very close to what will probably be the best.
But I don't think that is the point here really. I think we should be out to find the (combination of) metric(s) that will ultimately be best in predicting team wins. Thinking we could consistently beat Vegas with that methodology is probably naive, there are literally billions of $'s at stake for them, none for us. So the quest should probably be to come as close as possible.
That's also why I'm not too hot on predictions where it wasn't clearly explained what produced the predictions. We don't gain any sort of wisdom for the future and the whole thing just turns into some kind of internet-dick-waving
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 246
PostPosted: Sun Jan 23, 2011 7:00 pm Post subject: Reply with quote
I'll definitely take part in this next year.
(Or maybe in the playoffs, if you guys do that?)
Allocation of minutes probably plays as big a role in predictions as the metrics themselves. I think a more objective measure of each metric is to retrodict the previous season using the minutes that were actually played. Sure, that doesn't help prediction, because you still have to actually get the minutes right. But it can help us compare the relative merits of the various metrics. Of course, it won't help us compare with Vegas or Hollinger, which are black boxes, as far as we're concerned.
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
EvanZ
Joined: 22 Nov 2010
Posts: 217
PostPosted: Sun Jan 23, 2011 7:02 pm Post subject: Reply with quote
Mogilny wrote:
EvanZ wrote:
back2newbelf wrote:
pre-season over-under
That's impressive. Do they have their own stat guys calculate this? Or is it based on how people bet?
The accuracy of prediction markets is often amazing. Alot of interesting stuff has been written about it, Robin Hanson over att overcomingbias.com sometimes posts intriguing sectors of application on his blog if you are interested.
Thanks, I'll check it out.
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 776
PostPosted: Sun Jan 23, 2011 7:22 pm Post subject: Reply with quote
Not sure b2nb if you are referring to my predictions mainly, partially or not... but I'll just note that I explained my method in response to your question on page 5 of this thread. 11 of the 12 sources used in the first cut were metric-based, 9 known, 2 unknown and the ESPN panel. While the contest is a showcase for pure metrics against each other, mine is heavily a blended metric with a little extra. I think there ought to be blends under study too and one of your comments agrees with that.
I haven't backward mapped what the blended stats weights produced are for this grand mix as it would be quite tricky. If it did really well. maybe that would be worthwhile. I have the data on the blend before the adjustment too so I can see what the blended metric said equally weighted, with my weights and after what I added on top of it. So I am learning a few things and that really is my main focus.
Simpler mixes of fewer metrics would be easier to study. That can still be done and I might look at trying to optimize such a mix on this season and last season and then see what it can do in the future. I don't know if I'll join the contest in the future: but, if I do, I'll probably use EZPM100 in the mix of a blended metric and probably draw from a more limited set of other resources based on how they did this season. I think there is wisdom in the masses but more in some places than others.
Just as a mini-test I shifted to a 5 resource metric, quickly changed the weights and got results slightly better than Vegas. Whether it would last I don't know- yet; but it is suggestive of the possible promise of weighted metrics at team level performance projection.
Good team level performance projection can have value to teams with season ahead of them. An indication a bit ahead of game results to go get the piece you need to get higher, or try for the deal to get back in contention or be more willing to maybe scrape it and start over.
Comparing metrics against Vegas is a simple way to see differences in the metrics. It is a valuable comparison I think because in the last 4 years it has always been 1st or 2nd.
While they are helpful even without knowing exactly what they do, you don't have to stop and treat them as a brick wall. Trying to backward map exactly what the Hollinger and Vegas predictions thought of specific players and assumed on minutes and how they weighted stat contributions would be difficult but you might be able to do fairly well with enough technique and effort. It might be easier to simply optimize a metric of your own construction that tracked well with them over several years, if that is possible. I think you could get pretty useful insight out of it at the level of players and stat weights. I am sure there are a good number of people that try to do this for betting purposes but I think teams could probably benefit from doing this too as another source of analysis at season level (and probably at game level too).
If you weren't betting or just trying to beat Vegas in a little contest, you might think just focus on matching up a metric with the actual results. But there might be value in both matching up with the best projection and the actual results and the former might be more valuable in the long-run.
Last edited by Crow on Mon Jan 24, 2011 12:59 pm; edited 1 time in total
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 42
PostPosted: Mon Jan 24, 2011 12:12 am Post subject: Reply with quote
Wanted to pop in and mention that the "wisdom of the crowds" element of Vegas Regular Season Win Totals is particularly sharp because the general public typically doesn't bet these (NBA isn't on their yet radar in the heart of football, and the public isn't fond of bets where they have to let their money sit in sportsbook coffers for several months to see if they win).
So, the Vegas numbers in this instance represent what the oddsmakers think, shaped by what the most avid professional wagerers think...
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 776
PostPosted: Thu Jan 27, 2011 1:02 pm Post subject: Reply with quote
Was the Vegas line for Miami, one of which was at 63.5 at the start of the season, the true estimate or the true estimate adjusted by expected bettor response to it and set to make money for Vegas?
It looks like under 63.5 is going to win. Not sure which way the sharpies or self-considered sharpies broke more than the other.
The discussed ESPN panel average of 61 wins will probably be too high. They have to win 76% just to get to 60. Hollinger is currently projecting 57.
My 63 probably was too high, but it was under the Vegas line, as were 2 other projections here.
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 42
PostPosted: Thu Jan 27, 2011 2:50 pm Post subject: Reply with quote
Depends on if you're referencing an opening line crow, or where it closed after the "wisdom of the crowds" expressed itself with bets.
Generally, the opening line from oddsmakers would reflect anticipating "expected bettor response...set to make money for Vegas?" theme...as oddsmakers take what they think themselves about the proposition, how they expect bettors to react, and see if they can put their sportsbooks in position to win money by either splitting the action and collecting a vigorish...or by taking a position against public sentiment that would yield a free point or two in a game, or a free half-win or a win in this kind of prop, in addition to that 11/10 vigorish in their favor.
(I've freelanced as a ghostwriter in that world to some degree, and that's what oddsmakers say).
The closing line would better reflect a "true estimate" in a wisdom of the crowds sense, because it reflects input from the market. There are some who believe that closing lines represent a "perfected market." Professional wagerers aim to "beat the closer" figuring that this means they're on the right side...and that the right side will win enough over the long haul to grind out a profit.
I'm not sure what the opening and closing numbers were way back then on the Heat's win total. Did it open at 64 or 65 and drop down to 63.5?
I know that many sharps were looking to bet the Under before it came up because oddsmakers had been quoted in the media about how dominant they expected the Heat to be. But, then the very respected Haralabos Voulgaris had projected 64 wins for "the big 3 and nine scrubs' with his simulation system (fourth box down at)
http://aloneinthecorner.com/
I think more sharps were on Under than Over just figuring that the likelihood of injury(ies) would make it difficult to reach the number....
Back to top
View user's profile Send private message
BobboFitos
Joined: 21 Feb 2009
Posts: 184
Location: Cambridge, MA
PostPosted: Thu Jan 27, 2011 3:20 pm Post subject: Reply with quote
Jeff Fogle wrote:
The closing line would better reflect a "true estimate" in a wisdom of the crowds sense, because it reflects input from the market. There are some who believe that closing lines represent a "perfected market." Professional wagerers aim to "beat the closer" figuring that this means they're on the right side...and that the right side will win enough over the long haul to grind out a profit.
For whatever it's worth, the closing line is the more accurate line, of course - since it reflects current injury statuses that may have been up in the air when the lines come out / and other miscellaneous information etc. That said, beating the opening line is what's important to a sports bettor.
I just reread what you wrote, and I'm pretty exhausted so maybe this is actually what you're saying. If that's the case, my apologies. The opening line -> closing line is used as a gauge to see how accurate you model is. If your initial line is closer to the closing line then the actual opening line, boom, you have the more accurate predictor model. (And then you can start printing $)
_________________
-Rob
Back to top
View user's profile Send private message AIM Address
Jeff Fogle
Joined: 11 Jan 2011
Posts: 42
PostPosted: Thu Jan 27, 2011 4:21 pm Post subject: Reply with quote
Think we're saying the same thing BF, but "beat" can be used in different ways...and I should have been clearer in expressing that.
If an analyst's methodology is superior to the opening number, it's beating the opener.
If a bettor is trying to get in before the line has moved against him and taken away value, then winning that race is beating the closer. At least, that's the term professional wagerers use when talking amongst themselves.
Say a line opens at favorite -2...and pro bettors like the favorite...because they think the line should be -3.5. They'll bet at -2, -2.5, and -3, then stop at -3.5. Maybe the line even moves to -4 because a bandwagon effect creates stragglers.
The bets at -2.5 and -3 "beat the closer" in terms of the race to get a good number.
If the selection then went on to win, the methodologies used to come up with the pick "beat the opener."
So, you could say pro bettors want to beat both!
Let's change it a bit though, and move the line the other way. A "perfected market" guy bets -2 because he thought the line should be -3.5. If the line went down to -1 or pick-em, he'd be saying to himself "Crap, I've got the wrong side" rather than "hurray, more points for me." The perfected market guys have such respect for the market that they would largely trust line moves over their own methodology. Their goal is to be in synch with the movement, but get in before "perfection" is attained. If they do that (beat the closer), they'll win more than they lose.
I'm always surprised by how little this mindset cares about the actual final score. They figure their work is done when they beat the closer, and everything will come out fine in the wash over the long haul.
You want a predictive formula that beats the openers, and a betting strategy that beats the closers is a way to say it I guess. Though, if a formula exists that is so good against the number that it throws the perfected market theory out the window (and I've had it resting on a ledge myself in many conversations), then all that matters is matching reality better than the market does.
Back to top
View user's profile Send private message
BobboFitos
Joined: 21 Feb 2009
Posts: 184
Location: Cambridge, MA
PostPosted: Thu Jan 27, 2011 5:15 pm Post subject: Reply with quote
Jeff Fogle wrote:
Think we're saying the same thing BF, but "beat" can be used in different ways...and I should have been clearer in expressing that.
.
Yeah, upon re reading your post again, I think we are in fact saying the same thing, I was just tripped up by word usage. All good!
_________________
-Rob
Back to top
View user's profile Send private message AIM Address
Crow
Joined: 20 Jan 2009
Posts: 776
PostPosted: Thu Jan 27, 2011 5:42 pm Post subject: Reply with quote
The line of the site I checked in early summer was 64.5 wins but I am not sure if that was the opening line. It might have been higher. It was 63.5 in late September.
An expectation 64 in the early summer would have suggested a different action then than later. In July you would have, if you followed thru right then, gone under the line barely. By late September / October with the same expectation you'd have gone over barely.
Spurs at 20-1 for the title looked like good odds in the early summer. Assuming you were ok with probably an 80+% chance of losing.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 776
PostPosted: Sat Feb 12, 2011 6:19 pm Post subject: Reply with quote
Hollinger's playoff ratings give the Celtics, Heat, Spurs. Lakers and Mavs about a 68% chance of winning the title. Much of the remaining chance is given to Chicago and Orlando. I'd guess that these 5 have a somewhat higher total chance to win the title and, of the remaining chance, Chicago and Orlando may be getting a bit too much of it, because of average point differential and the chance to get to the title game.
Looking at regular season performance against east and west playoff level teams and not trying to weigh everything right now, I'd say the Celtics have a better chance to get to the finals than the Heat (Hollinger's numbers say the reverse) but then they are probably close to equal on chance to win that series once there.
The Lakers appear to be better than the Spurs and Mavs against the playoff west in general but weaker against the playoff east, though this is for all 8 and not just the top 2 or 4 and their performance against the east so far may be more about concentration than ability.
If the Spurs win the west I'd say they are favored to win the title. If the Mavs do, I'd say toss-up. If the Lakers get to the Finals, I am not sure if I'd say even or lean slightly toward Boston against them but I'd lean slightly towards the Lakers against the Heat.
Back to top
View user's profile Send private message
huevonkiller
Joined: 25 May 2010
Posts: 11
Location: Miami, Fl
PostPosted: Sat Feb 12, 2011 6:49 pm Post subject: Reply with quote
Crow wrote:
Hollinger's playoff ratings give the Celtics, Heat, Spurs. Lakers and Mavs about a 68% chance of winning the title. Much of the remaining chance is given to Chicago and Orlando. I'd guess that these 5 have a somewhat higher total chance to win the title and, of the remaining chance, Chicago and Orlando may be getting a bit too much of it, because of average point differential and the chance to get to the title game.
Looking at regular season performance against east and west playoff level teams and not trying to weigh everything right now, I'd say the Celtics have a better chance to get to the finals than the Heat (Hollinger's numbers say the reverse) but then they are probably close to equal on chance to win that series once there.
The Lakers appear to be better than the Spurs and Mavs against the playoff west in general but weaker against the playoff east, though this is for all 8 and not just the top 2 or 4 and their performance against the east so far may be more about concentration than ability.
If the Spurs win the west I'd say they are favored to win the title. If the Mavs do, I'd say toss-up. If the Lakers get to the Finals, I am not sure if I'd say even or lean slightly toward Boston against them but I'd lean slightly towards the Lakers against the Heat.
I wouldn't say so, any numbers to support those conclusions?
From what I've seen out of the Heat, they're 30-4 in James' last 34 games, and that's with injuries. I understand why Hollinger would want to weigh the latter parts of the season more. They were 9-8 at the start of the year, although basketball-reference still had them as a top tier team even then.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 776
PostPosted: Sat Feb 12, 2011 7:25 pm Post subject: Reply with quote
You are right to emphasize more recent for the Heat to allow adjustment into the new team and I was mainly focused on one dimension of the overall issue, but they are the clear weakest of the big 5 against the top 8 of their own conference on net efficiency for the season and only 8-7 against them. Meanwhile the Celtics are second only to the Lakers against the top 8 of their own conference on net efficiency and are 14-2 against that group.
The Heat might be or probably are better for the last 20-30 games, haven't checked the detail, but I don't think they've entirely closed the distance on these criteria in comparison with the others in the top 5 and especially Boston.
The Heat are the weakest on W-L against the league's best 10 but are probably better against the western teams in this group than the east so I rated the road to the finals as perhaps harder for them than the Finals itself. And I hoped others might share their perspectives on the end game prospects too.
After the '04 Pistons, the champ has usually been at or above 109 on offensive efficiency and always better than 105 on defensive efficiency. This is not an absolute but 3 of the named top 5 this season meet both criteria, Dallas misses by a little on defense, Boston by a bit more on offense.
Orlando misses mildly on offense, Chicago misses on offense by a pretty wide margin but their really strong defense could offset that. It did for 2 title teams since the '04 Pistons and it certainly did for those Pistons.
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 42
PostPosted: Sat Feb 12, 2011 8:22 pm Post subject: Reply with quote
Generally seeing:
Miami +200
Boston +300
Lakers +300
San Antonio +400 or +500
Orlando +1000 (lower in some places)
Dallas +1200
Chicago +1800
Those are moneylines. Risk $100 to win $200 with Miami, to win $300 with Boston, etc..
Miami opens at -1 or -1.5 on the road in Boston Sunday (just saw those). Boston's shorthanded though, so not a perfect proxy. Market saying Miami's 4 or 4.5 points better than Boston tomorrow on a neutral court.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 776
PostPosted: Sat Feb 12, 2011 10:03 pm Post subject: Reply with quote
I'll have to think more about those numbers.
But, the Vegas estimate just prior to the start of the season of the Spurs win total is looking off by probably the largest amount (way low). I guess they still aren't as high on them as their current record and SRS for some reason(s).
SRS has the Spurs ahead of the Heat and the Celtics only about 1 point behind the Heat. B-R's Adjusted Efficiencies has the Heat ahead of the Spurs by about a 1/3rd of a point and the Celtics behind by about 1 1/3rd. Hollinger's power rating is similar to SRS.
The Heat have had a very strong last 5 games and the Celtics have been just neutral but I don't know how much that figures in the title hopes bet. I wouldn't think much.
I can see the 3 teams ahead of them being the more popular bets and affecting the rest.
At first look, that Mavs price looks relatively attractive.
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 42
PostPosted: Sat Feb 12, 2011 11:51 pm Post subject: Reply with quote
Think you're right about popularity influencing those numbers crow. Spurs just don't resonate with bettors the way Kobe, LeBron, or the Celtics do.
Spurs beating the market about 63% of the time this year too...meaning prices continue to be behind the reality curve...
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Sun Feb 13, 2011 12:33 am Post subject: Reply with quote
Looks the Heat have an even split against the spread, Celtics -2 games, Lakers -4, Mavs +5 and Spurs +14. Spurs have the 2nd highest % of ATS covers, Mavs 7th best. Lakers 8th worst.
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 300
PostPosted: Sun Feb 13, 2011 8:04 am Post subject: Reply with quote
If SAS is good enough to win down the stretch with Duncan playing 12 minutes and Ginobili playing 8 minutes coming off the bench like last night, they should be well rested enough to get it done in the post-season. Laughing
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
huevonkiller
Joined: 25 May 2010
Posts: 15
Location: Miami, Fl
PostPosted: Sun Feb 13, 2011 4:04 pm Post subject: Reply with quote
Crow wrote:
You are right to emphasize more recent for the Heat to allow adjustment into the new team and I was mainly focused on one dimension of the overall issue, but they are the clear weakest of the big 5 against the top 8 of their own conference on net efficiency for the season and only 8-7 against them. Meanwhile the Celtics are second only to the Lakers against the top 8 of their own conference on net efficiency and are 14-2 against that group.
The Heat might be or probably are better for the last 20-30 games, haven't checked the detail, but I don't think they've entirely closed the distance on these criteria in comparison with the others in the top 5 and especially Boston.
The Heat are the weakest on W-L against the league's best 10 but are probably better against the western teams in this group than the east so I rated the road to the finals as perhaps harder for them than the Finals itself. And I hoped others might share their perspectives on the end game prospects too.
After the '04 Pistons, the champ has usually been at or above 109 on offensive efficiency and always better than 105 on defensive efficiency. This is not an absolute but 3 of the named top 5 this season meet both criteria, Dallas misses by a little on defense, Boston by a bit more on offense.
Orlando misses mildly on offense, Chicago misses on offense by a pretty wide margin but their really strong defense could offset that. It did for 2 title teams since the '04 Pistons and it certainly did for those Pistons.
Are these new Heat the weakest against the top level teams? That's all that matters.
If you're saying they're not a completely different team, I don't buy that. The Superstars look very comfortable now and the numbers support it. LeBron was at 23 PER, Wade at 21 PER when they were 9-8. So Crow you need to be more specifiic.
In a 7 game series, with lets say 4 games in Miami, would the Heat be favored to lose against Boston? Just today for example the homecourt got the Celtics through a win. Let's see how that plays out later.
If Zydrunas gets benched for Joel Anthony earlier in the third, they win, or if Mike Miller makes 1 out of his four missed three pointers? All that is correctable at home especially.
I've already seen BBR's rankings, seems pretty objective to me.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Sun Feb 13, 2011 8:29 pm Post subject: Reply with quote
You see more of the Heat and know them better than I do. I offered an opinion based on a few stats that I think are important. I don't necessarily expect to sway you.
But as for recent stats for the "new" adjusted Heat, it looks like the Heat are about 7-6 against playoff level teams since mid-December and not much different than their near .500 performance against that level of teams for the early part of the season. Better than the first 2 weeks sure, but not different against playoff teams in the last 2 months vs the first 7-8 weeks. Smashing the also-rans consistently doesn't mean a whole lot to me in projecting the playoffs. It means more to some others.
In a quick and rough check it looks like Boston, San Antonio, the Lakers and Dallas all have had better records against playoff teams in the last 2 months than Miami, maintaining this difference for the whole season. Against the top 10 has also been less favorable for the Heat for the whole season compared to these other 4 top teams.
For an early sketch of playoff prospects of the top 5 I went by the actual regular season record against east and west playoff teams, FWIW to any.
Back to top
View user's profile Send private message
huevonkiller
Joined: 25 May 2010
Posts: 15
Location: Miami, Fl
PostPosted: Mon Feb 14, 2011 1:46 pm Post subject: Reply with quote
Crow wrote:
You see more of the Heat and know them better than I do. I offered an opinion based on a few stats that I think are important. I don't necessarily expect to sway you.
But as for recent stats for the "new" adjusted Heat, it looks like the Heat are about 7-6 against playoff level teams since mid-December and not much different than their near .500 performance against that level of teams for the early part of the season. Better than the first 2 weeks sure, but not different against playoff teams in the last 2 months vs the first 7-8 weeks. Smashing the also-rans consistently doesn't mean a whole lot to me in projecting the playoffs. It means more to some others.
In a quick and rough check it looks like Boston, San Antonio, the Lakers and Dallas all have had better records against playoff teams in the last 2 months than Miami, maintaining this difference for the whole season. Against the top 10 has also been less favorable for the Heat for the whole season compared to these other 4 top teams.
For an early sketch of playoff prospects of the top 5 I went by the actual regular season record against east and west playoff teams, FWIW to any.
It is just coincidence that they lost those games when LeBron/Bosh got hurt? I mean we're talking about the playoffs, Shaq teams used to be like that in the regular season too. I'd actually favor the Heat more in the playoffs with those shorter rotations.
28 point loss first game without LeBron to Denver, Chicago without LeBron, New York , Atlanta OT? The Heat should be favored to win against all those teams in the playoffs.
They've lost two games, one to Dallas and one to Boston yesterday. For the purposes of the East let's discuss Boston. Will they get homecourt in the playoffs and does it even matter? We know they beat the Cavs last year (and Boston was the best team when healthy at the start of the year), but now we see what a terrible team they actually are especially defensively.
Have these New Heat lost at home against the Boston Celtics yet? That is probably the more important question, and Neil has similar sentiments.
http://www.basketball-reference.com/blog/?p=8811
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Mon Feb 14, 2011 3:18 pm Post subject: Reply with quote
Star injuries hurt the Heat a bit against top teams vs not being injured, but some of the other top teams have as much or more missed games by stars.
Without the first round how a team performed vs. elite teams vs. a team's record against all teams is real close to the same for playoffs at home. Overall record appears to be a bit better at determining road strength. But sample size is a difference and may be a factor. A comparison of top 10 to mid 10 to bottom 10 would be sample size neutral. That is what is still needed as Scott suggested at B-R.
It might not be wise to declare vs. elite or vs. overall the right or better choice if they are close on equal sized samples. It might be worthwhile to see how good each does in projecting the performance of particular teams over a 3 year stretch, especially the largely intact over that time teams.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Mon Feb 14, 2011 4:10 pm Post subject: Reply with quote
Averages are one thing, but what about just for the champ?
Rank for champ (last 6 seasons)
Regular season W-L 2.6
Regular season SRS 2.8
Regular season W-L against top16 2.0
Regular season W-L against top10 1.5
Value in predicting who becomes champ is what I based my preference for performance against top 10 over overall record on.
The '07-08 Celtics had the advantage on the Lakers on all these criteria. In '08-09 the Celtics still had the SRS advantage and a better W% against top16 and were only down modestly on overall W-L. Performance against the top 10 however swung dramatically and massively in the Lakers' favor and they went on to win the title. In '09-10 the Lakers had the advantage on all these criteria but especially on performance against the top 10. By performance against the top 10 no one was anywhere close to the Lakers. Same for the 2008 and 2009 champs.
So for the last 6 seasons there was one really good team against the top 10 3 times and all won the title. Once there was two good teams against the top 10 and the better team against the top 10 won. Twice there was three good teams against the top 10. Once the best won and once the 4th place team on this criteria pulled the upset (Miami with Shaq)
Best regular season W-L picked one winner in 6 seasons with one tie / non-decision. Regular season SRS called 3 winners. Regular season W-L against top16 called 4 winners. Regular season W-L against the top 10 called 5 winners. Maybe a longer study would change the results but regular season W-L against the top 10 looks stronger than overall record recently.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Mon Feb 14, 2011 9:57 pm Post subject: Reply with quote
Found Sagarin data for performance against their top 10 for 4 of the 5 years prior to the initial 6 year study. Of these 4 seasons the #1 team on W-L against the top 10 won the title 3 times. So for 10 total seasons it is 8 title winners were #1 on this and 2 were 4th. Looks like a pretty good measure to me.
Is anything better? It isn't overall record or overall SRS. With the added years, best regular season W-L picked 4 total winners in 11 seasons with one tie / non-decision (no missing data gives an extra year). Regular season SRS called 5 winners in 11 seasons.
80% winners vs. about 37-45%. You decide what is "better". If the focus is finding the title winner, it looks pretty strong toward best W-L against the top 10 (though definitions on the top 10 vary and may affect the results some).
(If you really care about separating 2nd and 3rd round winners and losers it is a closer race but one affected by sample size.)
Back to top
View user's profile Send private message
huevonkiller
Joined: 25 May 2010
Posts: 15
Location: Miami, Fl
PostPosted: Tue Feb 15, 2011 2:08 am Post subject: Reply with quote
Crow wrote:
Found Sagarin data for performance against their top 10 for 4 of the 5 years prior to the initial 6 year study. Of these 4 seasons the #1 team on W-L against the top 10 won the title 3 times. So for 10 total seasons it is 8 title winners were #1 on this and 2 were 4th. Looks like a pretty good measure to me.
Is anything better? It isn't overall record or overall SRS. With the added years, best regular season W-L picked 4 total winners in 11 seasons with one tie / non-decision (no missing data gives an extra year). Regular season SRS called 5 winners in 11 seasons.
80% winners vs. about 37-45%. You decide what is "better". If the focus is finding the title winner, it looks pretty strong toward best W-L against the top 10 (though definitions on the top 10 vary and may affect the results some).
(If you really care about separating 2nd and 3rd round winners and losers it is a closer race but one affected by sample size.)
If you really care about playoff basketball, you wouldn't be so nonchalant about LeBron James getting injured he's like his own 40 win team. Lol right Cavs? Dwyane Wade and Chris Bosh are pretty decent too. Wade is still the better than Kevin Durant and Chris Bosh has become a very good defender.
I'll accept that barometer though if you want. First because when the trio is together they've dominated every team, and lost two toss-up games. Those losses to Denver, Chicago, Atlanta, New York mean absolutely nothing.
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3616
Location: Hendersonville, NC
PostPosted: Tue Feb 15, 2011 3:37 am Post subject: Reply with quote
Crow wrote:
.. In '09-10 the Lakers had the advantage on all these criteria but especially on performance against the top 10. By performance against the top 10 no one was anywhere close to the Lakers. ..
Did you subtract out the Lakers' 16-7 postseason record?
Sagarin on USA Today ranks teams with RS and PO records combined.
LA played 105 games, going 32-21 vs (eventual) Top10 teams.
That puts them at 16-14 going into the playoffs.
This still doesn't tell us exactly who is going to end up as a top 10 team. That depends on the playoffs.
http://www.usatoday.com/sports/sagarin/nba0910.htm
Quote:
... record versus teams in these
rating's CURRENT top 10...
Ranked by regular season W% vs eventual Top 10 teams.
Code:
vs top 10 playoff total reg.sea.
tm W% W L W L W L
Cle .591 2 4 15 13 13 9
Den .586 2 4 19 16 17 12
Dal .556 2 4 17 16 15 12
Orl .545 2 4 14 14 12 10
LAL .533 16 7 32 21 16 14
Uta .483 4 6 18 21 14 15
Phx .481 6 4 19 18 13 14
SAS .414 4 6 16 23 12 17
Okl .414 2 4 14 21 12 17
Bos .409 11 8 20 21 9 13
.500 51 51 184 184 133 133
In this particular season, Atlanta may have dropped out of the top 10 by way of their postseason (4-3 vs Mil, 0-4 vs Orl).
If so, even a snapshot of the end-of-season records 'vs Top 10' would not be the same as the W% seen here.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
Mike G
Joined: 14 Jan 2005
Posts: 3616
Location: Hendersonville, NC
PostPosted: Tue Feb 15, 2011 8:04 am Post subject: Reply with quote
Nothing in last season's monthly progress suggests the Lakers were about to peak going into the playoffs.
Code:
LAL W L W% Pts Opp Diff
Nov 12 2 .86 105.9 96.4 9.5
Dec 12 3 .80 104.1 98.1 6.0
Jan 12 5 .71 102.9 95.9 7.0
Feb 8 4 .67 98.3 94.1 4.2
Mar 9 6 .60 101.5 100.7 .8
Apr 3 4 .43 95.0 96.6 -1.6
http://www.basketball-reference.com/tea ... 10/splits/
Kobe sat the last 2 games (1-1).
They wind up the season at SRS of 4.78 .
Then proceed to beat a bunch of teams that were supposedly about as good:
Code:
opp W L Pts Opp Diff OpSRS LAL
Okl 4 2 95.5 93.8 1.70 3.55 5.25
Uta 4 0 109.3 102.0 7.25 5.33 12.58
Phx 4 2 113.5 109.3 4.17 4.68 8.85
Bos 4 3 90.6 87.1 3.43 3.37 6.80
16 7 7.93
The LAL column is just Opponent SRS + PPG Diff.
(I won't call it their SRS in the series, but I don't know what else it would be.)
Per game, the Lakes were 7.93 points above average.
Per series, the average is +8.37
In playoffs, they were 3.15 PPG better than in the season.
To not punish them for sweeping Utah -- per series, their improvement was (8.37-4.78) 3.59 ppg.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Tue Feb 15, 2011 11:58 am Post subject: Reply with quote
Thanks Mike for the reminder that Sagarin on USA Today ranks teams with RS and PO records combined. I had noticed that before but didn't process and remember it while doing this analysis. It could change things considerably.
Whether best against top 10 regular season still does better than overall record or SRS on recent average in predicting the title winner or not is a question that would a revised study to answer.
The '09-10 Lakers were the next closest to the '05-06 Heat among the last 6 champs in terms of low-end regular season accomplishment so they may not not be a typical recent case for the relationship between regular season and playoffs.
Last edited by Crow on Tue Feb 15, 2011 12:49 pm; edited 1 time in total
Back to top
View user's profile Send private message
huevonkiller
Joined: 25 May 2010
Posts: 15
Location: Miami, Fl
PostPosted: Tue Feb 15, 2011 12:47 pm Post subject: Reply with quote
Crow wrote:
Thanks Mike for the reminder that Sagarin on USA Today ranks teams with RS and PO records combined. I had noticed that before but didn't recall it for this analysis. It could change things considerably, but I'll have to check if against top 10 regular season still does better than overall record or SRS on recent average or not.
The '09-10 Lakers were the next closest to the '05-06 Heat among the last 6 champs in terms of low-end regular season accomplishment so they may not not be a typical case for the relationship between regular season and playoffs.
Looks pretty typical to me actually. Two teams that got their superstars healthy made a run. 2006 Shaquille O'Neal played 46% of the minutes at his position in the regular season. Then 68% in the post-season, and the counterpart PER he allowed also improved (defense) .
Similarly the Lakers' core in 2010 was injured, then they righted themselves. Kobe got his knee drained, etc.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 824
PostPosted: Tue Feb 15, 2011 12:51 pm Post subject: Reply with quote
The regular season performance against the top 10 by the typical recent title winner is my focus.
Re-checking whether the very best on top 10 for just the regular season predicts the title winner, the right calls drop from 8 of 10 to 4 of 10. But the title winner is still usually top 2-3 on this.
Given the data it might be better to say that performance against the top 10 regular season and overall are about the same as independent predictors of title winners, not either one clearly better than the other.
The combination may be more powerful. Title winners very rarely fall outside the top 3 on at least one- that is, only once in 10 seasons, the one where Shaq's light minutes skewed the predictive power of regular season data.
Looking at top 3s this season (W%, SRS and against the top 10), you end up with these top contenders: Celtics, Heat, Spurs... and the Bulls and Mavs both +2 against the top 10 and essentially tied for 3rd on that though the Bulls would have a slight lead by W%. The Lakers fall short of the top 3 cutoffs each time. But account for Bynum's injury and being a 2 time returning champ who is perhaps waiting to turn it on again and you get back to the original 5 top contenders and the Bulls, at least at this moment given schedules to date.
All opponents in a top 10 are not of equal strength of course. Maybe the historical survey could be SOS schedule weighted or limited to top 5-6 to see if that matters much. Conference strength affects the raw W%. Being #1 on one criteria but not the other could be compared for each combination vs being #1 on both (or more than 2 if SRS is included). The study could be refined with more time.
Last edited by Crow on Thu Mar 24, 2011 8:11 pm; edited 1 time in total
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 300
PostPosted: Tue Feb 15, 2011 2:24 pm Post subject: Reply with quote
Sorry...getting back on top of the thread...Laughing
I know this is about "win" prediction, but I was thinking it may be more useful, in terms of developing models, to do the comparison with point differential directly. Most of our models are working with that, right?
Sacramento has only won 12 games. They "should" have won 16 according to their current p.d. One of the WoW bloggers (Alex @ Sports Skeptic) has been crunching some numbers using WP and ezPM. He found that ezPM has SAC winning 21 at this point. It looks bad (+9) compared to actual wins, but by wins according to p.d. it's about +5.
I know Vegas won't care. They don't have a p.d. bet (I assume). But in terms of the models...
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Re: 2010-11 NBA Win Predictions
back2newbelf
Joined: 21 Jun 2005
Posts: 272
PostPosted: Thu Mar 24, 2011 5:24 pm Post subject: Reply with quote
Code:
SanAntonio Spurs 63.57
Chicago Bulls 59.25
LosAngeles Lakers 58.07
Boston Celtics 57.53
Miami Heat 56.78
Dallas Mavericks 55.89
OklahomaCity Thunder 52.9
Orlando Magic 52.49
Denver Nuggets 49.17
PortlandTrail Blazers 46.48
NewOrleans Hornets 45.81
Memphis Grizzlies 45.68
Atlanta Hawks 44.88
Houston Rockets 43.53
Philadelphia 76ers 42.94
Phoenix Suns 41.71
NewYork Knickerbockers 40.96
Utah Jazz 40.33
Indiana Pacers 37.21
GoldenState Warriors 34.33
Milwaukee Bucks 33.45
Charlotte Bobcats 33.34
LosAngeles Clippers 31.69
Detroit Pistons 30.01
NewJersey Nets 27.29
Toronto Raptors 24.13
Sacramento Kings 21.96
Washington Wizards 21.19
Minnesota Timberwolves 20.47
Cleveland Cavaliers 16.93
Code:
b2n 6.75
Vegas 5.52
JH 5.44
KP 7.79
KD 6.59
Dsmok1 6.3
Crow 6.37
schtevie 7.26
WoW 7.24
WS 6.58
SPM(bbr) 7.26
SRS 8.25
“41” 10.71
(lastyear+41)/2 8.16
_________________
http://stats-for-the-nba.appspot.com/
Back to top
View user's profile Send private message
greyberger
Joined: 27 Sep 2010
Posts: 50
PostPosted: Thu Mar 24, 2011 10:12 pm Post subject: Reply with quote
Looks like John Hollinger could quit his day job.
Back to top
View user's profile Send private message Send e-mail
Jeff Fogle
Joined: 11 Jan 2011
Posts: 68
PostPosted: Sun Mar 27, 2011 11:16 pm Post subject: Reply with quote
Thanks back2new...didn't get a chance until now to thank you for posting the update.
Do you have any thoughts, or do others, about what the thresholds might be for evaluating the predictions. Obviously John Hollinger deserves kudo's based on that chart. Is "pretty good" within a point of Vegas (say 6.52 or better)? Or two points (7.52 or better)?
Or, maybe, is 5.52 disappointing from Vegas, and the market and/or stat nation should be aiming for something closer to 4.0?
Not sure how to evaluate it beyond ranking from closest to furthest. Generally Vegas is the benchmark in the prediction field...so I'm confident JH deserves pats on the back...
Thanks again for posting that...looking forward to seeing the season-end report...
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 816
PostPosted: Mon Mar 28, 2011 12:01 am Post subject: Reply with quote
In the first contest here in 2007-8 my early meta-metric approach of using a loose blend of other metrics and then making some quick adjustments produced the lowest average error of 8.4 and the rest of the field of pure metric based predictions and not necessarily pure predictions trailed by about one point. The blended metric approach may have helped reduce the impact of systemic errors some. A form of regression to the mean of other predictions.
http://tinyurl.com/47gctbd
In 2008-9 Neil Paine won with an SPM based model that produced an average error of 7.7, narrowly besting John Hollinger. A wider comparison showed Bill Simmons with a slight win over Vegas.
http://sonicscentral.com/apbrmetrics/vi ... &start=150
In 2009-10 back2newbelf won here with an Adjusted +/- based model that produced an average error of 6.7 and beat Vegas by 2/3rds of a point and another poster (Cysco?) beat Vegas as well.
http://tinyurl.com/477pd8s
In the past several people said they regretted not regressing expectations back toward the mean some or some more. I felt that way. I think that generally would have helped with average error. I don't know how much the metric contestants have applied this change in recent runs.
It looks like there will be substantial improvement again this season on the order of possibly another full point.
The Vegas line referenced has been better than almost all listed predictions every year. I'd say it is at least "pretty good" to be within a point of Vegas. Beating Vegas once is notable but still probably not enough to "plan to do it for real" next time with an as presented metric because of the vigorish applied to bets and because no predictor in these listings has beaten Vegas twice in the timespan so far. Within 2 points of Vegas isn't bad for trying to predict all of them instead of trying to cherry-pick a few teams. I'd guess within 1.5 - 2 points of Vegas will probably beat a good number of other media and fan predictions but it would come down to compiling the errors of those others.
(For predictors with 2 or more entries including this season as of now, I think I have a slight lead on back2newbelf for best average rank. Neil Paine and John Hollinger have competed every year and both have 2 strong finishes and lesser ones that hurt their average rank.)
Injuries, trades and coaching moves will somewhat limit how much further the average error can be reduced but it probably can go down further.
(I tested a modification of my meta-metric weights about a month ago filtering the included metrics down to fewer of the best past performers and found it was possible to fit this season's data better with those weights than what I used before the season. Maybe that will help next season if I do it or maybe not.)
Maybe somebody gets under 5 next season.
An average error of 4 - 5 might be close to the limit to achieve once or on a regular basis, but hard to say for sure what the typical limit is at this point based on a few seasons. Long term retrodiction metric testing would probably be the way to go if one wanted to give it the time to improve precision of the predictor and then perhaps use that refined projection capability to either try to win "the contest" or improve pre-season stories or beat Vegas or help a team or just improve one's own perspective.
Last edited by Crow on Mon Mar 28, 2011 11:04 am; edited 1 time in total
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3597
Location: Hendersonville, NC
PostPosted: Mon Mar 28, 2011 9:13 am Post subject: Reply with quote
Crow, nice summary. You wrote:
Quote:
In the first contest here in 2007-8 my early meta-metric approach of using a loose blend of other metrics and then making some quick adjustments produced the lowest average error of 8.4
Reviewing that 2008 thread, I found this comment by Cherokee_ACB:
Quote:
To put it into perspective, Vegas (error=7.3) beats you all, but at least you all beat Bill Simmons (9.8). Last season, Vegas' predictions errored by just 5.1 on average, while bloggers did poorly.
It seems '08 was just a wild and crazy year for predictions. Everyone was 20+ wins too high for Chi and Mia. And 15+ too low with Por and LAL.
Our predictions probably haven't really gotten 'better'; the league's just been more predictable[/quote]
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
Crow
Joined: 20 Jan 2009
Posts: 816
PostPosted: Mon Mar 28, 2011 10:48 am Post subject: Reply with quote
Good added detail about 2007-8. I didn't re-read it all to find that.
It is "hard to say for sure what the typical limit is at this point based on a few seasons".
4 seasons is dangerously short to be sure that an apparent trend is real sustainable in the future progress or all real progress (for this or other things). I still think there has been some, but not sure how much. More time might tell more if the interest continues.
Growth in number of entries over time also helped the trend toward a lower winning average error thru more chances to be a strong outlier.
Another year of bigger unexpected stories could break the clean trend. Maybe the future lowest error will tend to generally fall between the recently experienced high and low.
I tried not to over-conclude from the results but there was a base case for progress to at least hint at in response to the questions asked and perhaps draw out the next round(s) of analysis.
Looks like there have been 6-12 teams per season where the average error was above 10 or at least near it with many individual predictions over 10. 2007-8 was at the high end of that range, this season appears to be failing at the low end, That accounts for a portion of the overall trend but, without doing the exact math, it appears to be only a minority part of the overall trend, probably less than 1 of the near 3 point average error improvement from 2007-8 to this season. The error this season will probably creep up some more from where it is right now based on where it was earlier but probably not a lot.
Last edited by Crow on Mon Mar 28, 2011 12:29 pm; edited 1 time in total
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 280
PostPosted: Mon Mar 28, 2011 11:29 am Post subject: Reply with quote
My main question is whether the differences in predictions are primarily due to the actual metrics being used or the predicted allocation of minutes. It would be nice after the season is finished to see retrodicted results for the various metrics. That way we could at least take the playing time predictions out of the mix and focus on the metrics themselves.
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
huevonkiller
Joined: 25 May 2010
Posts: 15
Location: Miami, Fl
PostPosted: Mon Mar 28, 2011 2:09 pm Post subject: Reply with quote
Jeff Fogle wrote:
back2newbelf, any updates on how the predictions are doing compared to actual performances now that we're in the final stages of the season?
Was fun to go back and read all the early-thread thoughts about Miami. Turned out they've mostly avoided serious injuries yet still have to win out to finish 60-22. Great team for working through the process of full season analysis and trying to figure out what all the indicators mean. Good to review what people were thinking before the season started too.
Page 4 of this thread shows the first listing of Vegas estimates, for anyone who just wants to see how the market compared to what's happened.
Since December Miami has played like a 63-66 win team (upper 60's if Big three are intact). Those estimations weren't that bad at all, I don't think anyone expected Miami's superstars to play so poorly in the first month of the season.
Further while Miami has struggled, I think people lose sight of the fact that a 1 point loss is not the same as the 2009 Cavs getting owned by elite teams. Certain tactics Miami has been using are flawed, but this team has developing chemistry (two ball-dominant perimeter superstars is very rare) and has a young coach.
With some ball movement down the stretch this team can probably win half of their close games. Those who guarantee otherwise make me question their basketball knowledge. Benching Mike Miller for James Jones would probably have added key victories too, just as an example.
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 68
PostPosted: Mon Mar 28, 2011 4:09 pm Post subject: Reply with quote
Thanks Crow, all, for the comprehensive responses. Vegas looks to be a solid benchmark year in and year out. Would be fun to crack the mystery of what's in those projections that is eluding some of the logical stat/human projections.
Back to top
View user's profile Send private message
BobboFitos
Joined: 21 Feb 2009
Posts: 199
Location: Cambridge, MA
PostPosted: Mon Mar 28, 2011 8:25 pm Post subject: Reply with quote
EvanZ wrote:
My main question is whether the differences in predictions are primarily due to the actual metrics being used or the predicted allocation of minutes. It would be nice after the season is finished to see retrodicted results for the various metrics. That way we could at least take the playing time predictions out of the mix and focus on the metrics themselves.
This this this
_________________
-Rob
Back to top
View user's profile Send private message AIM Address
Crow
Joined: 20 Jan 2009
Posts: 816
PostPosted: Mon Mar 28, 2011 10:52 pm Post subject: Reply with quote
Thanks Jeff.
What did the Vegas line foresee particularly better this year than most?
That the Cavs would fall farther than most thought, though Kelly Dwyer went far more extreme and will end up low by just a few.
That Detroit would be near .400. John Hollinger and the ESPN panel average called that closely too.
That Minnesota wouldn't gain much ground. The ESPN panel average called that probably a little closer.
I'd say Vegas shared in all of the widespread big misses. Its strength must be mostly coming from being a little closer here and there or maybe avoiding adding big misses that weren't widespread.
A simple and even blend of back2newbelf's RAPM and DSMok1's ASPM didn't improve the results beyond the best of the two this year.
Besides Vegas and Hollinger, the ESPN panel average did well. Blending many worked for pretty well for them in this lowest average error contest.
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 68
PostPosted: Mon Mar 28, 2011 11:52 pm Post subject: Reply with quote
When the year is in the books, crow, let's take a closer look at differences between Vegas and some of the projections that missed. I think you're right that it's a lot about coming closer here and there rather than correctly anticipating big surprises.
Let's start with last year's final team records...see where Vegas made adjustments from those...maybe it's a case of being more conservative in anticipating changes than others, and there's an inherent stability in the league giving the consistency of the have's and the consistency of the have not's. Just throwing that out off the top of my head. Want to wait until final records are in the books.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 816
PostPosted: Tue Mar 29, 2011 12:43 am Post subject: Reply with quote
Yes, last year to this year Vegas prediction change would be important to view. And maybe second half of last season to prediction or some such split. And maybe look at playoff success or failure too.
Somebody in an earlier year thread also raised the variable of team age. One could add team stability or team experience together.
East and West prediction - actuals and inter-conference records / expectations would be worthwhile too. Last season top 5-10 offenses and defenses. Previous home and road and low rest strength and whether to expect it again or expect regression to the mean on the tails.
Potentially one could also look at teams lead by a top 5-10 star and teams lacking a top 20 star. And the multiple top 20 star teams. Alpha dog contests and switches?
You could consider whether adding Coaching Adjusted +/- on top of the player data correlates well with predictions and / or actuals. Who has more of what they have done well or not well with? What strategies could the other coaches try to mess that up compared to neutral expectations and how likely is that? How is going to get fired / hired and when and with what impact?
Which teams by their language in the media or other signs sound and look above or below on unity, focus, confidence, etc?
Maybe look at pace or high pace - strong offense and low pace - strong defense subgroups.
If 3 point shooting is particularly volatile how dependent on that shot were teams last season compared to average? How reliant were they on 3 point shot defense and success there?
Did teams who rely on driving for shots and foulshots overachievers last season and this season and did it accelerate?
Any trend with high and low assist rate teams?
Bench strength and clarity.
If you looked at the game consistency and the clutch performance data, what can one make of it in general?
The strength of top 5-10 lineups and whether teams can field those from last season in this season or not. Or even player pair availability and expected usage.
Does shooting well and defending the shot well carry a greater in actual wins than it might appear from the season average data?
Does offensive possession usage on teams looking clearcut or it is a luxury of options or a potentially troubling competition / issue to sort out? Who handles that well and not- at player, coach and organization level?
What if you tried to predict injury frequency and injury impact? Would you gain edge on the competition or lose ground because it is too hard to do well? Is it too hard to do better than neutral?
What if you tried to predict trades to some degree? The lopsided salary dumps and the trade the future for now and now for the future ones? Could you at least add an estimated value impact * probability for the 4-10 trade situations with the most potential to occur? Is that to much or another example of what you might have to do to get edge?
What about big market / small market splits. It shouldn't matter in winning but does it affect predictions (Vegas and "the pack") and does it appear to affect winning at all?
Any new additions or subtractions to the "gets superstar calls" list?
I guess you could look at payroll too. If it is a proxy for GM valuations does it translates well or not? Model in the moral hazard dimension of player contract situation. And maybe anticipated team responses to signs of that or expectations of that. In-season team payroll and roster flexibility- cap space, exceptions, non-guaranteed contracts, open slots or slots easily opened if opportunity arises.
What about the benefits and moral hazards of GMs and owners?
If you put the team over and underachievers thru a complete wash, what stats are the subgroups of teams high or low on?
Any spike in win achievement among teams with 6+ rotation guys positive on Adjusted +/- or greater than expected loss from having a star with moderate to heavy negative Adjusted +/- or multiple such starters or rotation guys?
Defensive breakdown of teams with 3 or more bad rotation defenders or at least key lineup breakdowns with 2-3+ bad ones on the court?
If you spent a lot of time looking at and working with the data perhaps one could beat the average accuracy of typical minute projections too. Maybe especially for rookies and of course their predicted performance is tough but also an opportunity to exceed the norm.
How will player location changes affect their performance due to elevation change or degree or travel miles and rest levels or quality of medical staff or amount of weekend games or "team discipline" or city distractions or friendship reunions and breakups? Player familiarity with and success level against new conference and division and their familiarity with the player.
Who has good and well targeted player - mentor pairings and might expect to get more good things out of it especially from young players than without looking at that consideration?
There is probably more. Easier and quicker to list than to research and estimate impacts, but brainstorming is an early step in the process.
Last edited by Crow on Tue Mar 29, 2011 2:05 pm; edited 2 times in total
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 68
PostPosted: Tue Mar 29, 2011 11:13 am Post subject: Reply with quote
No, that would take way too long. Let's just look at last year's records (lol).
Agree that all of those and more could be influences. I think we'll find though that many of the positives will cluster (well-run teams make good trades, emphasize the best percentage strategies, pace their players well, hire coaches and mentors that mesh well with their talent, while poorly run teams don't). That may be at the heart of the consistency over the years with many of the same teams being in the same general spots over and over again.
Maybe looking only at the teams who had extreme changes from one year to the next will help isolate things. But, even then, you're dealing with small sample sizes in a very dynamic universe. Or, the answers may come from studying the quilts rather than focusing so closly on each thread of the quilt. Hard to say.
My take from the industry is that:
*Oddsmakers tend to look at the quilts with just a few little threads. What did this team do last year? What have they done the last few years? What's different coming into this season (new players or coaches)? They're not digging deep into stats/minutes/etc... If a team won 50 last year, and 48 the year before, and it's a veteran team with the same coach, the number will be in that range.
*Bettors (and it's almost exclusively the "sharps" who bet regular season win totals in the NBA because the public isn't into them) may focus more on the specifics you mention.
I know Haralabob pops in occasionally to this forum. Maybe he can shed some light on his thinking process about these propositions.
Definitely think the factors you've listed are very relevant. Maybe be tough to separate the things that go hand in hand...
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 816
PostPosted: Tue Mar 29, 2011 2:00 pm Post subject: Reply with quote
Thanks for the industry prespective.
Yes, a strong positive culture covers a lot of descriptive ground and it may not be necessary to deal with all the detail if you have a reasonable basis for some positive adjustment.
Though a strong positive culture manifests in specific ways and it might be worth it to a big bettor or a team to try to deal with all the complexity. Many positives (and even some neutrals) could add up to more than the sum of the parts so even a detailed but static review is probably not the ultimate review. The interactions are important. A massive dynamic model might be great if really well done. If time or expertise is insufficient to make it really great then using a simpler model would probably be safer.
Back to top
View user's profile Send private message
12 of 12
Crow
Joined: 20 Jan 2009
Posts: 817
PostPosted: Fri Apr 01, 2011 5:13 pm Post subject: Reply with quote
Very good question.
There appear to be several issues. One of them is using the vague term "Vegas line" in the discussion.
I first referenced http://www.sportsbook.com on Sept 22 in response to back2newbelf's question on page 3 of this thread. I believe it had an opening line but I can't see what that page said then, at least immediately.
Then on page 4 Ed Kupfer cited the Painted Area article on predictions http://thepaintedarea.blogspot.com/2010 ... tions.html
... and included Vegas values in his post. The Painted Area had a Vegas line of unspecified origin and date, I don't know if it was the opening line or the current line as of then. Given the Miami value of 64.5 I lean towards thinking it was the opening line of some book.
There are differences between what the PA article showed for Vegas and what Ed posted. It appears to me (now) that Ed rounded almost all of the PA stated Vegas line values down by .5 or in some cases 1 but not all for the list he posted in this thread.
I am not sure what back2newbelf used for Vegas in his tracking.
Last edited by Crow on Fri Apr 01, 2011 8:31 pm; edited 2 times in total
Back to top
View user's profile Send private message
Ed Küpfer
Joined: 30 Dec 2004
Posts: 786
Location: Toronto
PostPosted: Fri Apr 01, 2011 6:47 pm Post subject: Reply with quote
Crow wrote:
It appears to me (now) that Ed rounded almost all of the PA stated Vegas line values down by .5 or in some cases 1 but not all for the list he posted in this thread.
I can't remember exactly what I did there, but I usually adjust those numbers so the probabilities for all the teams sum to one.
_________________
ed
Back to top
View user's profile Send private message Send e-mail
Crow
Joined: 20 Jan 2009
Posts: 817
PostPosted: Fri Apr 01, 2011 8:25 pm Post subject: Reply with quote
I assumed that was the rationale and should have added that but I had something else come up to deal with.
If your listed values were used for contest error tracking, the adjustments represent a step away from that "Vegas line" but the Vegas values stated in the PA article summed to 18 wins more than available so something had to be done to address the issue by your approach or another.
If your listed values were not used for contest tracking, then your specific rounding approach is not an issue.
If the actual Vegas line with .5 win increments was used in the contest as is, then being able to pick between 2 win numbers was a possible advantage over most other predictions in the test. Most did not consider that an available or appropriate option. The difference in available choices for win predictions makes for a bit uneven average error "performance" comparison. If you want to sweat fractions or tiny fractions of a point of average error.
Last edited by Crow on Sat Apr 02, 2011 12:42 pm; edited 3 times in total
Back to top
View user's profile Send private message
haralabob
Joined: 11 Apr 2007
Posts: 27
PostPosted: Sat Apr 02, 2011 12:56 am Post subject: Reply with quote
The issue is these lines are rarely 50-50 propositions.
They'll have a line like;
Mem Over 41.5 -130
under +110
this probably equates to a line of around 42.5 and not 41.5 since you are penalized by betting on the over.
Either way I'd be fairly confident that the consensus line moves by the sharps who bet these season win totals (I don't) would probably beat most of the predictors out there.
Hollingers stuff looked pretty good tho.
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 69
PostPosted: Sat Apr 02, 2011 10:33 am Post subject: Reply with quote
H, what do you think lies at the heart of the advantage sharps have over the various stat model "predictors?" I'd understand if you don't want to discuss something that might give away an advantage you have over the market down the road. But, if you're willing to discuss it, what elements could the models add into their mix that would improve their performance?
I know we're talking about various models and methodologies. But, at least in this thread, there are some prominent ones where the parameters are fairly well known. Why do they fall short in your opinion?
Joined: 21 Jun 2005
Posts: 272
PostPosted: Thu Mar 24, 2011 5:24 pm Post subject: Reply with quote
Code:
SanAntonio Spurs 63.57
Chicago Bulls 59.25
LosAngeles Lakers 58.07
Boston Celtics 57.53
Miami Heat 56.78
Dallas Mavericks 55.89
OklahomaCity Thunder 52.9
Orlando Magic 52.49
Denver Nuggets 49.17
PortlandTrail Blazers 46.48
NewOrleans Hornets 45.81
Memphis Grizzlies 45.68
Atlanta Hawks 44.88
Houston Rockets 43.53
Philadelphia 76ers 42.94
Phoenix Suns 41.71
NewYork Knickerbockers 40.96
Utah Jazz 40.33
Indiana Pacers 37.21
GoldenState Warriors 34.33
Milwaukee Bucks 33.45
Charlotte Bobcats 33.34
LosAngeles Clippers 31.69
Detroit Pistons 30.01
NewJersey Nets 27.29
Toronto Raptors 24.13
Sacramento Kings 21.96
Washington Wizards 21.19
Minnesota Timberwolves 20.47
Cleveland Cavaliers 16.93
Code:
b2n 6.75
Vegas 5.52
JH 5.44
KP 7.79
KD 6.59
Dsmok1 6.3
Crow 6.37
schtevie 7.26
WoW 7.24
WS 6.58
SPM(bbr) 7.26
SRS 8.25
“41” 10.71
(lastyear+41)/2 8.16
_________________
http://stats-for-the-nba.appspot.com/
Back to top
View user's profile Send private message
greyberger
Joined: 27 Sep 2010
Posts: 50
PostPosted: Thu Mar 24, 2011 10:12 pm Post subject: Reply with quote
Looks like John Hollinger could quit his day job.
Back to top
View user's profile Send private message Send e-mail
Jeff Fogle
Joined: 11 Jan 2011
Posts: 68
PostPosted: Sun Mar 27, 2011 11:16 pm Post subject: Reply with quote
Thanks back2new...didn't get a chance until now to thank you for posting the update.
Do you have any thoughts, or do others, about what the thresholds might be for evaluating the predictions. Obviously John Hollinger deserves kudo's based on that chart. Is "pretty good" within a point of Vegas (say 6.52 or better)? Or two points (7.52 or better)?
Or, maybe, is 5.52 disappointing from Vegas, and the market and/or stat nation should be aiming for something closer to 4.0?
Not sure how to evaluate it beyond ranking from closest to furthest. Generally Vegas is the benchmark in the prediction field...so I'm confident JH deserves pats on the back...
Thanks again for posting that...looking forward to seeing the season-end report...
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 816
PostPosted: Mon Mar 28, 2011 12:01 am Post subject: Reply with quote
In the first contest here in 2007-8 my early meta-metric approach of using a loose blend of other metrics and then making some quick adjustments produced the lowest average error of 8.4 and the rest of the field of pure metric based predictions and not necessarily pure predictions trailed by about one point. The blended metric approach may have helped reduce the impact of systemic errors some. A form of regression to the mean of other predictions.
http://tinyurl.com/47gctbd
In 2008-9 Neil Paine won with an SPM based model that produced an average error of 7.7, narrowly besting John Hollinger. A wider comparison showed Bill Simmons with a slight win over Vegas.
http://sonicscentral.com/apbrmetrics/vi ... &start=150
In 2009-10 back2newbelf won here with an Adjusted +/- based model that produced an average error of 6.7 and beat Vegas by 2/3rds of a point and another poster (Cysco?) beat Vegas as well.
http://tinyurl.com/477pd8s
In the past several people said they regretted not regressing expectations back toward the mean some or some more. I felt that way. I think that generally would have helped with average error. I don't know how much the metric contestants have applied this change in recent runs.
It looks like there will be substantial improvement again this season on the order of possibly another full point.
The Vegas line referenced has been better than almost all listed predictions every year. I'd say it is at least "pretty good" to be within a point of Vegas. Beating Vegas once is notable but still probably not enough to "plan to do it for real" next time with an as presented metric because of the vigorish applied to bets and because no predictor in these listings has beaten Vegas twice in the timespan so far. Within 2 points of Vegas isn't bad for trying to predict all of them instead of trying to cherry-pick a few teams. I'd guess within 1.5 - 2 points of Vegas will probably beat a good number of other media and fan predictions but it would come down to compiling the errors of those others.
(For predictors with 2 or more entries including this season as of now, I think I have a slight lead on back2newbelf for best average rank. Neil Paine and John Hollinger have competed every year and both have 2 strong finishes and lesser ones that hurt their average rank.)
Injuries, trades and coaching moves will somewhat limit how much further the average error can be reduced but it probably can go down further.
(I tested a modification of my meta-metric weights about a month ago filtering the included metrics down to fewer of the best past performers and found it was possible to fit this season's data better with those weights than what I used before the season. Maybe that will help next season if I do it or maybe not.)
Maybe somebody gets under 5 next season.
An average error of 4 - 5 might be close to the limit to achieve once or on a regular basis, but hard to say for sure what the typical limit is at this point based on a few seasons. Long term retrodiction metric testing would probably be the way to go if one wanted to give it the time to improve precision of the predictor and then perhaps use that refined projection capability to either try to win "the contest" or improve pre-season stories or beat Vegas or help a team or just improve one's own perspective.
Last edited by Crow on Mon Mar 28, 2011 11:04 am; edited 1 time in total
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3597
Location: Hendersonville, NC
PostPosted: Mon Mar 28, 2011 9:13 am Post subject: Reply with quote
Crow, nice summary. You wrote:
Quote:
In the first contest here in 2007-8 my early meta-metric approach of using a loose blend of other metrics and then making some quick adjustments produced the lowest average error of 8.4
Reviewing that 2008 thread, I found this comment by Cherokee_ACB:
Quote:
To put it into perspective, Vegas (error=7.3) beats you all, but at least you all beat Bill Simmons (9.8). Last season, Vegas' predictions errored by just 5.1 on average, while bloggers did poorly.
It seems '08 was just a wild and crazy year for predictions. Everyone was 20+ wins too high for Chi and Mia. And 15+ too low with Por and LAL.
Our predictions probably haven't really gotten 'better'; the league's just been more predictable[/quote]
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
Crow
Joined: 20 Jan 2009
Posts: 816
PostPosted: Mon Mar 28, 2011 10:48 am Post subject: Reply with quote
Good added detail about 2007-8. I didn't re-read it all to find that.
It is "hard to say for sure what the typical limit is at this point based on a few seasons".
4 seasons is dangerously short to be sure that an apparent trend is real sustainable in the future progress or all real progress (for this or other things). I still think there has been some, but not sure how much. More time might tell more if the interest continues.
Growth in number of entries over time also helped the trend toward a lower winning average error thru more chances to be a strong outlier.
Another year of bigger unexpected stories could break the clean trend. Maybe the future lowest error will tend to generally fall between the recently experienced high and low.
I tried not to over-conclude from the results but there was a base case for progress to at least hint at in response to the questions asked and perhaps draw out the next round(s) of analysis.
Looks like there have been 6-12 teams per season where the average error was above 10 or at least near it with many individual predictions over 10. 2007-8 was at the high end of that range, this season appears to be failing at the low end, That accounts for a portion of the overall trend but, without doing the exact math, it appears to be only a minority part of the overall trend, probably less than 1 of the near 3 point average error improvement from 2007-8 to this season. The error this season will probably creep up some more from where it is right now based on where it was earlier but probably not a lot.
Last edited by Crow on Mon Mar 28, 2011 12:29 pm; edited 1 time in total
Back to top
View user's profile Send private message
EvanZ
Joined: 22 Nov 2010
Posts: 280
PostPosted: Mon Mar 28, 2011 11:29 am Post subject: Reply with quote
My main question is whether the differences in predictions are primarily due to the actual metrics being used or the predicted allocation of minutes. It would be nice after the season is finished to see retrodicted results for the various metrics. That way we could at least take the playing time predictions out of the mix and focus on the metrics themselves.
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir
Back to top
View user's profile Send private message
huevonkiller
Joined: 25 May 2010
Posts: 15
Location: Miami, Fl
PostPosted: Mon Mar 28, 2011 2:09 pm Post subject: Reply with quote
Jeff Fogle wrote:
back2newbelf, any updates on how the predictions are doing compared to actual performances now that we're in the final stages of the season?
Was fun to go back and read all the early-thread thoughts about Miami. Turned out they've mostly avoided serious injuries yet still have to win out to finish 60-22. Great team for working through the process of full season analysis and trying to figure out what all the indicators mean. Good to review what people were thinking before the season started too.
Page 4 of this thread shows the first listing of Vegas estimates, for anyone who just wants to see how the market compared to what's happened.
Since December Miami has played like a 63-66 win team (upper 60's if Big three are intact). Those estimations weren't that bad at all, I don't think anyone expected Miami's superstars to play so poorly in the first month of the season.
Further while Miami has struggled, I think people lose sight of the fact that a 1 point loss is not the same as the 2009 Cavs getting owned by elite teams. Certain tactics Miami has been using are flawed, but this team has developing chemistry (two ball-dominant perimeter superstars is very rare) and has a young coach.
With some ball movement down the stretch this team can probably win half of their close games. Those who guarantee otherwise make me question their basketball knowledge. Benching Mike Miller for James Jones would probably have added key victories too, just as an example.
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 68
PostPosted: Mon Mar 28, 2011 4:09 pm Post subject: Reply with quote
Thanks Crow, all, for the comprehensive responses. Vegas looks to be a solid benchmark year in and year out. Would be fun to crack the mystery of what's in those projections that is eluding some of the logical stat/human projections.
Back to top
View user's profile Send private message
BobboFitos
Joined: 21 Feb 2009
Posts: 199
Location: Cambridge, MA
PostPosted: Mon Mar 28, 2011 8:25 pm Post subject: Reply with quote
EvanZ wrote:
My main question is whether the differences in predictions are primarily due to the actual metrics being used or the predicted allocation of minutes. It would be nice after the season is finished to see retrodicted results for the various metrics. That way we could at least take the playing time predictions out of the mix and focus on the metrics themselves.
This this this
_________________
-Rob
Back to top
View user's profile Send private message AIM Address
Crow
Joined: 20 Jan 2009
Posts: 816
PostPosted: Mon Mar 28, 2011 10:52 pm Post subject: Reply with quote
Thanks Jeff.
What did the Vegas line foresee particularly better this year than most?
That the Cavs would fall farther than most thought, though Kelly Dwyer went far more extreme and will end up low by just a few.
That Detroit would be near .400. John Hollinger and the ESPN panel average called that closely too.
That Minnesota wouldn't gain much ground. The ESPN panel average called that probably a little closer.
I'd say Vegas shared in all of the widespread big misses. Its strength must be mostly coming from being a little closer here and there or maybe avoiding adding big misses that weren't widespread.
A simple and even blend of back2newbelf's RAPM and DSMok1's ASPM didn't improve the results beyond the best of the two this year.
Besides Vegas and Hollinger, the ESPN panel average did well. Blending many worked for pretty well for them in this lowest average error contest.
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 68
PostPosted: Mon Mar 28, 2011 11:52 pm Post subject: Reply with quote
When the year is in the books, crow, let's take a closer look at differences between Vegas and some of the projections that missed. I think you're right that it's a lot about coming closer here and there rather than correctly anticipating big surprises.
Let's start with last year's final team records...see where Vegas made adjustments from those...maybe it's a case of being more conservative in anticipating changes than others, and there's an inherent stability in the league giving the consistency of the have's and the consistency of the have not's. Just throwing that out off the top of my head. Want to wait until final records are in the books.
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 816
PostPosted: Tue Mar 29, 2011 12:43 am Post subject: Reply with quote
Yes, last year to this year Vegas prediction change would be important to view. And maybe second half of last season to prediction or some such split. And maybe look at playoff success or failure too.
Somebody in an earlier year thread also raised the variable of team age. One could add team stability or team experience together.
East and West prediction - actuals and inter-conference records / expectations would be worthwhile too. Last season top 5-10 offenses and defenses. Previous home and road and low rest strength and whether to expect it again or expect regression to the mean on the tails.
Potentially one could also look at teams lead by a top 5-10 star and teams lacking a top 20 star. And the multiple top 20 star teams. Alpha dog contests and switches?
You could consider whether adding Coaching Adjusted +/- on top of the player data correlates well with predictions and / or actuals. Who has more of what they have done well or not well with? What strategies could the other coaches try to mess that up compared to neutral expectations and how likely is that? How is going to get fired / hired and when and with what impact?
Which teams by their language in the media or other signs sound and look above or below on unity, focus, confidence, etc?
Maybe look at pace or high pace - strong offense and low pace - strong defense subgroups.
If 3 point shooting is particularly volatile how dependent on that shot were teams last season compared to average? How reliant were they on 3 point shot defense and success there?
Did teams who rely on driving for shots and foulshots overachievers last season and this season and did it accelerate?
Any trend with high and low assist rate teams?
Bench strength and clarity.
If you looked at the game consistency and the clutch performance data, what can one make of it in general?
The strength of top 5-10 lineups and whether teams can field those from last season in this season or not. Or even player pair availability and expected usage.
Does shooting well and defending the shot well carry a greater in actual wins than it might appear from the season average data?
Does offensive possession usage on teams looking clearcut or it is a luxury of options or a potentially troubling competition / issue to sort out? Who handles that well and not- at player, coach and organization level?
What if you tried to predict injury frequency and injury impact? Would you gain edge on the competition or lose ground because it is too hard to do well? Is it too hard to do better than neutral?
What if you tried to predict trades to some degree? The lopsided salary dumps and the trade the future for now and now for the future ones? Could you at least add an estimated value impact * probability for the 4-10 trade situations with the most potential to occur? Is that to much or another example of what you might have to do to get edge?
What about big market / small market splits. It shouldn't matter in winning but does it affect predictions (Vegas and "the pack") and does it appear to affect winning at all?
Any new additions or subtractions to the "gets superstar calls" list?
I guess you could look at payroll too. If it is a proxy for GM valuations does it translates well or not? Model in the moral hazard dimension of player contract situation. And maybe anticipated team responses to signs of that or expectations of that. In-season team payroll and roster flexibility- cap space, exceptions, non-guaranteed contracts, open slots or slots easily opened if opportunity arises.
What about the benefits and moral hazards of GMs and owners?
If you put the team over and underachievers thru a complete wash, what stats are the subgroups of teams high or low on?
Any spike in win achievement among teams with 6+ rotation guys positive on Adjusted +/- or greater than expected loss from having a star with moderate to heavy negative Adjusted +/- or multiple such starters or rotation guys?
Defensive breakdown of teams with 3 or more bad rotation defenders or at least key lineup breakdowns with 2-3+ bad ones on the court?
If you spent a lot of time looking at and working with the data perhaps one could beat the average accuracy of typical minute projections too. Maybe especially for rookies and of course their predicted performance is tough but also an opportunity to exceed the norm.
How will player location changes affect their performance due to elevation change or degree or travel miles and rest levels or quality of medical staff or amount of weekend games or "team discipline" or city distractions or friendship reunions and breakups? Player familiarity with and success level against new conference and division and their familiarity with the player.
Who has good and well targeted player - mentor pairings and might expect to get more good things out of it especially from young players than without looking at that consideration?
There is probably more. Easier and quicker to list than to research and estimate impacts, but brainstorming is an early step in the process.
Last edited by Crow on Tue Mar 29, 2011 2:05 pm; edited 2 times in total
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 68
PostPosted: Tue Mar 29, 2011 11:13 am Post subject: Reply with quote
No, that would take way too long. Let's just look at last year's records (lol).
Agree that all of those and more could be influences. I think we'll find though that many of the positives will cluster (well-run teams make good trades, emphasize the best percentage strategies, pace their players well, hire coaches and mentors that mesh well with their talent, while poorly run teams don't). That may be at the heart of the consistency over the years with many of the same teams being in the same general spots over and over again.
Maybe looking only at the teams who had extreme changes from one year to the next will help isolate things. But, even then, you're dealing with small sample sizes in a very dynamic universe. Or, the answers may come from studying the quilts rather than focusing so closly on each thread of the quilt. Hard to say.
My take from the industry is that:
*Oddsmakers tend to look at the quilts with just a few little threads. What did this team do last year? What have they done the last few years? What's different coming into this season (new players or coaches)? They're not digging deep into stats/minutes/etc... If a team won 50 last year, and 48 the year before, and it's a veteran team with the same coach, the number will be in that range.
*Bettors (and it's almost exclusively the "sharps" who bet regular season win totals in the NBA because the public isn't into them) may focus more on the specifics you mention.
I know Haralabob pops in occasionally to this forum. Maybe he can shed some light on his thinking process about these propositions.
Definitely think the factors you've listed are very relevant. Maybe be tough to separate the things that go hand in hand...
Back to top
View user's profile Send private message
Crow
Joined: 20 Jan 2009
Posts: 816
PostPosted: Tue Mar 29, 2011 2:00 pm Post subject: Reply with quote
Thanks for the industry prespective.
Yes, a strong positive culture covers a lot of descriptive ground and it may not be necessary to deal with all the detail if you have a reasonable basis for some positive adjustment.
Though a strong positive culture manifests in specific ways and it might be worth it to a big bettor or a team to try to deal with all the complexity. Many positives (and even some neutrals) could add up to more than the sum of the parts so even a detailed but static review is probably not the ultimate review. The interactions are important. A massive dynamic model might be great if really well done. If time or expertise is insufficient to make it really great then using a simpler model would probably be safer.
Back to top
View user's profile Send private message
12 of 12
Crow
Joined: 20 Jan 2009
Posts: 817
PostPosted: Fri Apr 01, 2011 5:13 pm Post subject: Reply with quote
Very good question.
There appear to be several issues. One of them is using the vague term "Vegas line" in the discussion.
I first referenced http://www.sportsbook.com on Sept 22 in response to back2newbelf's question on page 3 of this thread. I believe it had an opening line but I can't see what that page said then, at least immediately.
Then on page 4 Ed Kupfer cited the Painted Area article on predictions http://thepaintedarea.blogspot.com/2010 ... tions.html
... and included Vegas values in his post. The Painted Area had a Vegas line of unspecified origin and date, I don't know if it was the opening line or the current line as of then. Given the Miami value of 64.5 I lean towards thinking it was the opening line of some book.
There are differences between what the PA article showed for Vegas and what Ed posted. It appears to me (now) that Ed rounded almost all of the PA stated Vegas line values down by .5 or in some cases 1 but not all for the list he posted in this thread.
I am not sure what back2newbelf used for Vegas in his tracking.
Last edited by Crow on Fri Apr 01, 2011 8:31 pm; edited 2 times in total
Back to top
View user's profile Send private message
Ed Küpfer
Joined: 30 Dec 2004
Posts: 786
Location: Toronto
PostPosted: Fri Apr 01, 2011 6:47 pm Post subject: Reply with quote
Crow wrote:
It appears to me (now) that Ed rounded almost all of the PA stated Vegas line values down by .5 or in some cases 1 but not all for the list he posted in this thread.
I can't remember exactly what I did there, but I usually adjust those numbers so the probabilities for all the teams sum to one.
_________________
ed
Back to top
View user's profile Send private message Send e-mail
Crow
Joined: 20 Jan 2009
Posts: 817
PostPosted: Fri Apr 01, 2011 8:25 pm Post subject: Reply with quote
I assumed that was the rationale and should have added that but I had something else come up to deal with.
If your listed values were used for contest error tracking, the adjustments represent a step away from that "Vegas line" but the Vegas values stated in the PA article summed to 18 wins more than available so something had to be done to address the issue by your approach or another.
If your listed values were not used for contest tracking, then your specific rounding approach is not an issue.
If the actual Vegas line with .5 win increments was used in the contest as is, then being able to pick between 2 win numbers was a possible advantage over most other predictions in the test. Most did not consider that an available or appropriate option. The difference in available choices for win predictions makes for a bit uneven average error "performance" comparison. If you want to sweat fractions or tiny fractions of a point of average error.
Last edited by Crow on Sat Apr 02, 2011 12:42 pm; edited 3 times in total
Back to top
View user's profile Send private message
haralabob
Joined: 11 Apr 2007
Posts: 27
PostPosted: Sat Apr 02, 2011 12:56 am Post subject: Reply with quote
The issue is these lines are rarely 50-50 propositions.
They'll have a line like;
Mem Over 41.5 -130
under +110
this probably equates to a line of around 42.5 and not 41.5 since you are penalized by betting on the over.
Either way I'd be fairly confident that the consensus line moves by the sharps who bet these season win totals (I don't) would probably beat most of the predictors out there.
Hollingers stuff looked pretty good tho.
Back to top
View user's profile Send private message
Jeff Fogle
Joined: 11 Jan 2011
Posts: 69
PostPosted: Sat Apr 02, 2011 10:33 am Post subject: Reply with quote
H, what do you think lies at the heart of the advantage sharps have over the various stat model "predictors?" I'd understand if you don't want to discuss something that might give away an advantage you have over the market down the road. But, if you're willing to discuss it, what elements could the models add into their mix that would improve their performance?
I know we're talking about various models and methodologies. But, at least in this thread, there are some prominent ones where the parameters are fairly well known. Why do they fall short in your opinion?