More recovered threads related to the draft

Home for all your discussion of basketball statistical analysis.
Post Reply
Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 12:40 am

Inclued thread titles (I probably missed some):

NCAA players and the draft

Evaluating the Draft Class

evaluating college players

euro-league stats

Draft Projection Accuracy?

2009 NBA Draft Grades

Who Had the Best Draft Day?

2009 NBA DRAFT - QSE ver7.0

Draft Pick Value

More


Hollinger's Draft Rater articles

Isiah Thomas' Draft Record 1995-2005


College PER and the draft

Blazers stat-based draft system



2009 NBA DRAFT - QSE ver7.0




JonathanG



Joined: 20 Apr 2005
Posts: 36


PostPosted: Wed Apr 20, 2005 3:57 pm Post subject: NCAA players and the draft Reply with quote
Hi there guys, first time posting here. I've been browsing for a little while now (I was chatting with Dan Rosenbaum today and he gave me the heads up on this place) and this appears to be a great little community you have going here with a wealth of fascinating information (that I may or may not always understand).

I've met a few of you in person in Chicago last year at the pre-draft camp, and know others through Hoopsworld, RealGM and maybe some back and forth occasional emailing. If you don't know me, I run the NBA draft site http://www.draftcity.com with some other people and do a lot of NCAA, European and High school player evaluations for Draftcity and other private sources.

I did a search here for the keyword "ncaa" and "draft" before I posted, trying to find if someone here already dabbled in this, and I found a little bit of interesting info, but not TOO much.

I hope Kevin doesn't get too pissed off for the spammy nature of this message, but I'm always looking for ways to make both myself and the website better in terms of scouting and evaluating talent for the draft, and the statistical part (which REALLY seems to be taking off lately) is obviously a place where I could or should be doing a better job at. I'm not as math oriented as most people here, so a lot of this stuff goes over my head. From reading some of the stuff that Kevin wrote for Hoopsworld when he wrote it and in hindsight (he looks like a genius in some instances, Marquis Daniels being the main one), I know that there is some value to this, its just a matter (as always) of doing it "right". Regardless, its always a fascinating read for people like me who want to get as many different perspectives as possible. We're all about actually.

What I'd ultimately like to accomplish is find some kind of middle ground between traditional scouting (watching a guy with your own eyes) and more cut and dry numbers type stuff. Everyone (whether its us or the actual decision makers) is always looking to get a leg up on the competition, and I think that it would be great to educate the many people who will be doing their own research on the draft, whether its the NBA and European teams who do a great deal of research (sometimes on DraftCity from what they've told me) but might need to be pushed a little into thinking outside the box, or just the casual fans who like looking at stats (especially when it tells them what they want to hear), but might not always have the patience or knowledge to read between the lines.

I'm not exactly sure how this would work, that's kind of why I am posting this very ambiguous message, but I would love to eventually mix some kind of statistical analysis with the scouting oriented stuff that we do over on DraftCity. In terms of reach, I don't think it would be a stretch to say that pretty much everyone in the business visits our site at least occasionally, along with a LOT of regular fans who are just doing their own research about who their team might pick. We've had 250,000 unique visitors so far since January 1st, almost 100,000 uniques in the 20 days of April, 200,000 visits this month and over half a million total visits on the year so far.

If anyone out there is already working on some statistical analysis that is somehow related to the draft, and might be looking for a professional place to publish it, we'd love to get involved somehow. Whether that means letting you do your thing on your own, getting in the mix ourselves with the numbers and coming up with stuff on our own, or whatever other ideas you might have, we'd love to talk about it.

In terms of compensation, financially there isn't TOO much we can offer (maybe something nominal) but besides the exposure and niche audience we can talk about maybe getting tickets to the draft, getting you access to certain events, connections with the right people to take this further (up to a certain point of course), or some other semi-attractive things that may or may not interest you. We've spent a helluva lot more money then we've made since this site started, foolishly thinking that people would be willing to support the site through donations (Evil or Very Mad) but we're willing to do what's needed to make this work.

Please share your thoughts about this (even if its just generally related to stats and the draft) either on this thread, via PM or by emailing me.

Thanks guys, and please don't ban me Kevin Very Happy
Last edited by Crow on Thu Jun 09, 2011 1:15 am, edited 1 time in total.

Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

Re: More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 12:43 am

Ben F.



Joined: 07 Mar 2005
Posts: 391


PostPosted: Sun Mar 27, 2005 9:13 pm Post subject: Evaluating the Draft Class Reply with quote
With the NCAA tournament winding down, it reminds us that not too far away are...the NBA playoffs - or for those fans whose teams are already down and out, the NBA Draft. Which of the stars of the Final Four will succeed in the NBA, and which will flare out?

Bill Simmons wrote an interesting article about this recently (although absolutely subjective), and it also reminds me of another discussion on the old newsgroup last year, but the question remains, how do you evaluate the upcoming draft class?

With college basketball, there's a great variability in the talent of the teams a lot of these programs play against. Is a breakout in the NCAA tournament a good measure, a la Dwyane Wade? How do you evaluate a Chris Taft, a sophomore with a lot of upside (at least it seems), who still hasn't put up the numbers expected of him in college?

I have found that ESPN has play-by-play logs for all top 25 games, and that you can usually find the play-by-play for almost any team on their college website - however, this doesn't include substitution information, so +/- is out of the question. What else can you do?

I guess I'll leave the question up to all you minds out there.
Back to top
View user's profile Send private message
jambalaya



Joined: 30 Jan 2005
Posts: 282


PostPosted: Mon Mar 28, 2005 8:26 pm Post subject: evaluating college players Reply with quote
there are a few college players that arent that hard to evaluate and project as pros. maybe the top 8 players or so. after that i agree it gets quite hard to judge by stats.

some rules of thumb i use:

if you dont rebound in double figures in college, you're not likely to rebound 7+ in the pros.

if you dont shoot 40% from the college three point line you're not likely to be a 35% NBA three point shooter

there are usually few good passers in the college game. unless you average 5 assists or more, dont tell me they are strong in that area. NBA backup points and tweeners these days seem more often the product of 2-4 years post graduation work - in the NBA, NBDL or kickin around and hopefully/eventually adjusting their mindset and playing style.

if i were able to evaluate game by game data, i'd throw out any game a big man played against under 6'10 guys who are decent talent opponents as not meaningful. the evaluation may end up being based just on their 6-12 games against size and talent that most closely resembles the nba.

guards i'd look closely for higher than average consistency game to game and strong fourth quarter play.

there is a lot more you could say on the topic of course. identifying good and bad habits. quality of instruction, feel for the game, etc.

i dont think the draft camps and individual workouts are that reliable but they seem to get weighted pretty heavily.

international players i no longer give any real credence to stats or draft site evaluations. i think it has to be personal opinion from seeing them play live and frequently by someone that has longtime experience evaluating players. i no longer pretend to be able to find clues who will and wont make it by looking at what's on paper about them.
Back to top
View user's profile Send private message
jeffpotts77



Joined: 18 Feb 2005
Posts: 150
Location: Cambridge, MA

PostPosted: Tue Mar 29, 2005 9:18 am Post subject: Reply with quote
If by some strange circumstance, there is anyone here who hasn't read John Hollinger's Pro Basketball Forcast, he has developed a system of evaluating Euro-league stats which he translates into per-40 min. nba production. It's really interesting.
Back to top
View user's profile Send private message Yahoo Messenger
jambalaya



Joined: 30 Jan 2005
Posts: 282


PostPosted: Wed Mar 30, 2005 3:23 pm Post subject: eurostat conversion Reply with quote
would you be willing to say a couple of sentences about the approach? is it based on the stats of all international players who have tried the nba or a smaller sample? or some simple rules of thumb?
Back to top
View user's profile Send private message
jeffpotts77



Joined: 18 Feb 2005
Posts: 150
Location: Cambridge, MA

PostPosted: Thu Mar 31, 2005 9:10 am Post subject: euro-league stats Reply with quote
I don't want to mis-represent his approach so if anyone notices that I'm missing something important, chime right in.

Basically it's based on Euro-league players who have played in the nba as well as nba players who later played in the Euro-league. He looks at how their stats changed from one stop to the other. On average, shooting percentages are higher in the Euroleague and rebounding totals are lower, and he simply translates them. I think you should read the Pro Basketball Forcast (w/ E. Brand on the cover) for a better explanation.
Back to top
View user's profile Send private message Yahoo Messenger

Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

Re: More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 12:45 am

Ilardi



Joined: 15 May 2008
Posts: 262
Location: Lawrence, KS

PostPosted: Thu Jul 31, 2008 2:56 pm Post subject: Draft Projection Accuracy? Reply with quote
I've enjoyed seeing draft projections this year from John Hollinger (and others), but have been wondering about the noise level inherent in the underlying regression models. John, when you derived your projected 3rd-year PER ratings, what were the typical standard errors of estimate? In other words, if we constructed a 95% confidence interval around your projections (e.g., Beasley at a PER of 19.19), what would those intervals look like?

Any information you - or others - can provide on this issue would be appreciated.
Back to top
View user's profile Send private message
bballfan72031



Joined: 13 Feb 2005
Posts: 54


PostPosted: Thu Jul 31, 2008 8:27 pm Post subject: Reply with quote
I've always thought using confidence intervals would be the best way to analyze prospects statistically.

I don't want to hijack the thread, but if anyone knows how to calculate confidence intervals in excel (preferably on individual outputs in regression analysis, such as TREND), any help would be immensely appreciated.
Back to top
View user's profile Send private message AIM Address
ErichDoerr



Joined: 06 Jul 2008
Posts: 15


PostPosted: Sat Aug 02, 2008 12:28 pm Post subject: Reply with quote
Ilardi,
I've done some work to try and recreate the Hollinger 3rd year PER projection. In this effort, my initial take away suggests the data set size alone should indicate a high noise level. In a previous thread here, John seems to have the same concerns, looking forward toward additional data points to improve accuracy.

From his articles, it appears John used data from the 2002 draft onward, which provides a sample size of ~230 players with NCAA backgrounds going to the NBA. From that group, only 115 had NBA playing time in the third year subsequent to their draft class, and 30 of them logged under 500 minutes. Given these requirements, I came up with a list of 86 players.

Breaking it down further, John took this group and split it among Bigs, Wings, and Points, giving me 27, 47, and 12 samples respectively. While regression analysis isn't my forte, these small sample sizes and the 16 explanatory variables may be enough in itself to give an experienced statistician a general idea of the level of noise present.

I have also done some work regressing the same 16 variables against John's listed results with some success, though my work is incomplete.

John, in case you happen to stop by, did you normalize height to position played or any other stat?
Back to top
View user's profile Send private message Send e-mail Visit poster's website
John Hollinger



Joined: 14 Feb 2005
Posts: 174


PostPosted: Mon Aug 04, 2008 6:43 am Post subject: Reply with quote
I actually included a bunch of "failures" in the sample as well -- if you only include players who played significant minutes in their third year the results will be biased as all hell. I put in anybody who got drafted plus everybody whose stats made them seem like OK prospects, so I had 100+ players at big and wing and something like 70 or 80 at the point. (all this work is sitting on my other computer so unfortunately I don't have it in front of me)

Obviously, how I handle the players who don't make it is very important to the method. For those who got a cup of coffee in the league but no more I gave them a PER of 9 or 9.5, for those who never even got a whiff I gave them a 7.5 I think . I didn't do anything special with height -- it was just one of the variables, expressed as inches above or below 6-0. .
Back to top
View user's profile Send private message AIM Address
Ilardi



Joined: 15 May 2008
Posts: 262
Location: Lawrence, KS

PostPosted: Tue Aug 05, 2008 12:28 pm Post subject: Reply with quote
John Hollinger wrote:
I actually included a bunch of "failures" in the sample as well -- if you only include players who played significant minutes in their third year the results will be biased as all hell. I put in anybody who got drafted plus everybody whose stats made them seem like OK prospects, so I had 100+ players at big and wing and something like 70 or 80 at the point. (all this work is sitting on my other computer so unfortunately I don't have it in front of me)

Obviously, how I handle the players who don't make it is very important to the method. For those who got a cup of coffee in the league but no more I gave them a PER of 9 or 9.5, for those who never even got a whiff I gave them a 7.5 I think . I didn't do anything special with height -- it was just one of the variables, expressed as inches above or below 6-0. .


John,

That's helpful info regarding your sample size, but it still doesn't address the question I raised regarding estimation errors . . . do you happen to have handy the standard errors of estimate from your model to pass along?
Back to top
View user's profile Send private message
John Hollinger



Joined: 14 Feb 2005
Posts: 174


PostPosted: Thu Aug 07, 2008 10:17 am Post subject: Reply with quote
Sorry for the delay, finally dug out the laptop and have the standard errors:

Point guards: 2.66
Wings: 2.82
Bigs: 3.04
Back to top
View user's profile Send private message AIM Address
Ilardi



Joined: 15 May 2008
Posts: 262
Location: Lawrence, KS

PostPosted: Thu Aug 07, 2008 11:33 am Post subject: Reply with quote
Many thanks.

So, it looks like there's a fair degree of "noise" in the predictive model right now, yes? For example, the 95% confidence interval around Beasley's projected 3rd year PER would range roughly from 13 to 25.

I wonder if these standard errors might not come down some in the future with more players in the database - or perhaps additional variables. Do you think adding athleticism/anthropometric data (standing reach, vertical leap, sprint, wingspan, etc.) from the pre-draft combine would be useful at all?
Back to top
View user's profile Send private message
John Hollinger



Joined: 14 Feb 2005
Posts: 174


PostPosted: Thu Aug 07, 2008 4:08 pm Post subject: Reply with quote
I have no doubt that the method will improve with more years of data. Right now I'm only working with six drafts, plus the past four are still basically works in progress. Then when you get down to the number of big men with talent like Beasley or Oden, you're down to a very small number very quickly.

I got the anthropomorphic data from the Orlando camp (standing reach, vertical, etc). and started playing with it but it's hard to incorporate because not everybody goes to the pre-draft camp, so it's a blank for the majority of players in the database.
Back to top
View user's profile Send private message AIM Address
tawtaw



Joined: 25 Jun 2008
Posts: 28
Location: Oregon

PostPosted: Thu Aug 07, 2008 7:37 pm Post subject: Reply with quote
John, are you using the draftexpress data base for that info?

I'm not sure how detailed it is as you go back six years, but it's really good for the more recent drafts.
Back to top
View user's profile Send private message
John Hollinger



Joined: 14 Feb 2005
Posts: 174


PostPosted: Thu Aug 07, 2008 8:57 pm Post subject: Reply with quote
I've got my sources, I'll leave it at that.

Perhaps a couple years down the road it will be useful because it seems like they're taking more measurements from more people. So far it's not, at least not in any way that I can divine.
Back to top
View user's profile Send private message AIM Address
tawtaw



Joined: 25 Jun 2008
Posts: 28
Location: Oregon

PostPosted: Thu Aug 07, 2008 11:07 pm Post subject: Reply with quote
Yeah, I could see how it'd be problematic if you didn't have the info for the majority of guys.

I'd really like to see a study someday about how well the pre-draft physical numbers do when it comes to predicting success.

I'd be willing to bet a PER-based analysis that factored in game pace and schedule strength of college prospects would predict NBA success every bit as well as combine numbers and draft position, if not better.
Back to top
View user's profile Send private message
ErichDoerr



Joined: 06 Jul 2008
Posts: 15


PostPosted: Sat Aug 09, 2008 12:06 pm Post subject: Reply with quote
John, have you considered using Game Score per minute to possibly increase your sample basis?
Back to top
View user's profile Send private message Send e-mail Visit poster's website
John Hollinger



Joined: 14 Feb 2005
Posts: 174


PostPosted: Sat Aug 09, 2008 9:23 pm Post subject: Reply with quote
Actually I create a PER for the entire NCAA.
Back to top
View user's profile Send private message AIM Address
jemagee



Joined: 05 Nov 2005
Posts: 129


PostPosted: Sat Aug 09, 2008 10:09 pm Post subject: Reply with quote
I wonder from a fan point of view (of the sixers and speights), how much your thoughts on Speights this season change after the season ending injury to Jason Smith (someone of whom i'm not fond of being on the sixers roster, i wish that scout that thought he's a 'starter' had convinced his team to trade for him)
Back to top
View user's profile Send private message
ErichDoerr



Joined: 06 Jul 2008
Posts: 15


PostPosted: Sun Aug 10, 2008 10:16 am Post subject: Reply with quote
John, perhaps I made a bad assumption or should clarify.

I thought your limitation in going back to NCAA seasons prior to 2002 was a lack of advanced NCAA data, including NCAA PER numbers. If this was the case, it may be possible to get the box score stats and assess a larger player pool based on Game Score.

If that is not your limitation in getting a bigger data pool, the suggestion has no bearing and I apologize for not being clearer in my earlier post.

Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

Re: More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 12:46 am

jmethven



Joined: 16 May 2005
Posts: 51


PostPosted: Sat Jun 27, 2009 7:12 pm Post subject: 2009 NBA Draft Grades Reply with quote
Yesterday I posted my 2009 NBA Draft grades for both conferences on my current blog (see my signature). To add some spice I ran every college prospect through a similarity scores database I have built over the last few summers. If anyone checks it out, I'd love some feedback and I could give more details into the system as well.
Back to top
View user's profile Send private message
Crow



Joined: 20 Jan 2009
Posts: 829


PostPosted: Sun Jun 28, 2009 4:19 am Post subject: Reply with quote
I'd be interested in hearing what attributes you included in the similarity system and the weights and why you choose them both.

With the similarity scores for the top guys being this close (looking) even small changes in the attributes and weights would have major consequences.



If the goal is projection I wonder if weighting the stat projection of a player by his similars' stat performances in the NBA should be done with recognition to the closeness of the draft rank of the player under study and the similar or, recognizing differing draft quality, to just the draft tier or perhaps if you went all out to adjusted draft rank, adjusted for the quality of the overall draft. At least if this is done after a draft. Before the draft I guess it could be done based on the consensus mock, maybe.

Where GMs draft players is imperfect, hit or miss but does show some level of of predictive power, I think, based on the retrospective Roland did at 82 games. Maybe it does better than not weighting by draft rank and tacitly assuming that draft rank doesn't matter as an indicator of future performance.

Or maybe you look at the top 5 similars and throw out one or two based on subjective judgment and take the straight average. Or do this and the weighting exercise outlined above.

Try these or other techniques and see which better match up to results for the timeframe and mass of players. It is fitting the model to the data but I'd guess more to be gained than lost by such adjustments if the timeframe / number of players is reasonably large as it appears to be. But maybe I am wrong.
Back to top
View user's profile Send private message
jmethven



Joined: 16 May 2005
Posts: 51


PostPosted: Sun Jun 28, 2009 2:16 pm Post subject: Reply with quote
Quote:
With the similarity scores for the top guys being this close (looking) even small changes in the attributes and weights would have major consequences.


Yes, this is absolutely the case. Especially looking at a narrow category like top 5 similar players, small changes can result in a completely different top 5. My categories were: Pts/40, Rbs/40, pure point rating, stls/pf, blks/pf, 3p/fga, 2p%, fta/fga, age, weight and height. I tried to use rate stats instead of per-minute stats when possible as I don't have team pace for all of the players in the database, which can obscure results.

I didn't weight anything because I strictly wanted to look at who was most similar. That may make the system less accurate for projection, I don't know.

Quote:
If the goal is projection I wonder if weighting the stat projection of a player by his similars' stat performances in the NBA should be done with recognition to the closeness of the draft rank of the player under study and the similar or, recognizing differing draft quality, to just the draft tier or perhaps if you went all out to adjusted draft rank, adjusted for the quality of the overall draft. At least if this is done after a draft. Before the draft I guess it could be done based on the consensus mock, maybe.


Yes, I thought about this and probably should have mentioned it more explicitly. Comparing draft rank seems like a good way to determine if a player would turn out better than his comps. Or comparing PER (ie, Zach Randolph is Blake Griffin's top comp but Griffin was drafted higher and had the better college PER - reasons why I think he will be the superior player).

I have a lot of data but ultimately, I'm not sure what the best way to look at it is. Or how to test its accuracy. Any more thoughts would be appreciated.
Back to top
View user's profile Send private message
Crow



Joined: 20 Jan 2009
Posts: 829


PostPosted: Mon Jun 29, 2009 2:45 am Post subject: Reply with quote
Pts/40, Rbs/40, pure point rating, stls/pf, blks/pf, 3p/fga, 2p%, fta/fga, age, weight and height

That is a fairly comprehensive list though there are some critiques that could be made-

Pts/40, Rbs/40 are affected by pace

Adding 3 pt FG% or using efg% instead of 2p% might be helpful

As for weight maybe they can be estimated by regression but simple cut I might do something like this

take the z-scores of the factors and weight
20 Pts/40, 10 Rbs/40, 5 pure point rating, 5 stls/pf, 5 blks/pf, 10 3p/fga, 10 2p%, fta/fga, 15 age, 5 weight and 15 height / 100

it is still offensive biased

or weight them differently as more appropriate for positions

heavier on pure point rating, stls/pf and 3p/fga for perimeter players, lighter on rebounds, blocks/pf and weight

lighter on pure point rating, stls/pf and 3p/fga for interior players, heavier on rebounds, blocks/pf height and weight and maybe age too (to try to capture defensive intelligence in a rough way)
Back to top
View user's profile Send private message
basketballvalue



Joined: 07 Mar 2006
Posts: 208


PostPosted: Tue Jul 07, 2009 10:06 am Post subject: Reply with quote
Interesting, I just caught up to this.

I noticed in your blog post that you listed the number of comps. How did you decide on the threshold for who is a comp and who is not?

Thanks,
Aaron
_________________
www.basketballvalue.com
Follow on Twitter
Back to top
View user's profile Send private message
jmethven



Joined: 16 May 2005
Posts: 51


PostPosted: Tue Jul 07, 2009 7:38 pm Post subject: Reply with quote
I cut off at 800, which is an arbitrary threshold, considering that the ranking system is arbitrary, but it was the point where I could get a good number of comps for all but a couple players. I am concerned that for players like Jordan Hill, saying he has 112 comps might not be very relevant. Then again, if I raised the threshold to cut his numbers down, a lot of other players (especially near the top of the draft) would have 5 or fewer valid comps, which wouldn't make the system very useful.

basketballvalue wrote:
Interesting, I just caught up to this.

I noticed in your blog post that you listed the number of comps. How did you decide on the threshold for who is a comp and who is not?

Thanks,
Aaron

Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

Re: More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 12:47 am

NickS



Joined: 30 Dec 2004
Posts: 384


PostPosted: Fri Jun 27, 2008 5:30 pm Post subject: Who Had the Best Draft Day? Reply with quote
Subjective or objective judgment, who had the best day -- judged against what one would expect from their situation going into the draft?

In alphabetical order:

GSW managed to balance an upside pick with Hendrix late in the draft. But does Hendrix fit in their up-tempo system? They may expect bigger things from Wright than either of their picks this draft

Houston didn't do anything flashy, but just managed their chips to try to come away with as much value as they could.

Miami may win by being stuck with Beasley, and I like Chalmers in the 2nd round.

Minnesota gets Kevin Love and Mike Miller which gives them a lot more offense immediately.

New Jersey gets two players, in Lopez and CDR who both went much later than expected.

Portland managed to come away with Bayless who had been considered a top-5 talent through most of the pre-draft season.
Back to top
View user's profile Send private message
Mountain



Joined: 13 Mar 2007
Posts: 1527


PostPosted: Sat Jun 28, 2008 1:42 pm Post subject: Reply with quote
Departing from the question as exactly pitched...
I think Miami had the best day, if you substitute "on draft only, regardless of position", because of Beasley. Chalmers was a nice kicker.


"-- judged against what one would expect from their situation going into the draft? -"

Rockets, Cavs and Nets I'd guess did among the best on draft assets acquired considering starting point.


Wolves and Pacers changed a lot overall. Not sure how it sorts out but their efforts might be up there too.

Pacers should be in thick of it for final playoff seeds. Not sure who falls short but I think they and the Bulls will put a lot of pressure on the incumbents. Wizards might be on the hottest seat? With free agency challenges and defensive rating.

Love fits better than Mayo. Overall the trade leaves them bigger. Still a jumble to organize effectively though and Wittman hasn't got strong evidence of ability to accomplish that. But McHale helped give him more of what he wanted for this next (last?) shot.
Back to top
View user's profile Send private message
jemagee



Joined: 05 Nov 2005
Posts: 129


PostPosted: Mon Jun 30, 2008 8:28 pm Post subject: Reply with quote
Me, cause the sixers picked the buy i called in february Smile I'm now 2 for 2 on february draft calls for the sixers, let's hope speights works out like that out guy i called from Arizona.

I believe portland did well because they came away with good players and good assets.

Must admit i thought seattle would be more active with thei 6 picks, was hoping the sixers would buy one of those 50K picks.

Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

Re: More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 12:49 am

Qscience



Joined: 22 Jun 2009
Posts: 70
Location: Phoenix, Arizona

PostPosted: Thu Jun 25, 2009 3:05 am Post subject: 2009 NBA DRAFT - QSE ver7.0 Reply with quote
I thought maybe I would place the QSE here for review:

1 Blake Griffin Oklahoma Iso-PF - O 222.6
2 James Harden Arizona State SG - Iso 211.1
3 Ty Lawson North Carolina TruePointG 187.1
4 Hasheem Thabeet Connecticut iso -Center 174.2
5 Terrence Williams Louisville CG - Def 164.9
6 Tyreke Evans Memphis CG - Def 161.0
7 Stephen Curry Davidson SG - Iso 160.5
8 Earl Clark Louisville SF - Def 143.1
9 Demar DeRozan USC Def-Wing -SF 127.8
10 B.J. Mullens Ohio State Center -high 118.8
11 DeJuan Blair Pittsburgh Iso-PF - O 112.7
12 Jordan Hill Arizona PF - Def 109.9
13 James Anderson Oklahoma State SG - Iso 105.1
14 Jonny Flynn Syracuse Pg >phys 103.9
15 Wayne Ellington North Carolina SG>neg 102.2
16 Jeff Pendergraph Arizona State PF 101.9
17 Gerald Henderson Duke Def-Wing -SF 98.6
18 Tyler Hansbrough North Carolina PF -wing 94.8
19 Derrick Brown Xavier SF - Iso 93.6
20 Jrue Holiday UCLA CG - Def 89.5
21 Nick Calathes Fr, 6-5 200(2008) Florida Pg = Paint 88.9
22 Chase Budinger Arizona SF - Range 88.7
23 Eric Maynor Virginia Commonwealth PG 88.4
24 Darren Collison So, 6-1(2007) UCLA TruePointG 86.4
25 Marcus Thornton LSU SG - Iso 85.6
26 Jodie Meeks Kentucky SG - role 82.3
27 Jeff Teague Wake Forest SG - Iso 79.9
28 Dante Cunningham Villanova SF - Def 77.0
29 Danny Green North Carolina Def-Wing -SF 75.8
30 Antonio Anderson Sr, Memphis CG - Def 74.7
31 DeMarre Carroll Missouri SF - paint 74.0
32 Jerel McNeal Marquette Pg >phys 72.5
33 Jerome Jordan Tulsa Center 71.1
34 Shawn Taggart Memphis Iso-PF - O 69.8
35 Josh Heytvelt Gonzaga IC/IPF 68.9
36 James Johnson Wake Forest SF - paint 68.7
37 Ahmad Nivins Saint Joseph's PF 68.4
38 Daniel Hackett USC CG - Def 68.2
39 Luke Nevill Utah Center 67.6
40 Sam Young Pittsburgh Def-Wing -SF 67.4
41 Austin Daye Gonzaga Def-Wing -SF 66.8
42 Robert Dozier Memphis Iso-PF - O 65.6
43 DaJuan Summers Georgetown IC/IPF 64.2
44 Patrick Mills Saint Mary's Pg >phys 64.1
45 Jon Brockman Washington PF - Def 63.0
46 Jeff Adrien Connecticut SF - paint 62.8
47 Alex Ruoff West Virginia Pg >phys 62.6
48 Dionte Christmas Temple SG>neg 62.2
49 Leo Lyons Missouri PF 61.8
50 Luke Harangody Notre Dame IC/IPF 61.4
51 Goran Suton Michigan State PF - Def 60.5
52 Wesley Matthews Marquette SF - paint 59.2
53 John Bryant Santa Clara Center 56.2
54 Micah Downs Gonzaga SG 56.2
55 Ibrahima Thomas Oklahoma State Center - Def 55.8
56 Jermaine Taylor UCF SG>neg 55.1
57 A.D. Vassallo Virginia Tech SG - role 54.8
58 Brandon Costner Fr, 6-9 North Carolina State IC/IPF 54.7
59 Taj Gibson USC Per/Block 54.5
60 Paul Harris Syracuse Def-Wing -SF 54.5
61 Byron Eaton Oklahoma State Pg >phys 54.2
62 A.J. Price Connecticut Pg >phys 53.9
63 Lester Hudson Tennessee-Martin Pg >phys 53.8
64 Charles Little Dayton SF - paint 53.3
65 Toney Douglas Florida State SG>neg 50.8
66 Connor Atchley Texas IC/IPF 50.3
67 Josh Carter , 6-7 Texas A&M SG - role 49.7
68 Sherron Collins Kansas Pg >phys 49.3
69 Jeff Allen Virginia Tech PF - Def 49.3
70 Aaron Jackson Duquesne Pg >phys 49.3
71 Lee Cummard Brigham Young SG>neg 48.9
72 JaMychal Green Alabama Per/Block 48.2
73 Arinze Onuaku Syracuse PF - Def 47.8
74 Tyrese Rice So, 6-0 Boston College Pg >phys 47.4
75 Ben Woodside North Dakota State Pg >phys 47.2
76 Paul Delany UAB sg 46.9
77 Patrick Beverley Fr, 6-1 Arkansas CG - Def 46.8
78 Kyle Spain San Diego State Def-Wing -SF 46.7
79 Curtis Jerrells Baylor Pg >phys 46.4
80 Dominic James So, 5-11(2007) Marquette Pg >phys 46.0
81 Chinemelu Elonu Texas A&M Iso-PF 45.7
82 Jeremy Pargo jr, 6-2, 208 Gonzaga Pg >phys 45.4
83 Chas McFarland Wake Forest PF - Def 44.9
84 Lawrence Hill So, 6-8 Stanford Def-Wing -SF 44.4
85 Gary Wilkinson Utah State PF -wing 43.7
86 Joe Krabbenhoft Wisconsin PF - Def 43.4
87 Ronald Steele So, 6-3 (2006) Alabama Pg >phys 43.4
88 Lorrenzo Wade San Diego State Def-Wing-SG 43.3
89 Michael Bramos Miami (OH) SG - role 42.4
90 Aron Baynes Washington State Center -high 42.2
91 Stefon Jackson UTEP SG - role 41.7
92 Eric Devendorf Syracuse SG - role 41.7
93 Jack McClinton Miami (FL) SG>neg 40.5
94 Robert Vaden UAB SG - role 40.5
95 Russell Robinson Sr, 6-2, 198 Kansas Pg >phys 40.4
96 Garrett Temple LSU Pg >phys 40.0
97 Lorrenzo Wade Jr, 6-6 San Diego State ComboGuard 39.6
98 David Padgett Jr, 7-0, 240 Louisville Center -high 39.5
99 Tony Gaffney Massachusetts SF - paint 39.3
100 Tasheed Carr Saint Joseph's Pg >phys 39.2
101 Hamady N'Diaye Rutgers Per/Block 39.1
102 Alade Aminu Georgia Tech PF - Def 39.0
103 Jeremy Chappell Robert Morris SG - role 38.7
104 Bryan Davis Texas A&M PF - Def 38.2
105 Alfred Aboya UCLA PF - Def 38.1
106 Korvotney Barber Auburn SF - Def 38.1
107 Gavin Grant Jr, 6-8 North Carolina State SG - role 38.1
108 Chester Frazier Soph, 6-2 Illinois Pg = Paint 37.8
109 Chris Johnson LSU Per/Block 37.8
110 Damian Johnson Minnesota SF - paint 36.9
111 Josh Shipp So, 6-5 UCLA SG - role 36.2
112 Dior Lowhorn San Francisco SF - paint 36.0
113 Kyle McAlarney Notre Dame Pg >phys 35.3
114 Deonta Vaughn Cincinnati Pg >phys 34.8
115 Ryan Toolson Utah Valley SG - role 34.6
116 Weyinmi Efejuku Providence SG - role 34.3
117 Landon Milbourne Maryland SF - paint 33.3
118 Edwin Ubiles Siena SF - paint 33.3
119 Chris Lowe Massachusetts Pg >phys 32.9
120 Vernon Goodridge La Salle Per/Block 32.4
121 Jeremy Hazell Seton Hall SG - role 32.1
122 Trent Meacham Illinois Pg >phys 32.0
123 Chamberlain Oguchi Illinois State SG - role 31.2
124 Taylor Rochestie Washington State Pg >phys 30.5
125 Lewis Clinch Georgia Tech SG - role 30.4
126 Rob Diggs George Washington Per/Block 30.4
127 Tiki Mayben Binghamton Pg >phys 30.2
128 Perry Stevenson Kentucky Per/Block 30.2
129 Marquise Kately Morgan State SG - role 30.0
130 Trevon Hughes Wisconsin SG - role 29.8
131 Kenny Hasbrouck Siena SG - role 29.6
132 Jeremy Mayfield Fr, 6-10 UAB PF - Def 29.4
133 Jeremy Wise Southern Miss Pg >phys 29.2
134 Tommie Liddell So, 6-4 Saint Louis SG - role 28.2
135 Ron Steele Alabama Pg >phys 27.5
136 Robert Mitchell Seton Hall SG - role 27.1
137 Bamba Fall So, 7-1 Southern Methodist Per/Block 27.0
138 Edgar Sosa Fr, 6-1 Louisville Pg >phys 27.0
139 Courtney Fells So, 6-5 North Carolina State SG - role 26.2
140 Keith Cothran Rhode Island SG - role 26.0
141 Uche Echefu So, 6-9 Florida State PF - Def 25.7
142 Wink Adams UNLV Pg >phys 25.5
143 Ramar Smith 6-2 Tennessee CG - Def 24.9
144 Taquan Dean --, -- Louisville Pg >phys 23.3
145 JR Inman So, 6-9 Rutgers PF - Def 23.3
146 Jeremis Smith Jr, 6-8 Georgia Tech SG - role 23.2
147 A.J. Abrams Texas Pg >phys 22.9
148 Juan Palacios Jr, 6-8 Louisville IC/IPF 22.6
149 Mike Mercer South Florida SG - role 21.9
150 Derrick Low Sr, 6-2 Washington State SG - role 20.7
151 Curtis Jerrells So, 6-1 Baylor Pg >phys 20.6
152 Brian Randle So, 6-8 Illinois Def-Wing -SF 20.3
153 Marcus Johnson Fr, 6-3 Dayton SG - role 16.8
154 Justin Hare Fr, 6-3 Belmont SG - role 13.3
155 Joey Henley Fr, 6-5 Sacred Heart Def-Wing -SF 13.1
156 Aaron Bruce Jr, 6-3 Baylor SG - role 11.7
157 Rob McKiver 6-3, 190 Houston SG - role 10.9

Last edited by Qscience on Sat Jun 27, 2009 12:31 am; edited 1 time in total
Back to top
View user's profile Send private message Visit poster's website
Crow



Joined: 20 Jan 2009
Posts: 812


PostPosted: Thu Jun 25, 2009 3:47 am Post subject: Reply with quote
Williams, DeRozan and Mullens seem rated high to me for their performance and for what I expect will happen but I haven't studied them closely. Thabeet high compared to what I expect in the NBA. Hansbrough might be low compared to what he actually does and I'd say certainly low compared to what he did in college.

Most of the rest of the top 20 looks pretty decent but that doesn't take an accurate analytic method.

Last edited by Crow on Fri Jun 26, 2009 5:30 am; edited 1 time in total
Back to top
View user's profile Send private message
supersub15



Joined: 21 Sep 2006
Posts: 273


PostPosted: Thu Jun 25, 2009 7:51 am Post subject: Reply with quote
Why do the numbers keep changing?
Back to top
View user's profile Send private message
Qscience



Joined: 22 Jun 2009
Posts: 70
Location: Phoenix, Arizona

PostPosted: Thu Jun 25, 2009 11:46 am Post subject: Reply with quote
supersub15 wrote:
Why do the numbers keep changing?


The biggest problem is getting inside information on players that did not attend the combine. Usually a few days before the draft I get some leaked information about particular things I ask for. When the team knows they will not draft that player usually I get a few tidbits.
Same holds true about a month later in Vegas, I will get a few more tidbits that will help out even though the draft is over it couldve possible helped a player by a few spots. Beyond that I will say that what i copied and pasted is the final.
Back to top
View user's profile Send private message Visit poster's website
Qscience



Joined: 22 Jun 2009
Posts: 70
Location: Phoenix, Arizona

PostPosted: Sat Jun 27, 2009 12:32 am Post subject: Reply with quote
Summer Rosters soon to be out
Back to top
View user's profile Send private message Visit poster's website
Crow



Joined: 20 Jan 2009
Posts: 812


PostPosted: Fri Aug 07, 2009 9:46 pm Post subject: Reply with quote
At this point your 2010 evaluations show only 4 guys over 110, fewer than any year you've listed. Part of this I assume is that the incoming freshman are not included because they do not yet have college stats to evaluate and I'd guess that the returning players show improvement over last season. But in your view how does this upcoming draft compare to previous classes? Do you recall or have evaluation numbers for them at a similar point in time?
Back to top
View user's profile Send private message
Qscience



Joined: 22 Jun 2009
Posts: 70
Location: Phoenix, Arizona

PostPosted: Sat Aug 08, 2009 6:41 pm Post subject: Reply with quote
Here were the updated totals which are not even on the website. I did not change them because I simply was doing a lot of traveling from Vegas Summer Leagues to a couple private meetings hoping to land a future consultant position.
Code:
rank
player
team
pos
Intangible
Evaluation
1
Evan Turner
Ohio State
CG - Def
10.2
162.0
2
Greg Monroe
Georgetown
PF
3.5
157.5
3
Manny Harris
Michigan
SG - Iso
13.7
151.0
4
Kalin Lucas
Michigan State
ComboGuard
9.1
127.6
5
Devin Ebanks
West Virginia
SF - Def
8.4
106.2
6
Willie Warren
Oklahoma
SG>neg
1.8
97.7
7
Cole Aldrich
Kansas
iso -Center
-3.3
95.7
8
D.J. Kennedy
St. John's
CG - Def
6.3
95.3
9
Omar Samhan
Saint Mary's
iso -Center
-4.2
94.5
10
Al-Farouq Aminu
Wake Forest
SF - Def
0.8
93.8
11
Patrick Patterson
Kentucky
PF
1.1
92.4
12
Demetri McCamey
Illinois
Pg = Paint
-1.2
88.3
13
Gordon Hayward
Butler
SF - Range
0.3
87.3
14
Mike Davis
Illinois
Per/Block
2.5
84.4
15
Trevor Booker
Clemson
Iso-PF
-1.8
83.2
16
Robbie Hummel
Purdue
SF - Def
-1.1
80.5
17
Craig Brackins
Iowa State
IC/IPF
2.6
79.1
18
Andrew Ogilvy 7-0 240
Vanderbilt
iso - Center - o
-8.0
77.3
19
K.C. Rivers So, 6-5
Clemson
CG - Def - r
0.3
75.6
20
Terrico White
Mississippi
SG - role
-3.3
74.5
21
Malcolm Delaney
Virginia Tech
SG - Iso
3.4
74.0
22
Damion James So, 6-7
Texas
Defense-only
3.3
73.7
23
Jarvis Varnado
Mississippi State
Defense-only
-3.8
73.6
24
Chris Singleton
Florida State
Defense-only
-5.9
73.2
25
Tyler Smith
Tennessee
ComboGuard
4.1
73.2
26
Matt Bouldin
Gonzaga
CG - Def
-2.2
72.5
27
Kenneth Faried
Morehead State
Defense-only
2.8
72.2
28
Dominique Jones
South Florida
ComboGuard
-2.1
72.2
29
Diamon Simpson
Saint Mary's
Defense-only
-2.9
70.6
30
JaJuan Johnson
Purdue
Per/Block
2.6
70.0
31
Marqus Blakely
Vermont
SF - Def
-1.1
69.8
32
Matt Howard
Butler
IC/IPF
-6.1
68.8
33
Stanley Robinson
Connecticut
SF - Def
-8.9
68.0
34
Chris Kramer
Purdue
Defense-only
-10.1
67.0
35
Kyle Singler
Duke
SF - Iso
-2.8
66.7
36
Ryan Thompson
Rider
SG - role
-0.2
66.6
37
Mike Tisdale
Illinois
Center -high
-9.0
66.1
38
Chris Wright
Georgetown
ComboGuard
-6.0
65.5
39
Ekpe Udoh Fr, 6-10
Michigan
Defense-only
-18.3
63.9
40
Ben McCauley So, 6-10
North Carolina State
PF - Def
-7.2
63.6
41
Sylven Landesberg
Virginia
Def-Wing-SG
0.6
62.2
42
Samardo Samuels
Louisville
PF
-1.2
62.0
43
Michael Dunigan
Oregon
3pt-Specialist
-20.6
61.8
44
Eric Hayes 6-4 195
Maryland
Pg >phys
-7.8
61.6
45
Solomon Alabi
Florida State
Defense-only
-9.5
61.3
46
Damian Saunders
Duquesne
Defense-only
-6.1
60.4
47
Wayne Chism
Tennessee
IC/IPF
-14.3
60.2
48
Larry Sanders
Virginia Commonwealth
Defense-only
-5.4
60.0
49
Jon Scheyer
Duke
3pt-Specialist
-5.6
58.0
50
Klay Thompson
Washington State
3pt-Specialist
-3.5
58.0
51
Gani Lawal
Georgia Tech
Defense-only
-5.0
57.9
52
Mark Payne
UC Davis
Pg >phys
-12.0
57.5
53
Dallas Lauderdale
Ohio State
PF - Def
-12.0
57.3
54
Da'Sean Butler
West Virginia
SF - paint
-3.5
57.3
55
Austin Freeman
Georgetown
ComboGuard
-6.2
57.2
56
Kevin Coble
Northwestern
SF - Range
-13.4
56.9
57
Iman Shumpert
Georgia Tech
Defense-only
-1.2
56.9
58
Jon Diebler
Ohio State
3pt-Specialist
-11.4
56.5
59
Scottie Reynolds
Villanova
Pg >phys
-4.4
55.1
60
Terrence Oglesby
Clemson
3pt-Specialist
-7.2
55.0
61
Wayne Chism So, 6-9, 245
Tennessee
Defense-only
-13.4
54.9
62
Jon Scheyer So, 6-5
Duke
3pt-Specialist
1.3
54.6
63
Jonathan Tavernari
Brigham Young
Defense-only
-4.4
54.1
64
Kemba Walker
Connecticut
CG - Def
-3.6
54.1
65
Damian Johnson
Minnesota
Defense-only
-11.3
54.0
66
Luke Babbitt
Nevada
SF - Range
-4.3
54.0
67
Ravern Johnson
Mississippi State
3pt-Specialist
-11.2
53.8
68
Obi Muonelo
Oklahoma State
SG - role
-2.2
53.3
69
Talor Battle
Penn State
Pg = Paint
-22.7
53.3
70
Josh Alexander So, 6-4
Stephen F. Austin
3pt-Specialist
-7.3
53.0
71
Lazar Hayward
Marquette
SG - role
-5.7
52.9
72
Brandon Ewing
Wyoming
Pg >phys
-4.9
52.4
73
Al Nolen 6-1 180
Minnesota
Defense-only
3.2
52.4
74
Greg Paulus jr, 6-2
Duke
3pt-Specialist
-1.1
52.1
75
E'Twaun Moore
Purdue
Defense-only
-6.8
52.0
76
Deon Thompson
North Carolina
Defense-only
-10.5
51.5
77
Paul George
Fresno State
Defense-only
0.1
51.5
78
Chandler Parsons
Florida
Def-Wing -SF
-12.7
51.2
79
Greivis Vasquez
Maryland
Pg=Playmaker
-9.1
51.2
80
Rakim Sanders
Boston College
Defense-only
-5.1
50.7
81
Cory Higgins
Colorado
Def-Wing-SG
0.3
50.6
82
Gerald Lee
Old Dominion
PF -wing
-5.4
50.6
83
DeShawn Sims
Michigan
PF - Def
-10.4
50.1
Back to top
View user's profile Send private message Visit poster's website
gabefarkas



Joined: 31 Dec 2004
Posts: 1313
Location: Durham, NC

PostPosted: Mon Aug 10, 2009 3:15 pm Post subject: Reply with quote
Qscience wrote:
Here were the updated totals which are not even on the website. I did not change them because I simply was doing a lot of traveling from Vegas Summer Leagues to a couple private meetings hoping to land a future consultant position.

Is this in addition to your working "with a couple NBA teams front office personnel" in which you consult with them "on many things"? I must say, I envy your time management skills.
Back to top
View user's profile Send private message Send e-mail AIM Address
Qscience



Joined: 22 Jun 2009
Posts: 70
Location: Phoenix, Arizona

PostPosted: Mon Aug 10, 2009 6:56 pm Post subject: Reply with quote
gabefarkas wrote:
Qscience wrote:
Here were the updated totals which are not even on the website. I did not change them because I simply was doing a lot of traveling from Vegas Summer Leagues to a couple private meetings hoping to land a future consultant position.

Is this in addition to your working "with a couple NBA teams front office personnel" in which you consult with them "on many things"? I must say, I envy your time management skills.


Gabe,
I have something I will send you privately hopefully stop your skepticism

Last edited by Qscience on Fri Aug 14, 2009 10:07 am; edited 2 times in total
Back to top
View user's profile Send private message Visit poster's website
Qscience



Joined: 22 Jun 2009
Posts: 70
Location: Phoenix, Arizona

PostPosted: Fri Aug 14, 2009 9:58 am Post subject: Reply with quote
..
Back to top
View user's profile Send private message Visit poster's website
battaile



Joined: 27 Jul 2009
Posts: 38


PostPosted: Fri Aug 14, 2009 6:30 pm Post subject: Reply with quote
Wow, the tone of some of these posts are pretty surprising. Keep up the good work Qscience.
Back to top
View user's profile Send private message Visit poster's website
Qscience



Joined: 22 Jun 2009
Posts: 70
Location: Phoenix, Arizona

PostPosted: Fri Aug 14, 2009 8:59 pm Post subject: Reply with quote
battaile wrote:
Wow, the tone of some of these posts are pretty surprising. Keep up the good work Qscience.


Well Battaile to be honest Gabe I think is defending this forum to an extent from people that come in and make claims without backing them up in public. I can respect that but also at the same point I was invited to the forums and asked to explain some of the things that I have done.
At that point, all I could say is that I am working with some front office people and that alot of what I have done was asked to be kept confidential.
In the end it leaves me on an island if someone gets aggressive and pushes for answers. Without divulging any confidential information I have possibly found a way to show him some information that "may" take away some of the skepticism without sharing confidential information.
If you go through the forum post it is littered with people that have had an influence on APBR or an NBA team in some way or another.
Besides the fact there are a ton of people with incredible websites that I cannot even imagine how long it took to make or what it takes to maintain. The more people that can succeed in this field the more credibility it will give to the field.

Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

Re: More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 12:56 am

DSMok1



Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains

PostPosted: Thu May 13, 2010 4:17 pm Post subject: Draft Pick Value Reply with quote
Justin Kubatko worked up a method for valuing draft picks at the Basketball Reference Blog last year, and I thought I would revise and expand upon it.

I went through every draft since the 1979 draft (that had 4 years completed) and compiled each player's win shares. However, I felt win shares didn't accurately reflect the player's value to the organization, so I implemented a replacement level of 0.025 WS/48. (That equals 10.25 wins per season). Then I calculated the Win Shares above Replacement for each player drafted, over their first 4 years.

Similar to Justin, I fitted a logarithmic curve to the data--but I didn't feel the curve accurately reflected the linearization of the second round, so I added a linear component as well. This I fitted using weighted least-squares regression, according to another curve estimating the standard deviation of each draft pick.

Here are the results:

WS above Replacement Estimate: y = -5.82652 * ln(0.83085*x - 0.02517) + 0.05883 * x + 20.10883
(R^2 = .922)

Std. Dev. Estimate: y = -1.3018 * ln(0.9812*x - 0.0089) - 0.0016 * x + 7.7374
(R^2 = 0.5)

The resulting graph:

Image
The error bars are equal to +/- 1 standard deviation.

And here are the results in tabular form:
Code:
Pick WS over Rep. Stdev
1 21.43 7.77
2 17.36 6.86
3 15.02 6.33
4 13.39 5.95
5 12.14 5.66
6 11.13 5.42
7 10.29 5.22
8 9.57 5.04
9 8.94 4.89
10 8.38 4.75
11 7.88 4.62
12 7.43 4.51
13 7.02 4.40
14 6.65 4.31
15 6.30 4.21
16 5.99 4.13
17 5.69 4.05
18 5.42 3.97
19 5.16 3.90
20 4.92 3.83
21 4.69 3.77
22 4.48 3.70
23 4.28 3.64
24 4.09 3.59
25 3.91 3.53
26 3.74 3.48
27 3.58 3.43
28 3.43 3.38
29 3.28 3.33
30 3.14 3.29
31 3.01 3.24
32 2.88 3.20
33 2.76 3.16
34 2.65 3.12
35 2.54 3.08
36 2.43 3.04
37 2.33 3.00
38 2.23 2.97
39 2.14 2.93
40 2.05 2.90
41 1.97 2.86
42 1.89 2.83
43 1.81 2.80
44 1.73 2.77
45 1.66 2.74
46 1.59 2.71
47 1.52 2.68
48 1.46 2.65
49 1.40 2.62
50 1.34 2.59
51 1.28 2.56
52 1.23 2.54
53 1.18 2.51
54 1.13 2.48
55 1.08 2.46
56 1.03 2.43
57 0.99 2.41
58 0.95 2.38
59 0.90 2.36
60 0.87 2.34


Perhaps soon, this can be incorporated into a model that also looks at the value of the controlled salaries of the players on a rookie contract.

Last edited by DSMok1 on Thu May 20, 2010 7:43 am; edited 2 times in total
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DSMok1



Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains

PostPosted: Fri May 14, 2010 9:55 am Post subject: Reply with quote
Another way to look at this is to evaluate the minimum, median, and maximum production from each pick over the 25 or so years. The following chart does that, including also the average/mean production from each pick as well.

At pick 45, the median production reaches 0; in other words, more than half of the players picked 45th or worse never produce any wins.

Note that the minimum wins actually drops as the picks get higher--the top picks get a bigger chance to prove themselves.

Image

Basically, the reason #45 is worth more than #60 is there is more "upside".
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DSMok1



Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains

PostPosted: Fri May 14, 2010 10:19 am Post subject: Reply with quote
And finally, here is a graph of the minimum, 25th percentile, median, 75th percentile, and maximum for each draft pick.

I only show the estimates, to avoid clutter.

Image

Below about pick 50, even the 75th percentile is at 0. In other words, don't expect them to play much! Also, many of the players in the 50s that produced did so after being stashed overseas for a while. I still measured their first 4 years in the NBA, for comparison's sake.

Here is the unsmoothed data:

Image
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DSMok1



Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains

PostPosted: Fri May 14, 2010 10:34 am Post subject: Reply with quote
One more visualization:

Image
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Mike G



Joined: 14 Jan 2005
Posts: 3617
Location: Hendersonville, NC

PostPosted: Fri May 14, 2010 3:38 pm Post subject: Reply with quote
Pretty cool. On the unsmoothed chart, odd-numbered picks are better than the previous picks.

Quote:
I felt win shares didn't accurately reflect the player's value to the organization, so I implemented a replacement level of 0.025 WS/48. (That equals 10.25 wins per season)

The '93 Mavs won 11 games (Pyth 10) with a lot of literal 'replacement' types. But they also had Derek Harper, pretty close to his prime.

Wouldn't any expected wins be of value to the organization? We've never seen a team of pure replacements, but they'd probably win a game or 2, not 10.

Anyway, how does it affect the curves when you set RL to .010 or .000 ?

This year, Monta Ellis was .023, Corey Brewer .020. They led their teams in minutes. No value? Replaceable players?
In all, 25 players who got at least 500 minutes, 56 who got at least 100, were under .025 WS/48.

Even at .000 WS/48 replacement level, you have 46 guys appearing; 3 (all Nets) went over 1000 minutes.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
DSMok1



Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains

PostPosted: Fri May 14, 2010 4:28 pm Post subject: Reply with quote
Mike G wrote:
Pretty cool. On the unsmoothed chart, odd-numbered picks are better than the previous picks.

Quote:
I felt win shares didn't accurately reflect the player's value to the organization, so I implemented a replacement level of 0.025 WS/48. (That equals 10.25 wins per season)

The '93 Mavs won 11 games (Pyth 10) with a lot of literal 'replacement' types. But they also had Derek Harper, pretty close to his prime.

Wouldn't any expected wins be of value to the organization? We've never seen a team of pure replacements, but they'd probably win a game or 2, not 10.

Anyway, how does it affect the curves when you set RL to .010 or .000 ?

This year, Monta Ellis was .023, Corey Brewer .020. They led their teams in minutes. No value? Replaceable players?
In all, 25 players who got at least 500 minutes, 56 who got at least 100, were under .025 WS/48.

Even at .000 WS/48 replacement level, you have 46 guys appearing; 3 (all Nets) went over 1000 minutes.


Heh. I guess I should have explained that better. Basically, that's the threshold used by Kevin Pelton for his WARP data; I think it's reasonable. The concept is that players available for "minimal" cost would combine to win about 10 games. For instance, any players that are basically bench sitters for average teams. I'm sure you're familiar with Wins Above Replacement in baseball, where the concept is far more precisely articulated and developed.

Also, don't forget that all of these "player valuation" systems are fallible. I chose Win Shares simply because that's what Neil had used in his previous system.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DJE09



Joined: 05 May 2009
Posts: 148


PostPosted: Tue May 18, 2010 6:13 am Post subject: Reply with quote
I looked into this a little after Mike G did some posting about it early in the last off-season, and got patterns similar to what you were presenting. However, I got to asking some questions:

1. What impact the number of teams drafting has?
For example the Spurs famously took Tony Parker with the 28th pick in round 1 of the 2001 Draft, but before 1995 there weren't 28th picks in the first round, so instead of being one of the best teams in the NBA, it was one of the worst teams, making their second pick. Also if you look back into the 80s when they had draft until you drop drafts, you see the odd player like Manute Bol being drafted 97th in 1983, amongst a whole heap of people who never even got burn in the NBA. The issue of expansion over the last 30 years makes any sort of analysis based on draft position really messy.

2. What impact does team drafting have on the quality of picks at that position?
For instance, it doesn't take much to realise that the Spurs and the Clippers have significantly different drafting success. I realise we probably don't have enough data to do that sort of analysis. But it makes our data much noisier. Neil, did you look at this, or was it just a wishful phantom memory I have?

2a. A corrollary to this, is when a team has a much higher pick than usual / trades 'up' to get an earlier pick to get a particular player. Does that tend to produce an above average player? I can't think of many successful instances of this off the top of my head. But I can think of times when teams got themselves two high draft picks which didn't really seem to set them up (Chicago Chandler and Curry spring to mind)...

3. Draft Depth. It is pretty clear that in general the Second Pick is much harder to take than the First. By this I mean, if we "know" the best player, they're gone by the 2nd pick, and if we don't know who the best player is in a draft, often we don't really know who the second best player is either (Durant / Oden being the except that proves this rule). Further, historically on average picks 2-5 generate about the same amount of WS. I think that it could be argued that this is as much about teams, what they (think they) need most to be (more) competative as much as talent. This is infamous in the Bowie over Jordan selection. There is only a limited amount of great talent, so by the time you get to the 5th pick, they have all been taken. (LBJ et al providing evidence for this assertion). The depth varies from draft to draft adding to the noise.

I really don't know what to think about all this. I feel it is a chicken and egg thing. Are players that talented when they are drafted, or do they develop into the productive player because of the situation they ended up in? Would De Jaun Blair have worked at, say Golden State (I'll go out on a limb and say he'd have gotten injured)? Would Kobe have been a flop if he was drafted by New Jersey instead of Kerry Kittles (kittles played about twice as many minutes as Kobe in 1st 2 seasons, whilst being similarly productive on a per minute basis...)? It has to be a bit of both. But that is just opinion. Case and point Brandon Jennings. Lots of teams getting slammed for not picking him up, but perhaps his development is in part of being in Milwalkee. How do even begin to seperate out those factors? Would Tyreke Evans have had his rookie season at the Clippers or Memphis? Agent 0 is imfamously immature in his crediting his motivation for his success (sic) to his low draft position.

Oh, yeah, If you start post 1985 draft, positions 3 and 5 don't look all that hot ...
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: Tue May 18, 2010 9:24 am Post subject: Reply with quote
DJE09 wrote:
I looked into this a little after Mike G did some posting about it early in the last off-season, and got patterns similar to what you were presenting. However, I got to asking some questions:

1. What impact the number of teams drafting has?
For example the Spurs famously took Tony Parker with the 28th pick in round 1 of the 2001 Draft, but before 1995 there weren't 28th picks in the first round, so instead of being one of the best teams in the NBA, it was one of the worst teams, making their second pick. Also if you look back into the 80s when they had draft until you drop drafts, you see the odd player like Manute Bol being drafted 97th in 1983, amongst a whole heap of people who never even got burn in the NBA. The issue of expansion over the last 30 years makes any sort of analysis based on draft position really messy.


I would argue that, in general, teams are taking the best player available/the best player for their needs, so the issue of whether it is a good team drafting or a bad team drafting is moot. The issue of high schooler vs. college star is interesting. I only look at the years covered by a "rookie contract," so the college stars would have an advantage. However, isn't it also the case that in general the teams would have to pay close to market value after the first contract anyway, thus mitigating the advantages of having picked the player?

DJE09 wrote:
2. What impact does team drafting have on the quality of picks at that position?
For instance, it doesn't take much to realise that the Spurs and the Clippers have significantly different drafting success. I realise we probably don't have enough data to do that sort of analysis. But it makes our data much noisier. Neil, did you look at this, or was it just a wishful phantom memory I have?


We know this sort of thing after the fact, but I don't know how much of an effect we could have assumed before the draft. In other words--were the Spurs good or lucky? Okay, so I think the Clippers were bad at drafting...

Here's what happened, though:
Code:
WS over Replacement
Team Picks Expected Actual Ratio
NOH 6 30 74 246%
PHO 69 245 375 153%
SAS 53 210 320 153%
SDC 8 56 83 149%
DET 53 263 354 135%
LAL 47 190 253 134%
CLE 61 282 373 132%
UTA 55 233 300 128%
MIL 54 260 318 122%
MEM 12 58 69 117%
KCK 15 72 83 116%
SEA 65 274 309 113%
HOU 58 264 280 106%
MIA 38 162 166 103%
NYK 55 251 258 102%
IND 60 271 277 102%
GSW 73 361 369 102%
POR 68 273 275 101%
CHH 24 154 149 97%
CHI 80 419 389 93%
MIN 38 192 177 92%
SAC 48 232 206 89%
BOS 52 257 226 88%
DAL 71 343 298 87%
WSB 52 209 182 87%
NJN 55 321 279 87%
ORL 39 238 206 86%
PHI 65 303 255 84%
DEN 62 288 228 79%
VAN 13 106 81 77%
ATL 71 302 225 74%
TOR 23 161 114 71%
WAS 18 94 54 58%
LAC 52 364 176 48%
CHA 6 55 22 40%
NOK 3 16 1 9%


DJE09 wrote:
2a. A corrollary to this, is when a team has a much higher pick than usual / trades 'up' to get an earlier pick to get a particular player. Does that tend to produce an above average player? I can't think of many successful instances of this off the top of my head. But I can think of times when teams got themselves two high draft picks which didn't really seem to set them up (Chicago Chandler and Curry spring to mind)...


I haven't looked into that. I rather doubt it's significant.

DJE09 wrote:
3. Draft Depth. It is pretty clear that in general the Second Pick is much harder to take than the First. By this I mean, if we "know" the best player, they're gone by the 2nd pick, and if we don't know who the best player is in a draft, often we don't really know who the second best player is either (Durant / Oden being the except that proves this rule). Further, historically on average picks 2-5 generate about the same amount of WS. I think that it could be argued that this is as much about teams, what they (think they) need most to be (more) competative as much as talent. This is infamous in the Bowie over Jordan selection. There is only a limited amount of great talent, so by the time you get to the 5th pick, they have all been taken. (LBJ et al providing evidence for this assertion). The depth varies from draft to draft adding to the noise.


I think that in most cases, there is a fairly steady continuum of talent. But it's only known after the fact. The reversal of #2 and #3 is more likely because of random scatter. And Len Bias and Jay Williams not turning out. I don't know why there was a consistent bias.

Here is a graph of the depth of each draft evaluated:

Image

And here is the average shape of talent, with +/- 1 stdev error bars:

Image
DJE09 wrote:
I really don't know what to think about all this. I feel it is a chicken and egg thing. Are players that talented when they are drafted, or do they develop into the productive player because of the situation they ended up in? Would De Jaun Blair have worked at, say Golden State (I'll go out on a limb and say he'd have gotten injured)? Would Kobe have been a flop if he was drafted by New Jersey instead of Kerry Kittles (kittles played about twice as many minutes as Kobe in 1st 2 seasons, whilst being similarly productive on a per minute basis...)? It has to be a bit of both. But that is just opinion. Case and point Brandon Jennings. Lots of teams getting slammed for not picking him up, but perhaps his development is in part of being in Milwalkee. How do even begin to separate out those factors? Would Tyreke Evans have had his rookie season at the Clippers or Memphis? Agent 0 is imfamously immature in his crediting his motivation for his success (sic) to his low draft position.

Oh, yeah, If you start post 1985 draft, positions 3 and 5 don't look all that hot ...


From what I have shown, the shape of talent curve varies widely from draft to draft. However, no one really knows it beforehand. (Perhaps they can guess--this year, they say that the mid-range talent is very deep.)

I suppose some of the difference in how a player performed depended on situation. However, I think that is often overrated. I believe most of what happens with the player depends on the player. If they are good enough, work hard enough, the team will get them on the floor. And all teams have a strong incentive to develop their players; I don't believe there is THAT big of a difference between teams.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Mike G



Joined: 14 Jan 2005
Posts: 3617
Location: Hendersonville, NC

PostPosted: Tue May 18, 2010 10:02 am Post subject: Reply with quote
Again, nice graphs!
The one labeled "Quality of Drafted Players" (and introduced as " a graph of the depth of each draft evaluated: ") -- can you tell us what that represents?

Is it still the first four years of players' WS?
And the (light yellow) high peak from the '84 draft is Jordan's 1st 4 years, while that of '87 is Robinson?

If so, then I'll mention that the first 4 years of the '85 draft vastly understates that group's eventual totals. They were 2nd to the previous year in career WS, I believe. But they were (I guess) nothing special thru 4 years.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
DSMok1



Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains

PostPosted: Tue May 18, 2010 10:46 am Post subject: Reply with quote
Mike G wrote:
Again, nice graphs!
The one labeled "Quality of Drafted Players" (and introduced as " a graph of the depth of each draft evaluated: ") -- can you tell us what that represents?

Is it still the first four years of players' WS?
And the (light yellow) high peak from the '84 draft is Jordan's 1st 4 years, while that of '87 is Robinson?

If so, then I'll mention that the first 4 years of the '85 draft vastly understates that group's eventual totals. They were 2nd to the previous year in career WS, I believe. But they were (I guess) nothing special thru 4 years.


Correct on all counts. However, wouldn't you say that the value to the team (in general) is from their first few years (after which the team pays close to market value for the player)?

In other words--the value of the contract to the team comes from not having to pay the player what they are worth on a true open market, like in baseball. If the average win costs $2.5 million dollars (the approximate payroll, divided by 31 (wins above replacement)*30), then the value of any player's contract is about their value in wins*2.5 million - contract price.

Obviously, the free agent market cost is above that price, primarily because teams have more of their (CBA defined) player money because of restricting the younger players' payrolls. Phil Birnbaum discussed the price for wins structure in a series of posts about baseball: http://sabermetricresearch.blogspot.com ... html#links, and Dan Rosenbaum wrote a journal article about this subject as well: http://web.archive.org/web/200710271553 ... ngnba1.pdf.

And this is the direction I want to research. What is the ACTUAL value of the higher draft picks, not in terms of wins, but in terms of the money saved by getting the wins at a controlled price?

Here's my first crack at it. Second rounders get really fuzzy because of their variable contracts.

Code:
Pick Value of Wins 4 Yr Contract Marginal Value
1 $53,570,000 $20,041,804 $33,530,000
2 $43,390,000 $17,936,256 $25,450,000
3 $37,560,000 $16,115,130 $21,440,000
4 $33,480,000 $14,532,895 $18,950,000
5 $30,350,000 $13,166,778 $17,180,000
6 $27,830,000 $11,962,023 $15,870,000
7 $25,720,000 $10,925,078 $14,790,000
8 $23,910,000 $10,013,568 $13,900,000
9 $22,340,000 $9,209,319 $13,130,000
10 $20,950,000 $8,750,603 $12,200,000
11 $19,700,000 $8,419,293 $11,280,000
12 $18,580,000 $8,097,008 $10,480,000
13 $17,560,000 $7,785,917 $9,770,000
14 $16,620,000 $7,487,751 $9,130,000
15 $15,760,000 $7,199,173 $8,560,000
16 $14,970,000 $6,841,065 $8,130,000
17 $14,230,000 $6,501,828 $7,730,000
18 $13,540,000 $6,179,628 $7,360,000
19 $12,900,000 $5,904,060 $7,000,000
20 $12,300,000 $5,670,647 $6,630,000
21 $11,730,000 $5,507,633 $6,220,000
22 $11,200,000 $5,349,948 $5,850,000
23 $10,700,000 $5,196,317 $5,500,000
24 $10,230,000 $5,045,667 $5,180,000
25 $9,780,000 $4,899,204 $4,880,000
26 $9,350,000 $4,739,066 $4,610,000
27 $8,950,000 $4,603,216 $4,350,000
28 $8,570,000 $4,575,824 $3,990,000
29 $8,200,000 $4,542,628 $3,660,000
30 $7,860,000 $4,509,812 $3,350,000


No wonder the top pick in the draft rarely gets traded!
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DSMok1



Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains

PostPosted: Tue May 18, 2010 10:53 am Post subject: Reply with quote
The top 10 players during their first 4 years in the league for 1984 and 1985:

Code:
1984
Pick Rnk Team Player WSaRP, first 4 Yrs
3 1 CHI Michael Jordan 48.4
5 2 PHI Charles Barkley 41.1
1 3 HOU Hakeem Olajuwon 35.6
16 4 UTA John Stockton 25.6
4 5 DAL Sam Perkins 21.9
18 6 IND Vern Fleming 19.9
9 7 KCK Otis Thorpe 19.8
7 8 SAS Alvin Robertson 17.5
46 9 POR Jerome Kersey 17.1
11 10 ATL Kevin Willis 15.6

1985
Pick Rnk Team Player WSaRP, first 4 Yrs
13 1 UTA Karl Malone 27.9
45 2 CLE Hot Rod Williams 24.5
23 3 LAL A.C. Green 23.6
24 4 POR Terry Porter 23.0
1 5 NYK Patrick Ewing 20.6
7 6 GSW Chris Mullin 19.1
4 7 SEA Xavier McDaniel 18.5
2 8 IND Wayman Tisdale 15.6
18 9 DET Joe Dumars 15.3
9 10 CLE Charles Oakley 14.0
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Carlos



Joined: 21 Jan 2005
Posts: 64
Location: Montevideo, Uruguay

PostPosted: Tue May 18, 2010 11:01 am Post subject: Reply with quote
VERY nice, but it would be interesting to see this data in terms of probability of drafting a player with x number of Winshares (from 5 to 10, from 10 to 15, etc.) at each spot. I remember seeing a study in 82games and it was somewhat striking how high the probability of drafting a bench player was after the 6th pick. My impression has always been that a lot of the value of the first 5 picks is that is very unlikely to do really badly (Darko badly) that high.
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: Tue May 18, 2010 11:32 am Post subject: Reply with quote
Carlos wrote:
VERY nice, but it would be interesting to see this data in terms of probability of drafting a player with x number of Winshares (from 5 to 10, from 10 to 15, etc.) at each spot. I remember seeing a study in 82games and it was somewhat striking how high the probability of drafting a bench player was after the 6th pick. My impression has always been that a lot of the value of the first 5 picks is that is very unlikely to do really badly (Darko badly) that high.


Hence my usage of the mean and median, along with percentiles up in the graphs higher up.

Or this:

Image
EDIT: the numbers are WS above Replacement thresholds.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Carlos



Joined: 21 Jan 2005
Posts: 64
Location: Montevideo, Uruguay

PostPosted: Tue May 18, 2010 12:04 pm Post subject: Reply with quote
Great. I had noticed the mean and median thing, but I think this shows more clearly the variation in value. After the fifth pick, you have a quite substantial chance of picking a bit player or a bust. Picking in the top 3 isn't a sure thing either, but your chances of picking a solid player or better are really good.
Back to top
View user's profile Send private message
Mike G



Joined: 14 Jan 2005
Posts: 3617
Location: Hendersonville, NC

PostPosted: Tue May 18, 2010 1:37 pm Post subject: Reply with quote
Looking at these charts, I still get the feeling there's an artificial horizon, beneath which we aren't seeing it all. That would be, of course, the .025 subtracted from everyone's WS/48 rates.

Losing more than half of their WS to the replacement threshold, during their 1st 4 years: Stackhouse, Jim Jackson, Griffith, Bradley, JWill, Hughes, Mark West, Delk, Bobby Jackson, Longley, Sealy, Chris Dudley, Christie, Skiles, Hawes, House, ...

I haven't looked up their draft positions, but I make a guess that they all played for bad-to-terrible teams early in their careers. Yet they all turned out respectably, after getting to better teams.

Some confounding conditions:
- Better rookies tend to start with bad teams initially.
- Players from bad teams are slammed by WS.
- The replacement threshold reduces their already meager WS even further.

So I repeat my initial question: How do these charts and tables look when RT = 0 ? or RT = .01 ?
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
Display posts from previous:

Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

Re: More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 12:58 am

Mike G



Joined: 14 Jan 2005
Posts: 3535
Location: Hendersonville, NC

PostPosted: Tue May 18, 2010 1:49 pm Post subject: Reply with quote
DSMok1 wrote:
.. the value to the team (in general) is from their first few years (after which the team pays close to market value for the player)..

Shoot, I thought after 4 years, a player has reached his peak/plateau; at which point you can invest in him heavily, or you can trade him for other really good players.

Isn't it really a whole lot better to actually have a 25-yr old Duncan or Malone or Ewing than to not have him? Most guys are pretty happy to have a team built around them -- or be integral to a stable team situation.

Code:
1985
Pick Rnk Team Player WSaRP, first 4 Yrs
13 1 UTA Karl Malone 27.9
45 2 CLE Hot Rod Williams 24.5
23 3 LAL A.C. Green 23.6
24 4 POR Terry Porter 23.0
1 5 NYK Patrick Ewing 20.6
7 6 GSW Chris Mullin 19.1
4 7 SEA Xavier McDaniel 18.5
2 8 IND Wayman Tisdale 15.6
18 9 DET Joe Dumars 15.3
9 10 CLE Charles Oakley 14.0

After 4 years, Hot Rod ranked over Ewing, Mullin, Porter, et al. Amazing.
Some of these guys developed slowly, others worked in losing environments. Ewing, probably both.

At this stage, I don't think their value has been maxed. I think for most, it's just being realized.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
DSMok1



Joined: 05 Aug 2009
Posts: 587
Location: Where the wind comes sweeping down the plains

PostPosted: Tue May 18, 2010 1:59 pm Post subject: Reply with quote
Mike G wrote:
Looking at these charts, I still get the feeling there's an artificial horizon, beneath which we aren't seeing it all. That would be, of course, the .025 subtracted from everyone's WS/48 rates.

Losing more than half of their WS to the replacement threshold, during their 1st 4 years: Stackhouse, Jim Jackson, Griffith, Bradley, JWill, Hughes, Mark West, Delk, Bobby Jackson, Longley, Sealy, Chris Dudley, Christie, Skiles, Hawes, House, ...

I haven't looked up their draft positions, but I make a guess that they all played for bad-to-terrible teams early in their careers. Yet they all turned out respectably, after getting to better teams.

Some confounding conditions:
- Better rookies tend to start with bad teams initially.
- Players from bad teams are slammed by WS.
- The replacement threshold reduces their already meager WS even further.

So I repeat my initial question: How do these charts and tables look when RT = 0 ? or RT = .01 ?


The charts would look the same except the players with a bunch of minutes would get boosted upwards.

The value of a player is absolutely tied to their difference from replacement level. 0 WS is not replacement level; in fact, it is an artificial horizon itself.

However, in totality, the difference would not really change anything. It would change the scale is all. It would probably help the higher picks that were given a longer leash in proving their ineptitude. (Michael Olowokandi, anyone?)
Back to top
View user's profile Send private message Visit poster's website
DSMok1



Joined: 05 Aug 2009
Posts: 587
Location: Where the wind comes sweeping down the plains

PostPosted: Tue May 18, 2010 2:04 pm Post subject: Reply with quote
Mike G wrote:
DSMok1 wrote:
.. the value to the team (in general) is from their first few years (after which the team pays close to market value for the player)..

Shoot, I thought after 4 years, a player has reached his peak/plateau; at which point you can invest in him heavily, or you can trade him for other really good players.

Isn't it really a whole lot better to actually have a 25-yr old Duncan or Malone or Ewing than to not have him? Most guys are pretty happy to have a team built around them -- or be integral to a stable team situation.

Code:
1985
Pick Rnk Team Player WSaRP, first 4 Yrs
13 1 UTA Karl Malone 27.9
45 2 CLE Hot Rod Williams 24.5
23 3 LAL A.C. Green 23.6
24 4 POR Terry Porter 23.0
1 5 NYK Patrick Ewing 20.6
7 6 GSW Chris Mullin 19.1
4 7 SEA Xavier McDaniel 18.5
2 8 IND Wayman Tisdale 15.6
18 9 DET Joe Dumars 15.3
9 10 CLE Charles Oakley 14.0

After 4 years, Hot Rod ranked over Ewing, Mullin, Porter, et al. Amazing.
Some of these guys developed slowly, others worked in losing environments. Ewing, probably both.

At this stage, I don't think their value has been maxed. I think for most, it's just being realized.


I picked 4 years because that is the amount of time on the rookie contract. After that point, the player is a restricted free agent, and the other teams will force the team to pay market value for their players. The rare exception is the "max" players, which occur at a rate of about 1 or 2 a year. Those players the teams will still get to keep at a relatively reduced price. After year 5, any team could have them. See Lebron.

Since I'm looking at the value of the draft pick itself to the team holding it, I'm looking at the time the player is on the rookie contract.

An interesting issue is late in the first round, where the teams don't always offer 3 year contracts with a 4th year option. In addition, losses to the team can be mitigated by not exercising the option. And in the second round, no telling what the contract looks like--making it hard to quantify the value of a second-rounder. I would argue, however, that it is quite possible that pick 31 is worth more than pick 30 because of the additional flexibility.
Back to top
View user's profile Send private message Visit poster's website
BobboFitos



Joined: 21 Feb 2009
Posts: 187
Location: Cambridge, MA

PostPosted: Tue May 18, 2010 3:10 pm Post subject: Reply with quote
DSMok1 wrote:


Code:
Pick Value of Wins 4 Yr Contract Marginal Value
1 $53,570,000 $20,041,804 $33,530,000
2 $43,390,000 $17,936,256 $25,450,000
3 $37,560,000 $16,115,130 $21,440,000
4 $33,480,000 $14,532,895 $18,950,000
5 $30,350,000 $13,166,778 $17,180,000
6 $27,830,000 $11,962,023 $15,870,000
7 $25,720,000 $10,925,078 $14,790,000
8 $23,910,000 $10,013,568 $13,900,000
9 $22,340,000 $9,209,319 $13,130,000
10 $20,950,000 $8,750,603 $12,200,000
11 $19,700,000 $8,419,293 $11,280,000
12 $18,580,000 $8,097,008 $10,480,000
13 $17,560,000 $7,785,917 $9,770,000
14 $16,620,000 $7,487,751 $9,130,000
15 $15,760,000 $7,199,173 $8,560,000
16 $14,970,000 $6,841,065 $8,130,000
17 $14,230,000 $6,501,828 $7,730,000
18 $13,540,000 $6,179,628 $7,360,000
19 $12,900,000 $5,904,060 $7,000,000
20 $12,300,000 $5,670,647 $6,630,000
21 $11,730,000 $5,507,633 $6,220,000
22 $11,200,000 $5,349,948 $5,850,000
23 $10,700,000 $5,196,317 $5,500,000
24 $10,230,000 $5,045,667 $5,180,000
25 $9,780,000 $4,899,204 $4,880,000
26 $9,350,000 $4,739,066 $4,610,000
27 $8,950,000 $4,603,216 $4,350,000
28 $8,570,000 $4,575,824 $3,990,000
29 $8,200,000 $4,542,628 $3,660,000
30 $7,860,000 $4,509,812 $3,350,000


No wonder the top pick in the draft rarely gets traded!


Wow, great data. We can draw a few inferences from this, but the primary one is teams that historically sell their late picks (for 3m or less!) are likely making a mistake. (Looking at you, Phoenix!)
_________________
-Rob
Back to top
View user's profile Send private message AIM Address
BobboFitos



Joined: 21 Feb 2009
Posts: 187
Location: Cambridge, MA

PostPosted: Tue May 18, 2010 3:12 pm Post subject: Reply with quote
DSMok1 wrote:
Carlos wrote:
VERY nice, but it would be interesting to see this data in terms of probability of drafting a player with x number of Winshares (from 5 to 10, from 10 to 15, etc.) at each spot. I remember seeing a study in 82games and it was somewhat striking how high the probability of drafting a bench player was after the 6th pick. My impression has always been that a lot of the value of the first 5 picks is that is very unlikely to do really badly (Darko badly) that high.


Hence my usage of the mean and median, along with percentiles up in the graphs higher up.

Or this:

Image

EDIT: the numbers are WS above Replacement thresholds.


Can you smooth this out?
_________________
-Rob
Back to top
View user's profile Send private message AIM Address
DSMok1



Joined: 05 Aug 2009
Posts: 587
Location: Where the wind comes sweeping down the plains

PostPosted: Tue May 18, 2010 3:17 pm Post subject: Reply with quote
BobboFitos wrote:


Can you smooth this out?


I wish I could create a smoothed 3D surface of win shares vs. pick vs. odds. I think that's a bit beyond me though.

The thresholds were arbitrary, simply to give a feel for the shape of the surface.
Back to top
View user's profile Send private message Visit poster's website
Mike G



Joined: 14 Jan 2005
Posts: 3535
Location: Hendersonville, NC

PostPosted: Tue May 18, 2010 3:32 pm Post subject: Reply with quote
DSMok1 wrote:

The value of a player is absolutely tied to their difference from replacement level. 0 WS is not replacement level; in fact, it is an artificial horizon itself.

However, in totality, the difference would not really change anything. It would change the scale is all. It would probably help the higher picks that were given a longer leash in proving their ineptitude. (Michael Olowokandi, anyone?)

Good example. Kandi, with a career year in year 4, climbed up to zero in career Win Shares.

Rather than confuse him with players who did something --
http://www.basketball-reference.com/pla ... i?id=NG7tG
... Why not give others their due?
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
DSMok1



Joined: 05 Aug 2009
Posts: 587
Location: Where the wind comes sweeping down the plains

PostPosted: Tue May 18, 2010 3:56 pm Post subject: Reply with quote
Mike G wrote:

Good example. Kandi, with a career year in year 4, climbed up to zero in career Win Shares.

Rather than confuse him with players who did something --
http://www.basketball-reference.com/pla ... i?id=NG7tG
... Why not give others their due?


They were at or below replacement level, so they contributed nothing to the team, other than taking up space. (Assuming the WS evaluation of them was correct). Seriously... replacement level is a well-known subject in sports player evaluation metrics; I don't really think it's worth getting into here since it's tangential to the subject at hand.
Back to top
View user's profile Send private message Visit poster's website
Mike G



Joined: 14 Jan 2005
Posts: 3535
Location: Hendersonville, NC

PostPosted: Wed May 19, 2010 6:54 am Post subject: Reply with quote
Well, I've been discussing replacement level with you, so it can hardly be inferred that I don't believe in it.
Quote:
They were at or below replacement level, so they contributed nothing to the team, other than taking up space.

Kevin Edwards played for Miami in their first 4 seasons. In that time, he was 3rd in minutes and points, 5th in Reb, 2nd in Ast, 1st in Stl. The team won 95 games, and he's credited with 4.1 WS.

You've removed the entirety of his WS by an assumption that .025 WS/48 = 0 . Nobody who was with that team would accept that he "contributed nothing".

There are a whole bunch of pretty good players for bad teams who are quantitatively better than the many true 'nothings' out there. You may be throwing away lots of good data when you define them as 'non contributors'.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
DSMok1



Joined: 05 Aug 2009
Posts: 587
Location: Where the wind comes sweeping down the plains

PostPosted: Wed May 19, 2010 8:00 am Post subject: Reply with quote
Mike G wrote:
Well, I've been discussing replacement level with you, so it can hardly be inferred that I don't believe in it.
Quote:
They were at or below replacement level, so they contributed nothing to the team, other than taking up space.

Kevin Edwards played for Miami in their first 4 seasons. In that time, he was 3rd in minutes and points, 5th in Reb, 2nd in Ast, 1st in Stl. The team won 95 games, and he's credited with 4.1 WS.

You've removed the entirety of his WS by an assumption that .025 WS/48 = 0 . Nobody who was with that team would accept that he "contributed nothing".

There are a whole bunch of pretty good players for bad teams who are quantitatively better than the many true 'nothings' out there. You may be throwing away lots of good data when you define them as 'non contributors'.


No, I think it's working fine. Kevin Edwards was an abysmal offensive player. Over the first 4 years: TS% .470, TRB% 6.4, AST% 18.8, TOV% 14.9. The only thing he was good at was steals: STL% 2.8.

I'm quite fine with those rates summing to a replacement-level player! Why did his coaches play him so much?
Back to top
View user's profile Send private message Visit poster's website
Mike G



Joined: 14 Jan 2005
Posts: 3535
Location: Hendersonville, NC

PostPosted: Wed May 19, 2010 8:41 am Post subject: Reply with quote
Because he was the best they could get, obviously.
In other words, no available replacement was as good.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
DLew



Joined: 13 Nov 2006
Posts: 222


PostPosted: Wed May 19, 2010 9:10 am Post subject: Reply with quote
Is that really obvious Mike? Are you sure it couldn't have just been a case of poor evaluation by the coaches? There is quite a bit of evidence that people, in all fields and aspects of life make mistakes and bad choices all the time. Not sure why we should assume that coaches always make the right ones. I'm guessing that there were several better replacements available for the NBA minimum during that time.
Back to top
View user's profile Send private message
DSMok1



Joined: 05 Aug 2009
Posts: 587
Location: Where the wind comes sweeping down the plains

PostPosted: Wed May 19, 2010 9:14 am Post subject: Reply with quote
Mike G wrote:
Because he was the best they could get, obviously.
In other words, no available replacement was as good.


I doubt it. I suspect that they believed he would develop into a better player, so they kept playing him despite him actually hurting the team. That's seen often with very young players--they are played despite not being the best player available, to develop the player for the future.

Also, very often bad teams end up playing replacement-level players (or even worse in terms of actual performance). Check out the Astros this year: Fangraphs player wins above replacement. Almost every player on the team is below replacement level!
Back to top
View user's profile Send private message Visit poster's website
Joe



Joined: 27 Sep 2009
Posts: 94
Location: Long Island, NY

PostPosted: Thu May 20, 2010 6:38 am Post subject: Reply with quote
DSMok1 wrote:


Code:
Pick Value of Wins 4 Yr Contract Marginal Value
1 $53,570,000 $20,041,804 $33,530,000
2 $43,390,000 $17,936,256 $25,450,000
3 $37,560,000 $16,115,130 $21,440,000
4 $33,480,000 $14,532,895 $18,950,000
5 $30,350,000 $13,166,778 $17,180,000
6 $27,830,000 $11,962,023 $15,870,000
7 $25,720,000 $10,925,078 $14,790,000
8 $23,910,000 $10,013,568 $13,900,000
9 $22,340,000 $9,209,319 $13,130,000
10 $20,950,000 $8,750,603 $12,200,000
11 $19,700,000 $8,419,293 $11,280,000
12 $18,580,000 $8,097,008 $10,480,000
13 $17,560,000 $7,785,917 $9,770,000
14 $16,620,000 $7,487,751 $9,130,000
15 $15,760,000 $7,199,173 $8,560,000
16 $14,970,000 $6,841,065 $8,130,000
17 $14,230,000 $6,501,828 $7,730,000
18 $13,540,000 $6,179,628 $7,360,000
19 $12,900,000 $5,904,060 $7,000,000
20 $12,300,000 $5,670,647 $6,630,000
21 $11,730,000 $5,507,633 $6,220,000
22 $11,200,000 $5,349,948 $5,850,000
23 $10,700,000 $5,196,317 $5,500,000
24 $10,230,000 $5,045,667 $5,180,000
25 $9,780,000 $4,899,204 $4,880,000
26 $9,350,000 $4,739,066 $4,610,000
27 $8,950,000 $4,603,216 $4,350,000
28 $8,570,000 $4,575,824 $3,990,000
29 $8,200,000 $4,542,628 $3,660,000
30 $7,860,000 $4,509,812 $3,350,000




Can you run this for OWS and DWS separately as well?
_________________
http://www.hoopdata.com
Back to top
View user's profile Send private message Visit poster's website
DSMok1



Joined: 05 Aug 2009
Posts: 587
Location: Where the wind comes sweeping down the plains

PostPosted: Thu May 20, 2010 8:01 am Post subject: Reply with quote
Joe wrote:

Can you run this for OWS and DWS separately as well?


Yes, I can. Something to note--more of the "rookie" wins come from defense than from offense; this is possibly related to the fact there is a significant team adjustment on defensive win shares.
Back to top
View user's profile Send private message Visit poster's website

Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

Re: More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 1:03 am

Joe



Joined: 27 Sep 2009
Posts: 94
Location: Long Island, NY

PostPosted: Thu May 20, 2010 8:30 am Post subject: Reply with quote
DSMok1 wrote:
Joe wrote:

Can you run this for OWS and DWS separately as well?


Yes, I can. Something to note--more of the "rookie" wins come from defense than from offense; this is possibly related to the fact there is a significant team adjustment on defensive win shares.


Yeah, I'm aware of the adjustment, which is why I wanted to see how it affects things when split.
_________________
http://www.hoopdata.com
Back to top
View user's profile Send private message Visit poster's website
DSMok1



Joined: 05 Aug 2009
Posts: 592
Location: Where the wind comes sweeping down the plains

PostPosted: Thu May 20, 2010 8:37 am Post subject: Reply with quote
Download the RAW DATA here:

Google Docs: Draft Pick Value Data
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DSMok1



Joined: 05 Aug 2009
Posts: 592
Location: Where the wind comes sweeping down the plains

PostPosted: Thu May 20, 2010 9:18 am Post subject: Reply with quote
Here's a start, Joe:

Image

Code:
OWS Value DWS Value Actual Cost Marginal Value
$28,673,949 $26,692,346 $20,041,804 $35,320,000
$23,662,943 $18,213,328 $17,936,256 $23,940,000
$20,427,147 $15,946,985 $16,115,130 $20,260,000
$18,055,336 $14,554,302 $14,532,895 $18,080,000
$16,196,425 $13,533,293 $13,166,778 $16,560,000
$14,677,595 $12,719,340 $11,962,023 $15,430,000
$13,400,909 $12,037,599 $10,925,078 $14,510,000
$12,305,457 $11,447,661 $10,013,568 $13,740,000
$11,350,766 $10,925,209 $9,209,319 $13,070,000
$10,508,569 $10,454,468 $8,750,603 $12,210,000
$9,758,341 $10,024,619 $8,419,293 $11,360,000
$9,084,709 $9,627,921 $8,097,008 $10,620,000
$8,475,858 $9,258,639 $7,785,917 $9,950,000
$7,922,513 $8,912,412 $7,487,751 $9,350,000
$7,417,257 $8,585,842 $7,199,173 $8,800,000
$6,954,069 $8,276,229 $6,841,065 $8,390,000
$6,527,990 $7,981,397 $6,501,828 $8,010,000
$6,134,893 $7,699,562 $6,179,628 $7,650,000
$5,771,299 $7,429,243 $5,904,060 $7,300,000
$5,434,254 $7,169,203 $5,670,647 $6,930,000
$5,121,228 $6,918,389 $5,507,633 $6,530,000
$4,830,034 $6,675,907 $5,349,948 $6,160,000
$4,558,772 $6,440,982 $5,196,317 $5,800,000
$4,305,779 $6,212,947 $5,045,667 $5,470,000
$4,069,592 $5,991,215 $4,899,204 $5,160,000
$3,848,917 $5,775,276 $4,739,066 $4,890,000
$3,642,605 $5,564,675 $4,603,216 $4,600,000
$3,449,628 $5,359,011 $4,575,824 $4,230,000
$3,269,067 $5,157,925 $4,542,628 $3,880,000
$3,100,093 $4,961,099 $4,509,812 $3,550,000


I didn't round the first numbers, obviously!
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DSMok1



Joined: 05 Aug 2009
Posts: 592
Location: Where the wind comes sweeping down the plains

PostPosted: Thu May 20, 2010 9:46 am Post subject: Reply with quote
Here is another graph of the historical depth of drafts--hard to read, I know, but rather cool anyway. Very Happy

Image

Back to top
View user's profile Send private message Send e-mail Visit poster's website
Joe



Joined: 27 Sep 2009
Posts: 94
Location: Long Island, NY

PostPosted: Thu May 20, 2010 10:03 am Post subject: Reply with quote
Seeing how DWS has a much smaller distribution than OWS (I'm fairly certain this applies both for rookie contracts and the league as a whole), shouldn't replacement level value for DWS be higher than OWS? That chart makes me wonder if perhaps DWS and OWS should have their replacement level values calculated independently and then have marginal value added back together based on the independent replacement level values to calculate total marginal value for WS.

If you applied that 0.0125 DWS/48 threshold as replacement level for the entire league, how many players actually come out below replacement level? And as compared to OWS?

I may be off base, but seeing those splits really makes me pause about setting the replacement level for DWS/OWS at the same number (an equal split of the replacement level for WS), though there certainly may be something I'm missing, making me sound like an idiot right now. Smile

edit: I'm also assuming you indeed did split the 0.025 replacement level equally for OWS/DWS, which I think is implied seeing how you didn't state otherwise. If you did do something different, then nevermind my rambling.
_________________
http://www.hoopdata.com
Back to top
View user's profile Send private message Visit poster's website
DSMok1



Joined: 05 Aug 2009
Posts: 592
Location: Where the wind comes sweeping down the plains

PostPosted: Thu May 20, 2010 11:14 am Post subject: Reply with quote
Another chart looking at the expected contribution percentiles for each pick (down to #54).

Image
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DSMok1



Joined: 05 Aug 2009
Posts: 592
Location: Where the wind comes sweeping down the plains

PostPosted: Thu May 20, 2010 12:11 pm Post subject: Reply with quote
Joe wrote:
Seeing how DWS has a much smaller distribution than OWS (I'm fairly certain this applies both for rookie contracts and the league as a whole), shouldn't replacement level value for DWS be higher than OWS? That chart makes me wonder if perhaps DWS and OWS should have their replacement level values calculated independently and then have marginal value added back together based on the independent replacement level values to calculate total marginal value for WS.

If you applied that 0.0125 DWS/48 threshold as replacement level for the entire league, how many players actually come out below replacement level? And as compared to OWS?

I may be off base, but seeing those splits really makes me pause about setting the replacement level for DWS/OWS at the same number (an equal split of the replacement level for WS), though there certainly may be something I'm missing, making me sound like an idiot right now. Smile

edit: I'm also assuming you indeed did split the 0.025 replacement level equally for OWS/DWS, which I think is implied seeing how you didn't state otherwise. If you did do something different, then nevermind my rambling.


I've been pondering this, and I don't think that the replacement levels need to be split and calculated separately. The replacement team is defined as 10 wins. The OWS will have more players below replacement level, purely because we have more precise offensive stats and can calculate the player's actual contributions more accurately. Conversely, the DWS calculation is more ambiguous and has more of a "regressed" look to it--there will be fewer players as far from the league average. I don't think that means the replacement level needs to be changed; I think it simply means that we cannot differentiate players as well on the defensive side of the ball.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Joe



Joined: 27 Sep 2009
Posts: 94
Location: Long Island, NY

PostPosted: Thu May 20, 2010 12:23 pm Post subject: Reply with quote
DSMok1 wrote:
Joe wrote:
Seeing how DWS has a much smaller distribution than OWS (I'm fairly certain this applies both for rookie contracts and the league as a whole), shouldn't replacement level value for DWS be higher than OWS? That chart makes me wonder if perhaps DWS and OWS should have their replacement level values calculated independently and then have marginal value added back together based on the independent replacement level values to calculate total marginal value for WS.

If you applied that 0.0125 DWS/48 threshold as replacement level for the entire league, how many players actually come out below replacement level? And as compared to OWS?

I may be off base, but seeing those splits really makes me pause about setting the replacement level for DWS/OWS at the same number (an equal split of the replacement level for WS), though there certainly may be something I'm missing, making me sound like an idiot right now. Smile

edit: I'm also assuming you indeed did split the 0.025 replacement level equally for OWS/DWS, which I think is implied seeing how you didn't state otherwise. If you did do something different, then nevermind my rambling.


I've been pondering this, and I don't think that the replacement levels need to be split and calculated separately. The replacement team is defined as 10 wins. The OWS will have more players below replacement level, purely because we have more precise offensive stats and can calculate the player's actual contributions more accurately. Conversely, the DWS calculation is more ambiguous and has more of a "regressed" look to it--there will be fewer players as far from the league average. I don't think that means the replacement level needs to be changed; I think it simply means that we cannot differentiate players as well on the defensive side of the ball.


That certainly may be part of it, but it doesn't exactly clash with common sense for the difference in distribution to actually be partially true. The notion that differentiation in defensive performance is less than offensive performance isn't exactly a radical notion by basketball evaluation standards. Further, if the stats are not exact enough, isn't there a good argument for replacement level not being exact enough as well? Replacement level is typically viewed as something like the 80th percentile of player performance in most sports, right? So if 99% of players are playing over the defensive replacement level in this methodology, that doesn't really make much sense, does it? Otherwise what we're talking about isn't really replacement level, seeing how everyone is adding wins.

I'm making an assumption here about where the replacement threshold is actually falling with DWS, as I don't have access to the entire league numbers.
_________________
http://www.hoopdata.com
Back to top
View user's profile Send private message Visit poster's website
BobboFitos



Joined: 21 Feb 2009
Posts: 188
Location: Cambridge, MA

PostPosted: Thu May 20, 2010 1:12 pm Post subject: Reply with quote
This thread has brought up a few other thread starters, like redefining exactly what we all mean by replacement level. Perhaps it's best served to split it up, so that we can delve a little further into the value of a draft pick here - all the while exploring the floor of a team of replacements in a new thread?..
_________________
-Rob
Back to top
View user's profile Send private message AIM Address
DSMok1



Joined: 05 Aug 2009
Posts: 592
Location: Where the wind comes sweeping down the plains

PostPosted: Thu May 20, 2010 4:55 pm Post subject: Reply with quote
My first attempt at smoothing the WSoRP vs. Draft Pick vs. Likelihood surface:

Image


The equation to get to that is REALLY messy.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DSMok1



Joined: 05 Aug 2009
Posts: 592
Location: Where the wind comes sweeping down the plains

PostPosted: Fri May 21, 2010 9:07 am Post subject: Reply with quote
A 2-D version of the above graph:

Image

I went back historically and estimated thresholds for various types of players.

Over 4 years, such as this evaluation, here are the levels:

30+ WSoRP: Superstar (~15 in the league in any 4 year span)
20-30 WSoRP: Star (~30 in the league over a 4 year span)
12-20 WSoRP: Solid Starter (~60)
5-12 WSoRP: Role Player
0-5 WSoRP: Bit Player
<0 WSoRP: Worthless Player
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DSMok1



Joined: 05 Aug 2009
Posts: 592
Location: Where the wind comes sweeping down the plains

PostPosted: Fri May 21, 2010 10:21 am Post subject: Reply with quote
Continuing to explore visualization options:

Image

Image
Back to top
View user's profile Send private message Send e-mail Visit poster's website
gabefarkas



Joined: 31 Dec 2004
Posts: 1313
Location: Durham, NC

PostPosted: Fri May 21, 2010 12:42 pm Post subject: Reply with quote
DJE09 wrote:
before 1995 there weren't 28th picks in the first round, so instead of being one of the best teams in the NBA, it was one of the worst teams, making their second pick. Also if you look back into the 80s when they had draft until you drop drafts, you see the odd player like Manute Bol being drafted 97th in 1983, amongst a whole heap of people who never even got burn in the NBA. The issue of expansion over the last 30 years makes any sort of analysis based on draft position really messy.

(I know I'm a little late with this reply...)

I think the talent pool expanded in line with the number of picks. At least, I think we have to assume that it did. In other words, those late-first-round foreign gems that teams started grabbing in the late 90's weren't even on the radar screen in the mid-80's. Or if they were, there were fewer of them and there were also fewer American-born players.

To me, the 1996 draft was the first one where you really see an influx of foreign talent, and even a few creeping into the first round (#14 Peja Stojakovic, #20 Zydrunas Ilgauskas, #23 Efthimi Rentzias, #25 Martin Muursepp). The 1998 draft was when the floodgates burst open (Dirk, Rasho Nesterovic, Mirsad Turkcan, Vladimir Stepania, Bruno Sundov).

Before '96, there you had players like: George Zidek (born in Czech Republic, but attended UCLA), Dragan Tarlac, Dejan Bodiroga, Constantin Popa (born in Romania, attended Miami), Aurelijius Zukauskas, Yinka Dare, Andrei Fetisov, William Njoku, Zeljko Rebraca, Gheorghe Muresan, Marcelo Nicola, Sasha Danilovic, Zan Tabak, Toni Kukoc, Stefano Rusconi, Vlade Divac, Dino Radja, Jose Vargas, Jorge Gonzalez, Rik Smits, Song Tao, Jose Antonio Montero, Sarunas Marciulionis, Franjo Arapovic, Georgi Glouchkov, Arvydas Sabonis, Drazen Petrovic, Stojko Vrankovic, Uwe Blab, Detlef Schrempf, Manute Bol, Milos Babic, Radisav Curcic, Geert Hammink, Swen Nater, Gundars Vetra, and a bunch more I'm probably missing.

There's only a handful of guys who actually had an impact (or even saw more than limited minutes) out of that big group (of which I'm leaving out a whole bunch). Muresan, Rebraca, Kukoc, Divac, Radja, Smits, Marciulionis, Sabonis, Petrovic, Schrempf, Nater. That's 10-15 players out of maybe 100-200 who were drafted. The highest-drafted players were Smits at #2 and Yinka at #14. Both of them went to American colleges though.

My point is that I think it makes complete sense to draw an analogy between what a late 1st round pick was in 1980 and a late first round pick in 2005. Even though the number of teams increased, the talent pool widened too. So a 28th pick in 2005 will net you a lot better player than a 28th pick in 1980.
Back to top
View user's profile Send private message Send e-mail AIM Address
DSMok1



Joined: 05 Aug 2009
Posts: 592
Location: Where the wind comes sweeping down the plains

PostPosted: Fri May 21, 2010 1:13 pm Post subject: Reply with quote
gabefarkas wrote:

(I know I'm a little late with this reply...)

I think the talent pool expanded in line with the number of picks. At least, I think we have to assume that it did. In other words, those late-first-round foreign gems that teams started grabbing in the late 90's weren't even on the radar screen in the mid-80's. Or if they were, there were fewer of them and there were also fewer American-born players.

To me, the 1996 draft was the first one where you really see an influx of foreign talent, and even a few creeping into the first round (#14 Peja Stojakovic, #20 Zydrunas Ilgauskas, #23 Efthimi Rentzias, #25 Martin Muursepp). The 1998 draft was when the floodgates burst open (Dirk, Rasho Nesterovic, Mirsad Turkcan, Vladimir Stepania, Bruno Sundov).

Before '96, there you had players like: George Zidek (born in Czech Republic, but attended UCLA), Dragan Tarlac, Dejan Bodiroga, Constantin Popa (born in Romania, attended Miami), Aurelijius Zukauskas, Yinka Dare, Andrei Fetisov, William Njoku, Zeljko Rebraca, Gheorghe Muresan, Marcelo Nicola, Sasha Danilovic, Zan Tabak, Toni Kukoc, Stefano Rusconi, Vlade Divac, Dino Radja, Jose Vargas, Jorge Gonzalez, Rik Smits, Song Tao, Jose Antonio Montero, Sarunas Marciulionis, Franjo Arapovic, Georgi Glouchkov, Arvydas Sabonis, Drazen Petrovic, Stojko Vrankovic, Uwe Blab, Detlef Schrempf, Manute Bol, Milos Babic, Radisav Curcic, Geert Hammink, Swen Nater, Gundars Vetra, and a bunch more I'm probably missing.

There's only a handful of guys who actually had an impact (or even saw more than limited minutes) out of that big group (of which I'm leaving out a whole bunch). Muresan, Rebraca, Kukoc, Divac, Radja, Smits, Marciulionis, Sabonis, Petrovic, Schrempf, Nater. That's 10-15 players out of maybe 100-200 who were drafted. The highest-drafted players were Smits at #2 and Yinka at #14. Both of them went to American colleges though.

My point is that I think it makes complete sense to draw an analogy between what a late 1st round pick was in 1980 and a late first round pick in 2005. Even though the number of teams increased, the talent pool widened too. So a 28th pick in 2005 will net you a lot better player than a 28th pick in 1980.


This chart disagrees with you about that last statement:

Image
The curves are about the same 1-60, each year. I think there wasn't a big change at least in this regard.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
DLew



Joined: 13 Nov 2006
Posts: 222


PostPosted: Fri May 21, 2010 3:22 pm Post subject: Reply with quote
Can that be true? It seems to me that with more teams in the league now (and therefore more games), there must be more Win Shares. So, even assuming the quality of players drafted did not change, we should see the number of win shares per draft increase roughly in line with the increase in overall win shares. Right?



page 4 of 4 missing

Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

Re: More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 1:16 am

page 1 of 4 missing

page 2

mavs128



Joined: 25 Feb 2005
Posts: 32
Location: Dallas, Texas

PostPosted: Fri Jun 16, 2006 10:31 am Post subject: Reply with quote
alphamale wrote:
If Toronto doesn't pick Tyrus Thomas, they would be making a grave mistake. Everybody knows that Toronto had one of the leagues worst defense. They need a shot blocker and rebounder like Thomas, he would also bolster their running game because of his athletism. A front line consisting of Bosh, Villeneuva, and Thomas would be quite formidable indeed.

I wouldn't be surprised if Thomas ends up with a very high similarity score to Marcus Camby. Hopefully he's not as injury prone as the Camster.


I dunno.....they're all PFs. Thomas is probably quick enough to guard SFs, but do you want Villanueva guarding centers? Tyrus measured at 6 feet, 7.25 inches w/0 shoes.
Back to top
View user's profile Send private message
Tango



Joined: 18 Mar 2005
Posts: 24


PostPosted: Fri Jun 16, 2006 10:38 am Post subject: Reply with quote
Richard Lu posted this article on 82games.
http://www.82games.com/nbadraft2006.htm

I think I understand his methodology. What I don't understand is how he calculated and correlated the success rates for the 2006 draft prospects with the similar players.

Anyone have any insights?
Back to top
View user's profile Send private message
cherokee_ACB



Joined: 22 Mar 2006
Posts: 157


PostPosted: Fri Jun 16, 2006 1:32 pm Post subject: Reply with quote
holymoly wrote:
I have compiled some stats on the the main european draft prospects coming from the euroleague. It has been posted on 82games.com, check it out


Let me complement your data with some +/- stats. I've also included some already drafted players with NBA potential
Code:

Player Team Min on-off +/-of +/-df DR% OR% DR+/- OR+/-
Bargnani Benetton 358 3.5 -5.9 10.0 71.3% 23.7% 3.2% -6.1%
Splitter TAU 519 -6.7 -6.5 0.5 70.7% 30.7% 1.5% -1.1%
Halperin Olimpija 469 -10.7 -2.7 -10.0 65.4% 31.5% -17.5% 4.4%
Markota Cibona 386 -0.4 -2.3 5.0 70.4% 26.0% 1.6% -3.6%
Vasilopoulos Olympiacos 424 -1.0 4.7 -9.1 70.7% 26.2% -1.7% -3.1%
Jankunas Zalgiris 446 2.3 13.0 -7.3 73.6% 34.3% -0.3% 6.0%
Perovic Partizan 332 8.5 6.2 4.5 70.6% 31.6% -5.5% -7.1%
Scola TAU 713 12.7 8.0 6.6 70.0% 33.3% 0.1% 7.4%
Lavrinovic Zalgiris 491 10.2 -1.4 9.2 72.8% 31.0% -2.8% -2.3%
Gelabale Madrid 451 1.4 5.9 -6.6 65.4% 39.2% -3.9% 4.8%
Javtokas Rytas 576 11.3 1.2 12.1 74.5% 34.8% 0.8% 3.6%


Methodological note: this is based on Euroleague play-by-plays for most games, but I had to discard some corrupt ones. Playing time is not much accurate because the pbp only report the minute at which each event occured. FT points are added to the players in court during the FT, not at foul time.

Now the comments. Bargnani and Splitter deffensive rebounding rates are not that high, but they do help increase their team rate to a good%. In the offensive end, Bargnani plays too far from the basket to rebound.

Bargnani's excellent deffensive +/- number can only be explained by Benetton awful defense. Even if he is not a good defender (from what I've seen, I don't think he is), his teammates are evenworse. With Bargnani in court they still allow 111 points per 100 possesions. Compare that with Splitter's 98 Pp100P. By the way, Splitter +/- stat in Spanish league is much better: +9.2 per 40 min.

Halperin's on-off data is not much significant, as he played 90% of his team time.

Perovic's lack of rebounding instinct clearly manifests in the team rates. On the other hand, he has solid +/- numbers but keep in mind he was playing in the worse team of the competition.
Back to top
View user's profile Send private message
tmansback



Joined: 12 Aug 2005
Posts: 129


PostPosted: Fri Jun 16, 2006 3:44 pm Post subject: Reply with quote
jmethven wrote:
Quote:
Now right now there no proof that PER in college translates to the NBA.


I think you'd be surprised by how well it translates. As mentioned by other posters, there are certain types of players it underrates, such as point guards. However, I've been looking at it for a couple years, and although I'm not sure how well it does for forecasting long-term success, it can give you a pretty good starting point for how a player will perform his first 2 or 3 years in the NBA. I haven't done an extensive study or anything, and there are indeed some stats that translate better than others (scoring is pretty random), but I think a ranking of players by college PER could match up well against a ranking of players by draft position, judged by NBA success. I don't think it's reliable enough to turn college scouting on its head or anything, but it's a pretty good step in that direction.


Well I would be interested in seeing the proof. I know for me I can name dozens of players each year that were productive college players that failed badly in the NBA.
Back to top
View user's profile Send private message
SGreenwell



Joined: 12 Feb 2005
Posts: 76
Location: Rhode Island

PostPosted: Sat Jun 17, 2006 11:49 am Post subject: Reply with quote
I've got a little web site on Atlantic 10 PERs here. The site has 2004 and 2005 up, but I've also completed 2006 and I'm working on 2002.

My method for calculating PER differs a little bit - I only use A-10 statistics to (hopefully) standardize level of play. A-10 teams regularly play 3 to 5 games a year against D1 doormats and NAIA schools. I also use a junk stat, gross per, to try and calculate the actual value a player adds to a team, as opposed to relying solely on a rate stat to rank players.

PG Mardy Collins of Temple rated out as the best player in 2005, and from my numbers in 2006, I believe he's in the top 3 this year as well. However, his PER of 23.05 (2005) and 22ish (2006) really isn't that great. For comparison's sake, Delonte West (26.32) and Jameer Nelson (27.02) both had PERs of 26+ in St. Joe's NCAA run in 2004. Both are now around league to above league average PGs in the NBA.

There's pros and cons to what I'm doing. I write for my college newspaper, so I'm actually more interested in using this as a tool to evaluate college teams, not necessarily NBA draft prospects. This is also part of the reason why I confine myself to A-10 stats.

My other interest is seeing if certain coaches (John Cheney, Phil Martelli, Jim Baron) have a demonstratable "improvement" effect on their players. That information would be interesting and valuable. Also, at some point I'd like to run the numbers on some leagues that have many more draftees (ACC, Big East), if I can find Conference Only stats for them.
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
Mark



Joined: 20 Aug 2005
Posts: 807


PostPosted: Sat Jun 17, 2006 7:28 pm Post subject: Reply with quote
I just followed a link to this article

http://lowpost.net/blog/2006/06/14/sort ... rafts.html

Looks pretty intersting.

FYI if you havent seen it already.
Back to top
View user's profile Send private message
asteroid



Joined: 13 Aug 2005
Posts: 9


PostPosted: Sun Jun 18, 2006 10:02 pm Post subject: Reply with quote
To reply to hpanic's post--your failing to take into effect the importance of physical attributes. For example, Fazekas who is very polished and has a nice soft touch, and puts up great numbers in college, might not make the transition too well to the pros cause of his body. This guy's weaker than Travis Knight and he's not a very good athlete. But this has not been a problem that much in college yet, but can catch up with you in a hurry when you try the NBA. This is why he's going back to school for the second year in a row after flirting with the draft. I love Fazekas's game, but I really worry about his body, I feels there is no way he can hold position and I personally do not think his rebounding numbers will translate.

Markota got athleticism issues. See like I posted in College PER thread, you really have problems with evaluating numbers across the board cause of the importance of size & athleticism. For example, look at Shelden, his numbers look really good, but his standing reach is 8-8, not really good. This is something that has to be taken into consideration. Cause maybe he's another Fizer, Rodney Rogers, Fortson, etc., etc. Guys who dominate in college but are too short or too unathletic to transfer that ability over.

Tyrus Thomas is in that sketchy range too cause he's even shorter than Swift. How do really evaluate his numbers. Cause he might have to play a totally different position in the NBA. I agree with the poster who made the observation that the range of talent is much wider in college from one game to the next.

What do you do with Redick's numbers? How do you evaluate them? Look at a guy like Alford, who's stats are very similar to JJ, and had a similar game & size. Look at Rodney Monroe's numbers. He was a 6-3 2-guard putting up huge numbers in the ACC. Perfect example is Langdon. He could not cut in the NBA cause of his size & lack of speed, but he is one of the better players in Europe cause he can match up at the 2-guard over there. This is where you have to be careful with Euro stats as well. Something to keep in mind with Markota.
Back to top
View user's profile Send private message
hpanic7342



Joined: 16 Feb 2005
Posts: 201


PostPosted: Mon Jun 19, 2006 7:57 am Post subject: Reply with quote
asteroid - I certainly agree that good numbers in college are no guarantee that a player will enjoy similar success in the NBA, and that calculations like PER for college and Euro players should be taken with a grain of salt. However, Fazekas's college numbers (he had the highest PER of everyone I calculated, higher than Noah, Williams, Thomas, Morrison, etc., and it wasn't even that close) are so good that it's hard for me to imagine him not making it as a rotation player in the NBA. Since he was projected as a second-rounder (though he's pulled, right? I thought I heard that...), that would make him an outright steal, since second rounders aren't expected to even make the roster.

Also, beware of criticizing Shelden's athleticism...he was the second-strongest player at the draft combine.

I don't really know what to make of Redick. If you're looking for athleticism markers, Redick didn't get a whole lot of steals while at Duke, which suggests that he may be a notch below what he needs to be. Then again, he did OK at the combine. Gansey is a similar player to Redick but was a prolific thief, which may indicate more pro potential. Then again, Gansey is a year and a half older, so I don't quite know what to make of him.
Back to top
View user's profile Send private message Send e-mail
SGreenwell



Joined: 12 Feb 2005
Posts: 76
Location: Rhode Island

PostPosted: Tue Jun 20, 2006 12:22 am Post subject: Reply with quote
I'd hesitate to put down players with great college stats for a lack of athleticism. For example, Andrew Bogut was one of the best rookies last year, despite finishing near the bottom in overall athleticism at the combine last year. I'm not positive, but I'd say guys like Trajan Langdon are probably the exception, but I also haven't seen PER numbers for college players that far back.

While guys with freakish athleticism probably have a higher peak than guys like Bogut and Reddick, there's always the chance they'll never develop. See Jerome Moiso or Kwame Brown. I think it depends on a team's need, whether they should draft for peak or take their chances with lower risk, lower peak prospects with solid college stats.
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
Mark



Joined: 20 Aug 2005
Posts: 807


PostPosted: Tue Jun 20, 2006 11:46 am Post subject: Reply with quote
Primarily it is relative athleticism to your counterpart matchup that matters.

Bogut as a big could rank low on speed and agility overall against smaller faster players, it would be useful to see his rank compared to centers and powerforwards actually in the league.

Vertical jump amount is less important than vertical reach after jump- can you get your hand high enough to block a shot or dunk? and most can, so then quickness of jump actually matters more. Vertical jump and vertical reach jumping and just standing matters in rebounding but so do other things- ball tracking, fighting for position, box out technique.

Bench press maybe easy to measure but it isnt the measure of needed on the strength, they'd do better measuring leg and whole body strength- can they push a football tackling sled simulating trying to push a Ben Wallace or can they hold ground on against a sled simulating Shaq pushing them.

Even looking at big men in the draft camp group can be misleading because there are going to be strong atheletes who arent drafted or dont play much because skills play a large role in decinding that. On the surface from his stats and generally acceptable reviews Bogut's athleticism seems enough, at least for this year's level of minutes, time split 3-1 more PF time over bigger stronger Cs, veterans around to help against tough matchups and style of play. Milwaukee handled Bogut with about as much care and patience as you could ask for with a #1.

Combine tested physical attributes and athletic scores of perimeter scorers may matter the most. Shooters it matters but basketball IQ/execution can overcome disadvantages especially if the coach gives you the time and the assistance necessary. Bigs need a better assessment of game strength, quickness moving 4-8 feet, and of course competitiveness/toughness.
Back to top
View user's profile Send private message
jmethven



Joined: 16 May 2005
Posts: 51


PostPosted: Tue Jun 20, 2006 2:30 pm Post subject: Reply with quote
Has anyone calculated PER for the SEC? I'm getting a beyond absurd 40.12 PER for Joakim Noah (conference only; he was actually most dominant in SEC play) and I want to know if this is verifiable.
Back to top
View user's profile Send private message
tmansback



Joined: 12 Aug 2005
Posts: 129


PostPosted: Tue Jun 20, 2006 7:00 pm Post subject: Reply with quote
SGreenwell wrote:
I'd hesitate to put down players with great college stats for a lack of athleticism. For example, Andrew Bogut was one of the best rookies last year, despite finishing near the bottom in overall athleticism at the combine last year. I'm not positive, but I'd say guys like Trajan Langdon are probably the exception, but I also haven't seen PER numbers for college players that far back.

While guys with freakish athleticism probably have a higher peak than guys like Bogut and Reddick, there's always the chance they'll never develop. See Jerome Moiso or Kwame Brown. I think it depends on a team's need, whether they should draft for peak or take their chances with lower risk, lower peak prospects with solid college stats.


Bogut it also a 7 footer. If you have size the need for athletism is less. How many non athletic players that also are undersized have success in the NBA.

If Reddick was playing PG I dont think his wingspan, arm length, and athletism would be an issue. SG though is were your going up against 6'6 players that have 6'10 wingspans and have 40 inch verticals. One your going up against a Luke Ridnour, Sam Cassell like athlete. The other your going up against the guys that compete in Dunk Competitions. If your measurements suggest your like a 6'2 guard your in serious trouble if you cant beat those players off the dribble. I think for Reddick he needs a system like the triangle offense or to play with a big guard or forward that can handle the PG duty. I think teams like the Lakers, Cleveland, Miami, Houston are perfect fits.
Back to top
View user's profile Send private message
Mark



Joined: 20 Aug 2005
Posts: 807


PostPosted: Tue Jun 20, 2006 9:09 pm Post subject: Reply with quote
I would not be surprised if Redick plays 1/3 of his minutes at point.

If the Rocket take him maybe more. The new Matt Maloney?
Back to top
View user's profile Send private message
StatGuru



Joined: 16 May 2006
Posts: 16


PostPosted: Wed Jun 21, 2006 1:10 pm Post subject: Reply with quote
Tango wrote:
Richard Lu posted this article on 82games.
http://www.82games.com/nbadraft2006.htm

I think I understand his methodology. What I don't understand is how he calculated and correlated the success rates for the 2006 draft prospects with the similar players.

Anyone have any insights?


What are you guy's thoughts on Ronnie Brewer being ranked at the top of this draft class based off this system?
Back to top
View user's profile Send private message AIM Address
holymoly



Joined: 30 Jul 2005
Posts: 63


PostPosted: Wed Jun 21, 2006 4:42 pm Post subject: Reply with quote
JJ is a specialist. A specialist can't carry a team but can contribute efficiently with other players on the floor using a high % team poss. With Yao Ming and McGrady, I feel Reddick could be a good fit for the Rockets.

Last edited by holymoly on Wed Jun 21, 2006 5:09 pm; edited 2 times in total

holymoly



Joined: 30 Jul 2005
Posts: 63


PostPosted: Wed Jun 21, 2006 5:01 pm Post subject: Reply with quote
Mark wrote:
Clean, detailed display for a handful of players, Bargnani and Splitter especially. Rankings by individual stats including a lot of the advanced stats very interesting, probably unique information at least to many American eyes. Thanks. I find the typical euroleague stat lines on the internet a little hard to read.


Thanks Mark. Yes the euroleague captures a few extras in the boxscores (e.g. blocks against, fouls commited against) which can cloud them but are also interesting.

Mark wrote:
Euroleague stats vs college stats raw has aparently been studied some; would like to here more details about such efforts, methods, findings to date if the researchers are willing to summarize for us. Your looking at the rankings adds some additional insight and then if possible comparing the ranking in the two environments even more helpful.


Any articles or research already done in this area would be very interesting to hear about.

Mark wrote:
Look forward to more of these stat sets from your ambitious project to cover the whole euroleague scene and history. I assume you plan a new site to showcase these stats? Wonder what euroleague PER is needed to "make it in the NBA? 20?


Yes once the project is complete a new site is the preferred option. But will be posting interim findings prior to this. Translating a PER from the euroleague for success in the NBA is the challenge. Lyn Greer had the highest PER this year of 29 and a TS% of 70% in the italian league. Should he be in the NBA? The first step will be looking at current succesful euro players (Ginobli) numbers as well as unsuccessful imports and drawing some analysis here.

Mark wrote:
Splitter 26th rebound rating for centers, low compared to expectations? Bargnani at 37th for power forwards but in writeup it is labelled "solid"?
I would interpret this data differently for them as players coming to the NBA. Splitter as a PF/C may do ok instead of as a pure center. Bargnani lack of strength at rebounding likely to get worse if he stays at PF. SF- I dont know.


Splitters also played with one of the best big men in the euroleague (luis scola) this year so there could be an impact there (As it will again with his draft withdrawl). Bargnani's lack of strength is highlighted by most scouts as his key weakness, but for a forward his rebound rate is solid (rank 37 out of 100+).

Mark wrote:
Splitter much better assist and turnover ratings among peers than Bargnani goes against surface image of these guys over here. Bargnani stronger blocker compared to peers does as well. At least compared to my fiorst impressions reading draft profiles with Bargnani as more of a ball handler and Splitter as more traditional post player. Different positions- there but maybe not hereas both project mainly as PFs?


Thats why I want to get these stats out there, to challenge the subjective analysis of some commentators.

Mark wrote:
Perovic's low rebounding a major factor in the dramatic cooling of NBA interest from 1-2 years ago?


The phrase "nail on head" springs to mind
Back to top
View user's profile Send private message Send e-mail MSN Messenger
tmansback



Joined: 12 Aug 2005
Posts: 129


PostPosted: Thu Jun 22, 2006 6:48 pm Post subject: Reply with quote
Mark wrote:
I would not be surprised if Redick plays 1/3 of his minutes at point.

If the Rocket take him maybe more. The new Matt Maloney?


Yeah really doesn't have PG skills though. He might have to play more. His 2 rebound per game average in college has to be one of the lowest numbers for any SG taken in the first round period.
Back to top
View user's profile Send private message
Mark



Joined: 20 Aug 2005
Posts: 807


PostPosted: Thu Jun 22, 2006 7:48 pm Post subject: Reply with quote
Thanks for the additional comments HM. Look forward to hearing more as your project advances.

Redick could play point the way somewhat like the way Damon Jones plays it- 2.1 assists, 4.5 3pt attempts in 20some minutes. He probably would do better on assists than that.
Back to top
View user's profile Send private message
tmansback



Joined: 12 Aug 2005
Posts: 129


PostPosted: Fri Jun 23, 2006 6:51 pm Post subject: Reply with quote
Yeah thats why I think the Lakers would be a good fit for him. Getting in the triangle offense with no real PG needed would be perfect for him. He could take on that Paxson, Kerr role. If not there a place like Cleveland, Miami, Houston would be a good fit since they have other players that dominate the ball. Cleveland sounds like they like Daniel Gibson to play a similar role.

I don't know if Reddick a better passer than Jones. Jones did have a great assist per minute number in Milwaukee. For the most part he has no playmaking skills though. I don't know how he averaged so many assist in Milwaukee though.
Back to top
View user's profile Send private message
tsherkin



Joined: 31 Jan 2005
Posts: 247


PostPosted: Mon Jun 26, 2006 3:31 pm Post subject: Reply with quote
hpanic7342 wrote:

- Jordan Farmar looks like a total bust.


I wonder why you said this?

It can't be because of his performance at the Orlando Pre-Draft camp and certainly not with his athletic test scores in mind.

I don't see him exploding into a HoFer anytime soon but he's got the physical ability to play in the NBA and he's got the passing instincts. What makes you think he's going to be a bust?

alphamale wrote:
If Toronto doesn't pick Tyrus Thomas, they would be making a grave mistake. Everybody knows that Toronto had one of the leagues worst defense. They need a shot blocker and rebounder like Thomas, he would also bolster their running game because of his athletism. A front line consisting of Bosh, Villeneuva, and Thomas would be quite formidable indeed.

I wouldn't be surprised if Thomas ends up with a very high similarity score to Marcus Camby. Hopefully he's not as injury prone as the Camster.


Personally, I think it'd be a grave mistake if we TOOK Thomas.

My first thought is that Thomas would get abused terribly if we had to use him at center and there is no way we play him at the 4 with Bosh and Villa ahead of him, not without trading one. Bosh isn't going anywhere and Villa is 4" taller than Thomas and has already proved himself of use to our team.

Weakside shot-blocking isn't going to dramatically alter our team's defense, though it'll help and so will his rebounding. Toronto's main problem, however, is that we get abused in iso situations in the post. Thomas won't help that. Every team in the league will look at him and go "Oh God, post him up!" Anyone with a 6'10+ post scorer will attack him and Thomas will be as effective as a matador. He's too short, athleticism notwithstanding, to effectively guard the post on a nightly basis. He doesn't have Elton Brand's build (though he does have the wingspan).

Our frontcourt is lost, with or without Thomas. That's going to have to be built up slowly. The best the Raptors could do is draft the best player available and look to improve our backcourt defense with free agents.

Besides, Thomas scares me, he reminds me a terribly large amount of Stromile Swift with maybe more brains.
Back to top
View user's profile Send private message
And1



Joined: 14 May 2006
Posts: 7


PostPosted: Mon Jun 26, 2006 4:10 pm Post subject: Reply with quote
Does anyone know where I can find individual college player data that goes back at least five or six years? The goal is to measure how well success at the college level predicts success in the NBA?
Back to top
View user's profile Send private message
bchaikin



Joined: 27 Jan 2005
Posts: 686
Location: cleveland, ohio

PostPosted: Mon Jun 26, 2006 4:50 pm Post subject: Reply with quote
try this:

www.rivals.com/bstats.asp
Back to top
View user's profile Send private message Send e-mail Visit poster's website
bchaikin



Joined: 27 Jan 2005
Posts: 686
Location: cleveland, ohio

PostPosted: Mon Jun 26, 2006 5:09 pm Post subject: Reply with quote
Shelden Williams is not only underrated, he was Duke's best player last season.

Sheldon Williams also suffered from Duke's perimeter-oriented offense. In most half-court attacks, he could have averaged 20+. Defensively, he's a weak-side demon. And he's strong enough to bang with all of the 4s and maybe a few 5s in the pros.

actually shelden williams' college stats are very similar to the college stats of patrick ewing (georgetown, 81-82 to 84-85), especially their sophmore, junior, and senior seasons. combined in those 3 seasons here are their averages:

-----------------------G---min/g--pts/g--reb/g--ast/g--pf/g--st/g--to/g--bs/g--ScFG%
p.ewing-----------106----31-----16------9.8----1.0---3.0---1.2---2.4---3.5------62%
s.williams---------100----31-----15----10.0----1.0---3.1---1.4---2.5---3.5------61%

He could be the steal of the draft.

his stats look great and he was the college defensive player of the year not once but twice. hope he lands with the cleveland cavs somehow...
Back to top
View user's profile Send private message Send e-mail Visit poster's website
tmansback



Joined: 12 Aug 2005
Posts: 129


PostPosted: Mon Jun 26, 2006 5:26 pm Post subject: Reply with quote
Pat Ewing though was 7 feet tall. He had a much better chance for his game to translate. Shelden is one of those players that how will his game react when he goes from being the biggest player or one of the biggest players on the floor in most of his games to being smaller than many NBA SGs and SFs. He has a large wingspan. He also though has a very short standing reach. He blocked so many shots in college will it translate to the pros. I looked back the last few years and can't find any shot blockers that standing reach was so short. My bet his career is similar to Mike Sweetney. I think Mike's numbers were better all around.
Back to top
View user's profile Send private message
Mike G



Joined: 14 Jan 2005
Posts: 3574
Location: Hendersonville, NC

PostPosted: Mon Jun 26, 2006 7:17 pm Post subject: Reply with quote
Comparing current NCAA players' stats to those of 20 years ago is like comparing a minor league to a major league. In the '80s, college ball was full of super players, because they didn't skip straight to the pros. I'd guess in Ewing's G'town days, the NCAA could've filled 5-6 NBA teams' rosters -- competitively. Now, of course, nobody good enough to play NBA is staying in school.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
hpanic7342



Joined: 16 Feb 2005
Posts: 201


PostPosted: Mon Jun 26, 2006 7:20 pm Post subject: Reply with quote
I wrote that Farmar looked like a bust because his numbers aren't very good. In particular, he's very, very turnover-prone. Like, more so than anyone else in this draft.

Also, I wrote that before his performance in the Orlando Pre-Draft camp (against no other first rounders), and before his 42-inch vertical was measured at the athletic combine. I wouldn't count on his being anything more than a journeyman.

Also, what do you mean by "Stromile Swift with maybe more brains"? That's like saying, "like Cliff Robinson, but a better rebounder."
Back to top
View user's profile Send private message Send e-mail
bchaikin



Joined: 27 Jan 2005
Posts: 686
Location: cleveland, ohio

PostPosted: Mon Jun 26, 2006 8:52 pm Post subject: Reply with quote
Pat Ewing though was 7 feet tall. He had a much better chance for his game to translate. Shelden is one of those players that how will his game react when he goes from being the biggest player or one of the biggest players on the floor in most of his games to being smaller than many NBA SGs and SFs.

????

smaller than whom? many nba SGs and SFS? shelden williams is 6'9" and 260 lbs - forget many, can you name even one nba SG or SF bigger than that?...

he's just 22 years of age and a college senior and he's already heavier than close to half of all nba startings Cs. how much bigger will he get once he's on an nba weight training program?...

He also though has a very short standing reach. He blocked so many shots in college will it translate to the pros. I looked back the last few years and can't find any shot blockers that standing reach was so short.

what does standing reach have to do with shot blocking? shot blocking is about ability and desire and hustle. shelden williams has been one of the best shot blockers on the college level for the past 3 seasons, and has been a better shot blocker on the college level than a number of current nba Cs were - including mark blount, erick dampier, jeff foster, chris kaman, chris mihm, lorenzen wright, michael olowokandi, jason collins, p.j. brown, tony battie, kurt thomas, and i'm sure a few others...

are you assuming that his shot blocking ability will wither away once on the pro level, simply because he's just 6'9" and has a short standing reach? last i checked ben wallace was 6'9" and he's been one of the league's leading shot blockers for the past few seasons...

Comparing current NCAA players' stats to those of 20 years ago is like comparing a minor league to a major league.

how do you figure this?...

In the '80s, college ball was full of super players, because they didn't skip straight to the pros.

how many college players each season out of all those that declare for the draft before their college eligibility is up make an nba team, let alone stick for more than a year or two? there are about 4000 div I college players each season. are you saying that the few that make the nba so dilute the college pool of talent that what's left is significantly worse and thus a player's ability simply cannot be gauged properly based on what he did in college?...

Now, of course, nobody good enough to play NBA is staying in school.

aahh... so no sense then in any nba team 48 hours from now drafting the likes of brandon roy, or shelden williams, or randy foye, or rodney carney, or j.j.redick, or hilton armstrong, or paul davis, or mardy collins, or any other current college senior because their level of competition has been so watered down these past few seasons by other college players leaving college early that their true abilities can't be properly judged based on the statistics they produced...
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Mike G



Joined: 14 Jan 2005
Posts: 3574
Location: Hendersonville, NC

PostPosted: Tue Jun 27, 2006 6:13 am Post subject: Reply with quote
bchaikin wrote:


Comparing current NCAA players' stats to those of 20 years ago is like comparing a minor league to a major league.

how do you figure this?...
...
because their level of competition has been so watered down these past few seasons by other college players leaving college early that their true abilities can't be properly judged based on the statistics they produced...


Well, a 15-and-10 center ain't what it used to be.

Back around 1979 in the Big Ten, for example, a center would go 2 games against Clark Kellogg and Herb Williams, 2 vs Jay Vincent (and Magic Johnson), 2 vs Mychal Thompson and Kevin McHale, vs Phil Hubbard, vs Ray Tolbert, vs Joe Barry Carroll, ...

All these guys would have skipped out of their best college years, if that were in vogue then as it is now. Such competition cuts into your production.

Maybe the above names weren't all in college at the same time; but my general argument is -- it's my understanding -- that all players who are judged 'NBA-ready' are strongly encouraged/motivated to leave college and go pro. Watching an NCAA Finals full of future NBA allstars is unthinkable nowadays. Remember Jordan/Worthy/Perkins? Or Olajuwon/Drexler?
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
tmansback



Joined: 12 Aug 2005
Posts: 129


PostPosted: Tue Jun 27, 2006 5:27 pm Post subject: Reply with quote
Quote:
are you assuming that his shot blocking ability will wither away once on the pro level, simply because he's just 6'9" and has a short standing reach? last i checked ben wallace was 6'9" and he's been one of the league's leading shot blockers for the past few seasons...


Ben Wallace is a long player. If I remember correctly him and Elton Brand were equal in standing reach at 9'2.

Quote:
Now, of course, nobody good enough to play NBA is staying in school.

aahh... so no sense then in any nba team 48 hours from now drafting the likes of brandon roy, or shelden williams, or randy foye, or rodney carney, or j.j.redick, or hilton armstrong, or paul davis, or mardy collins, or any other current college senior because their level of competition has been so watered down these past few seasons by other college players leaving college early that their true abilities can't be properly judged based on the statistics they produced...


You may not like it but I heard a Rick Nash say those exact same things last year. The main question you have to ask is if a kid is so good why is he still playing college basketball. Look at the final four and the top teams in each conference in the Nation. UCLA a team with freshman and sophmores that aren't even that good dominate the Pac-10. UNC looses there top 7 rotation player from a year ago and by the end of the year had maybe the best team in the ACC. LSU, Ohio State, Memphis, Florida all being teams that were suppose to be good in a year or two good now. The amount of early entry players and high school entries has changed the game of college basketball. You never had the NCAA dominated by so many young teams.

Pretty much all the guys on your list above were considered second round late first round talent even up till there Junior year. The longer they play the better they get. That suppose to happen though. I'm a UW fan and even I wonder about the players that Brandon Roy went up against. You rarely go up against any quality seniors. Most of the players guys like Reddick, Williams, Roy go up against now are sophmores and freshman. There not only physically more mature than those players but have huge advantages in experience. Go look at the list of upperclass big men that entered the NBA draft over the last few years. They all have been average NBA players at best. Yet they stuffed the stat sheet in college. Now look at some of the bigs that have been or could have been in the ACC with Shelden this year. Dwight Howard, Josh Smith, Al Jefferson, Chris Bosh all could be playing ACC basketball. Why I have a feeling Shelden benefited that those players left early. Guys like Shelden do a good job throwing there more mature bodies around in college but when they played the game with men do they have the skill to compete at a high level.
Back to top
View user's profile Send private message
Mark



Joined: 20 Aug 2005
Posts: 807


PostPosted: Tue Jun 27, 2006 6:51 pm Post subject: Reply with quote
[quote="StatGuru"]
Tango wrote:


What are you guy's thoughts on Ronnie Brewer being ranked at the top of this draft class based off this system?


I think Brewer is a very difficult case to project. I dont know what he will deliver but his strong prediction isnt convincing and makes me even more wary of the confidence of the method.

Situation obviously has a huge impact on NBA results.


bchaikin



Joined: 27 Jan 2005
Posts: 687
Location: cleveland, ohio

PostPosted: Tue Jun 27, 2006 9:00 pm Post subject: Reply with quote
all i did was to compare one player's current college stats to another's from 20 years ago...

my general argument is -- it's my understanding -- that all players who are judged 'NBA-ready' are strongly encouraged/motivated to leave college and go pro

The main question you have to ask is if a kid is so good why is he still playing college basketball.

how many college players every year declare for the draft and forfeit their college eligibility? then out of those how many make an nba team? if they are all "nba ready" it must be a high percentage right? i read on one website that 62 college underclassmen declared for the draft this year - how many of those will even get drafted, let alone make and play for an nba team? you don't think the vast majority of those players (that have dreams of playing in the nba) might have bettered themselves playing college ball longer and having a better chance of making the nba later?...

again there are some 4000 college players at the level of div I college ball these days, and every year a very small percentage of underclassmen declare for the draft, an even smaller percentage get drafted, and then an even smaller percentage than that even make an nba team. and you're arguement is that that very small percentage dilutes the talent level of div I ball so much that the stats produced by that large percentage remaining isn't indicative of a player's ability? especially one like shelden williams that comes form from a large program such as duke which far more than most college teams plays against some of the best college players year in and year out?...

there is no lock on any college player being as good in the nba as he was in college. but i'll bet if you look at the rosters of nba teams the majority of their players came from the top college div I basketball programs that played against other top college programs in div I...

UCLA a team with freshman and sophmores that aren't even that good dominate the Pac-10.

this the same ucla team that went all the way to the ncaa championship game? because you say they aren't even that good makes the rest of all div I ball not that good??

let's take this arguement about freshman and sophmores opting out of college diluting the talent pool of div I ball even further. so if all freshman, sophmores, juniors, and seniors are not playing college ball the stats produced by players aren't comparable from year to year? decade to decade? fyi there was a time not too long ago (well not too long ago from my perspective anyway) when freshman weren't even allowed to play varsity ball - does that then mean that during all those years of div I college ball the stats produced by players really didn't reflect the true abilities of the players playing (sophomores thru seniors) because a small percentage of freshmen that were really good and could have played on the varsity weren't allow to play on the varsity? this arguement seems never ending...
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Mike G



Joined: 14 Jan 2005
Posts: 3584
Location: Hendersonville, NC

PostPosted: Wed Jun 28, 2006 6:16 am Post subject: Reply with quote
bchaikin wrote:
...when freshman weren't even allowed to play varsity ball - does that then mean that during all those years of div I college ball the stats produced by players really didn't reflect the true abilities of the players playing ?...


Yes. When Alcindor (and later Walton) were playing in the 'warmup' game, the rest of the conference were well aware that they were getting a 1-year break. (Didn't UCLA's freshman team regularly beat the varsity, until Wooden called it off?)

Of course you may 'compare' players, from different eras, conferences, divisions, hemispheres, etc. It's rather speculative, though.

Whether there are 4000 or 4,000,000 college players filling roster spots, the competitive upper echelon have moved on. A player who looks NBA-ready in the tournament may have needed that last 30 games to get that good -- or at least, to build that consensus.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
tmansback



Joined: 12 Aug 2005
Posts: 129


PostPosted: Wed Jun 28, 2006 7:31 pm Post subject: Reply with quote
Quote:
again there are some 4000 college players at the level of div I college ball these days, and every year a very small percentage of underclassmen declare for the draft, an even smaller percentage get drafted, and then an even smaller percentage than that even make an nba team. and you're arguement is that that very small percentage dilutes the talent level of div I ball so much that the stats produced by that large percentage remaining isn't indicative of a player's ability? especially one like shelden williams that comes form from a large program such as duke which far more than most college teams plays against some of the best college players year in and year out?...


One it doesn't matter if 4000 or 400000 players play college basketball. I would think that especially on this board the effect of 1 player on the game of basketball would not be taken lightly. The fact that were talking about 40-60 players every year with the vast majority coming from the top 6-8 conferences where talking about a much smaller number than 4000. We could easily say that 200 players that would have been top competition for Shelden Williams have left or not entered college since he came to Duke. The saying goes "someone has to score".

Your saying Shelden Williams and JJ Reddick are not only not benefiting from playing against those players. They also would have played with some of those players. Does Shelden Williams average 20 PPG if Luol Deng is still playing at Duke. I guarantee JJ Reddick wouldn't have averaged 28 if Deng is there. How many HighSchool kids would have been at Duke right now that bolted to the NBA. I could name one right off the top of my head Shaun Livingston commited to Duke. He obviously didn't go but if he does If Livingston goes to Duke Shelden Williams is the 4th best player on that team. Thats the difference between now and college basketball now and 20 years ago. Not only is Shelden not going up against some of the better players he would have went up against but his role in the offense is enhanced with players leaving. Thats going on all over the nation.

Quote:
would think that especially on this board the effect of 1 player on the game of basketball would not be taken lightly.


I also said the entire nation was filled with young teams that dominated there conference. This was the new college basketball. No time in history did you have so many teams with young talent dominate the conferences in the nation. 3 of the 4 teams in the final four were extremely young. The 4th team was a mid-major. The level of play in basketball was horrible this year because some of the best high school players in the nation never went to college and the good players they did have left early.

Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

Re: More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 1:18 am

Mountain



Joined: 13 Mar 2007
Posts: 1527


PostPosted: Mon Jun 23, 2008 11:28 am Post subject: Hollinger's Draft Rater articles Reply with quote
John, I like the new third year PER projection scale.

I was wondering if you had any further observations on the indicator value of "program quality" (apart from your team strength adjustment of player stats due to competition for shots, rebounds, etc.).

Some GM's seem to stick to big name schools (top 20-30?) more than others and I wonder if that returns additional value on average, beyond what is indicated by your method. Portland is one tagged as going that way recently. Using the recruiting judgment of top coaches (with the most choice) as a future quality indicator or believing that they add value in terms of learning and locking players into the fundamentals makes sense but of course there will be guys who don't really take permanent advantage of it and guys who are better than the quality of their program origin.


I wonder if any additional defensive stat that includes shot defense ("defensive rating" or expected vs actual analysis of the box scores of player's main counterpart match-up) were added to the method if it would improve prediction of overall success (beyond just PER success).

I also wonder if players weighing over 250-260 in college or above a certain body fat % are giving off a warning signal that on average suggests likelihood of less than otherwise expected return.

Want to share anything further about your consideration of these or other issues?
Back to top
View user's profile Send private message
asimpkins



Joined: 30 Apr 2006
Posts: 245
Location: Pleasanton, CA

PostPosted: Mon Jun 23, 2008 2:04 pm Post subject: Reply with quote
I'm curious about Drew Gooden ranking as the 3rd best college prospect since 2002 -- after only Beasley and Chris Paul I think. Of course, any prediction system will have anomalies, but that one really stands out.

I know nothing about college basketball, but did that guy squander a lot of potential or what? Why does the system like him so much?
Back to top
View user's profile Send private message
kbomb



Joined: 09 Aug 2005
Posts: 9
Location: Gardena, CA

PostPosted: Mon Jun 23, 2008 3:50 pm Post subject: Reply with quote
is it just me or his year-3 projections look more like rookie year projections? i find it odd that none of the players studied has year-3 projection of PER over 20. to me, those numbers look closer to what they did in their rookie seasons.
Back to top
View user's profile Send private message
Mountain



Joined: 13 Mar 2007
Posts: 1527


PostPosted: Mon Jun 23, 2008 4:30 pm Post subject: Reply with quote
Gooden got to almost 20 PER in his year 3, then lost 20% off his shot attempts per 36 minutes in his contract year. How much of that was him / warranted? I don't know the story step by step in detail but I know they got him re-signed at modest cost but then they got lower production thereafter too. How much was health related? In Chicago his numbers bounced positive from the lackluster previous 2 years. Wonder which level of production he brings next season.
Back to top
View user's profile Send private message
THWilson



Joined: 19 Jul 2005
Posts: 164
Location: phoenix

PostPosted: Mon Jun 23, 2008 6:13 pm Post subject: Reply with quote
I'd be interested in seeing if there's a correlation between the model residuals and draft position.
Back to top
View user's profile Send private message
Kevin Pelton
Site Admin


Joined: 30 Dec 2004
Posts: 979
Location: Seattle

PostPosted: Mon Jun 23, 2008 6:51 pm Post subject: Reply with quote
asimpkins wrote:
I know nothing about college basketball, but did that guy squander a lot of potential or what? Why does the system like him so much?

Pretty much any system you use will put Gooden that high, I suspect. His college numbers were off the charts.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
94by50



Joined: 01 Jan 2006
Posts: 499
Location: Phoenix

PostPosted: Mon Jun 23, 2008 10:55 pm Post subject: Reply with quote
Mountain wrote:
Gooden got to almost 20 PER in his year 3, then lost 20% off his shot attempts per 36 minutes in his contract year. How much of that was him / warranted? I don't know the story step by step in detail but I know they got him re-signed at modest cost but then they got lower production thereafter too. How much was health related? In Chicago his numbers bounced positive from the lackluster previous 2 years. Wonder which level of production he brings next season.

And Bulls fans everywhere wonder with you.
Back to top
View user's profile Send private message
tenkev



Joined: 31 Jul 2005
Posts: 20
Location: Memphis,TN

PostPosted: Tue Jun 24, 2008 9:04 am Post subject: One huge problem with Hollinger's Draft Rater Reply with quote
A huge problem with Hollinger's draft rater is that he uses PER as the dependent value and the flaws inherent in PER will be inherent in the Draft Rater. PER does a horrible job measuring defense and non-box score stuff like the value of a big man with range and of drawing double teams and setting picks. When is Hollinger going to get on the adj. +/- wagon? Its by far the most relevant measure to use in any regression analysis like this. It is more accurate with fewer biases and its high variation is not that much of a factor when it is used in this way.
Back to top
View user's profile Send private message Send e-mail AIM Address
Ryan J. Parker



Joined: 23 Mar 2007
Posts: 711
Location: Raleigh, NC

PostPosted: Tue Jun 24, 2008 11:02 am Post subject: Reply with quote
Quote:
When is Hollinger going to get on the adj. +/- wagon?


The availability of college basketball play-by-play data leaves much to be desired.
Back to top
View user's profile Send private message Visit poster's website
tenkev



Joined: 31 Jul 2005
Posts: 20
Location: Memphis,TN

PostPosted: Tue Jun 24, 2008 1:08 pm Post subject: Reply with quote
I wasn't talking about adj. +/- for college years. I was talking about the running the same regression off of third year adj +/- instead of third year PER.
Back to top
View user's profile Send private message Send e-mail AIM Address
John Hollinger



Joined: 14 Feb 2005
Posts: 175


PostPosted: Tue Jun 24, 2008 2:03 pm Post subject: Reply with quote
Adj +/- appears to have at least as many warts as PER, actually, most notably the "bad backup effect." PER is at least controlled more or less by the player in question, as opposed to what happens when he leaves the court.

As for Gooden, this is where good off-court scouting can come in. This rates pro potential, not what they'll actually become, and Gooden and Sweetney are examples of why sometimes "potential" never becomes "actual". Gooden can't remember plays and never improved, and Sweetney ate himself out of the league.

Finally, as to why no 20+ projections -- I'm looking at that. It may just be that it's more difficult to peg somebody as a surefire star than we think ... or it may be that my method is too tightly bound around the midpoint. The beauty of this is as we get more data with each passing year, the projections should automatically improve.
Back to top
View user's profile Send private message AIM Address
Mountain



Joined: 13 Mar 2007
Posts: 1527


PostPosted: Tue Jun 24, 2008 2:10 pm Post subject: Reply with quote
Gooden in his year 3 rated the highest on adjusted +/- of any Cav, ahead of LeBron James. In year 4 Gooden ranked the worst Cav on the same measure.

Ferry takes charge and they shift from Silas to Brown, McInnis to Snow... and Gooden from burgeoning star to role-playing big?

In 04-05 Gooden-James worked well enough under Silas / McInnis (above James' average +/-) but under Brown/Snow Gooden became one of James' worst pairs.

How much blame should go to Gooden for weaker execution and how much is it due to different handling / perhaps role, less effective with and for him?

It is about team of course, not Gooden personally, and Cavs have had a fairly high level of success doing it their chosen way. But one step short of championship.


Gooden on his 3rd team by then (and now 4th) so the story is more complicated than just the numbers and I assume much of the buzz about him is fair. But he did improve from year 1 to 3 nicely on TS%, especially on DR%, assists, blocks, TOs. He can play well. And in short Bulls run was back to career average on TS%, above on both types of rebounding, at career high on assists, blocks and 2nd best on TOs. He can still play well in non-Brown / James - Cavs system...and least for a while.

It is true he has bounced around in his performance and that might continue too and it reflects on him and is not all about that system / role. But this year might help some with sorting out how much of the decline was "context" influenced. The system that was very good for James was apparently not as good for Gooden. Could things have been made to work better- for both and ultimately the team? More like under Silas (a big, perhaps both more sympathetic and more knowledgeable about how to use an inside big)? I don't know, just arriving to the issue. Those closer know how hard it was worked and everything involved. Not every player pairing or player=coach pairing or player-role assignment-team design works out.

Last edited by Mountain on Tue Jun 24, 2008 8:06 pm; edited 5 times in total
Back to top
View user's profile Send private message
Mountain



Joined: 13 Mar 2007
Posts: 1527


PostPosted: Tue Jun 24, 2008 4:12 pm Post subject: Reply with quote
Cavs better than +2 on adjusted +/-:

04-05 4
05-06 4
06-07 3
07-08 2 James and the traded Gooden

Interesting trend.

And James after being #2 in league on this measure in 05-06, declined in 06-07 and again in 07-08.


Over the same time-period Kobe's Lakers went from 4, 3, to I think 1... and then to 5 this season. LeBron and the Cavs might benefit from a similar movement back to broader group of positive +/- contributors- by support players "picking it up" or "the system maxing their potential contribution as well" or some combination.


With the errors associated with adjusted +/- measurements these trends might not be as clear as they appear. But it still gives food for thought.


Number of +2 or better players on NBA champions last 4 years: 5, 6, 4 and 6 this season.

Last edited by Mountain on Tue Jun 24, 2008 7:26 pm; edited 2 times in total
Back to top
View user's profile Send private message
kevC



Joined: 28 Jun 2007
Posts: 17


PostPosted: Tue Jun 24, 2008 7:12 pm Post subject: Reply with quote
John, how high would have Carl Landry have rated on your new system. I think he was your biggest miss in last year's draft.
Back to top
View user's profile Send private message
Mountain



Joined: 13 Mar 2007
Posts: 1527


PostPosted: Tue Jun 24, 2008 10:16 pm Post subject: Reply with quote
Number of +2 or better players on this year's other main contenders;

(from basketballvalue / heavily affected by playoffs)

Detroit 5
Orlando 3
Dallas 5
Houston 4
Phoenix 3
New Orleans 4
Utah 6
Denver 3
San Antonio 3

(Portland 5, but not Roy by this measure)


page 2 missing

Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

Re: More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 1:21 am

jeffpotts77



Joined: 18 Feb 2005
Posts: 150
Location: Cambridge, MA

PostPosted: Wed Oct 25, 2006 11:25 am Post subject: Isiah Thomas' Draft Record 1995-2005 Reply with quote
Curious about how well Isiah has drafted over the past decade? I did a very simple analysis using PER, found at basketball-reference.com and draft archives found at nbadraft.net. (Thanks to E.H. Munroe for the editing.)
Back to top
View user's profile Send private message Yahoo Messenger
Ben



Joined: 13 Jan 2005
Posts: 266
Location: Iowa City

PostPosted: Wed Oct 25, 2006 12:39 pm Post subject: Reply with quote
Nice post. Incidentally, I would've thought that #2 picks would have been better over the years.
Back to top
View user's profile Send private message
jeffpotts77



Joined: 18 Feb 2005
Posts: 150
Location: Cambridge, MA

PostPosted: Wed Oct 25, 2006 1:36 pm Post subject: Reply with quote
Ben wrote:
Nice post. Incidentally, I would've thought that #2 picks would have been better over the years.


Interesting that you point that out because I noticed the same thing. Darko Milicic played a large role in bringing that average down. It might have helped to put a footnote explaining this, but the post had way too many footnotes to begin with.

It would be extremely interesting to examine the average PER of every draft position in the first round over a large span of time. It would need to be weighted by minutes played; that way the Darkos wouldn't distort it too much. I'll ponder this but if anyone else beats me to it, and does it with finesse, I won't be too disappointed.
Back to top
View user's profile Send private message Yahoo Messenger
Ed Küpfer



Joined: 30 Dec 2004
Posts: 787
Location: Toronto

PostPosted: Wed Oct 25, 2006 2:59 pm Post subject: Reply with quote
jeffpotts77 wrote:
It would be extremely interesting to examine the average PER of every draft position in the first round over a large span of time.


Not PER, but I posted this to the Sonics froup a few years ago, using some wins above replacement hack.
_________________
ed
Back to top
View user's profile Send private message Send e-mail
jeffpotts77



Joined: 18 Feb 2005
Posts: 150
Location: Cambridge, MA

PostPosted: Wed Oct 25, 2006 3:10 pm Post subject: Reply with quote
Ed Küpfer wrote:


Not PER, but I posted this to the Sonics froup a few years ago, using some wins above replacement hack.


Very interesting! Correct me if I'm wrong, but by your measure, the third draft pick seems to generate slightly more wins than the second. Considering what my analysis showed, is there a trend here?
Back to top
View user's profile Send private message Yahoo Messenger
KeeneKaufmanWheeler



Joined: 02 Aug 2006
Posts: 72


PostPosted: Wed Oct 25, 2006 5:20 pm Post subject: Reply with quote
I don't think that the mystery of the underperforming #2 pick is limited to basketball. The NFL has it's share of Manning/Leaf and Bledsoe/Mirer drafts.
Back to top
View user's profile Send private message Visit poster's website
hpanic7342



Joined: 16 Feb 2005
Posts: 201


PostPosted: Thu Oct 26, 2006 12:28 am Post subject: Reply with quote
I liked your post. I wouldn't recommend weighting by minutes played if you're looking at how the picks have performed. Darko was just as much a #2 pick as Jason Kidd was...weighting by minutes would just cover up whatever systematic problems there might be with picking #2.
Back to top
View user's profile Send private message Send e-mail
KnickerBlogger



Joined: 30 Dec 2004
Posts: 180


PostPosted: Thu Oct 26, 2006 8:47 am Post subject: Reply with quote
I talked about this subject, but instead I looked at the players drafted in those years in Zeke’s Eye For The Draftee Guy
_________________
KnickerBlogger.Net - now indispensable!
Back to top
View user's profile Send private message Visit poster's website AIM Address Yahoo Messenger
jeffpotts77



Joined: 18 Feb 2005
Posts: 150
Location: Cambridge, MA

PostPosted: Thu Oct 26, 2006 10:31 am Post subject: Reply with quote
KnickerBlogger wrote:
I talked about this subject, but instead I looked at the players drafted in those years in Zeke’s Eye For The Draftee Guy


Nicely done! That was pre-Channing and you were cautiously optimistic. It's just such a shame that he traded away those draft picks to the Bulls.
Back to top
View user's profile Send private message Yahoo Messenger

Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

Re: More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 1:32 am

mojo



Joined: 09 Dec 2005
Posts: 10


PostPosted: Fri Jun 16, 2006 2:45 pm Post subject: College PER and the draft Reply with quote
I'm trying to find information re: NCAA to NBA success. The article mentioned in a previous thread on 82games is intriguing but his methodology is unclear. The best site I've found for college stats is kenpom. Any sites have college PER for teams and individuals? Also, any other mathematical analysis for the draft other than 82games, this site, and hoopsanalyst? Thanks.
Back to top
View user's profile Send private message
tmansback



Joined: 12 Aug 2005
Posts: 129


PostPosted: Sat Jun 17, 2006 6:34 pm Post subject: Reply with quote
I would think one of the major problems with translating college PER to the pros is the major difference in talent from one game to the next. In college you see a lot of players pad the stats against certain teams but against the stronger competition there production drops off drastically.
Back to top
View user's profile Send private message
asteroid



Joined: 13 Aug 2005
Posts: 9


PostPosted: Sun Jun 18, 2006 8:09 pm Post subject: Reply with quote
The two biggest problems you have when trying to translate the NCAA or Euro stats to NBA are height/length & athleticism. Every level of basketball that you go up, length & athleticism become more crucial.

Obviously, you have the dreaded undersized-two (Respert, Monroe, Ed Gray, etc.) or the undersized post guy (Lane, Rogers, Bryon Houston, endless list). Granted , the guy in the recent 82games study did use height, but what about the importance of arm reach. This definitely is a factor for guys like Brand, Boozer, & Randolph being able to produce in the pros. And this could definitely effect Sheldon this year cause his length was about 5-6 inches shorter than Brand or Okafor.

But one key variable missing from any formula trying to translate NCAA or foreign league stats to the pros is degrees of athleticism; can't put a measure on that. For example in the 82games study, he tries to compare Darius Washington to Arenas, but I would argue that Arenas is one of the best athletes in the NBA, while Darius is not in his league.

This will definitely also wreck havoc with Euroleague stats as well. Hollinger likes to pump Macijauskas. Granted, the guy should be seeing more playing time then he got from Bryon this year, but his translated Euro stats are going to be screwy cause he's 6-3 & not a good athlete. Sort of the same problems you could have trying to evaluate Redick's numbers this draft season. Scola's another guy you have to be careful with, cause he's in that cautious range of 6-8/6-9 and his athleticism could be questioned.

Another thing that is hard to put in a formula is the style of play or role. Look at a guy like Vince Carter, here's guy who was held in check and used in a particular way by Dean, and only averaged at the most 15/game at UNC.

Crow
Posts: 6251
Joined: Thu Apr 14, 2011 11:10 pm

Re: More recovered threads related to the draft

Post by Crow » Thu May 19, 2011 1:35 am

Eli W



Joined: 01 Feb 2005
Posts: 402


PostPosted: Mon Apr 04, 2005 11:17 am Post subject: Blazers stat-based draft system Reply with quote
Quote:
At the core of the secret are the Blazers' hiring of a brainy millionaire who was featured in a best-selling book, a bevy of high-tech computer programs funded by owner Paul Allen, and an open-minded way of looking at statistics.

This much is known: The mystery millionaire, who used an alias while being featured in the book and has requested to stay anonymous in conjunction with the Blazers' project for three more months, does quantitative analysis on basketball statistics for the Blazers.


http://www.oregonlive.com/sports/oregon ... 175340.xml
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Dan Rosenbaum



Joined: 03 Jan 2005
Posts: 541
Location: Greensboro, North Carolina

PostPosted: Mon Apr 04, 2005 1:03 pm Post subject: Re: Blazers stat-based draft system Reply with quote
John Quincy wrote:
Quote:
At the core of the secret are the Blazers' hiring of a brainy millionaire who was featured in a best-selling book, a bevy of high-tech computer programs funded by owner Paul Allen, and an open-minded way of looking at statistics.

This much is known: The mystery millionaire, who used an alias while being featured in the book and has requested to stay anonymous in conjunction with the Blazers' project for three more months, does quantitative analysis on basketball statistics for the Blazers.


http://www.oregonlive.com/sports/oregon ... 175340.xml

Thanks for passing this on. As teams use more sophisticated methods of analyzing statistics, I think the biggest danger is taking the numbers too seriously. Analyzing statistics is more than just writing sophisticated programs. It is understanding biaes in the data, understanding sampling variation, understanding limitations of the data, etc. Developing such understanding is tough and sometimes brilliant folks - folks who are way smarter than I am - who are great programmers are not very good at understanding nuances of their data. Even formal training is not a cure for this. For example, based on press accounts, I have always been skeptical of how well Winston (a professor at Indiana) and Sagarin understand the nuances of their data; I just have a hard time figuring out why they say some of the things they say about what they do. I try not to say things like this very often, but from my vantage point some of things they say just sound really, really naive.

Over time I am coming more and more to the conclusion that the real selling point of a good statistical analyst is not the methodology he tries to sell. But instead his (or her) greatest value comes in telling non-stats folks when the data just does not have much to say, when basing a decision on a gut instinct probably is no worse than basing it on statistical analysis. And that takes good judgment - good judgment that comes through a combination of training, experience, and skill.

It is funny, but I trust a lot of you more with data analysis than I do some of my economist brethren who have a lot more formal training. More and more over time I have realized that understanding the nuances of data is a skill and there are lots of folks with great math skills and great programming skills who just struggle with the nuances of data.

Now I am not saying that these folks in Portland are programmers who don't understand data, but when I hear folks appearing to oversell what they they are doing, the alarm bells in my head start to go off.
Back to top
View user's profile Send private message Send e-mail Visit poster's website Yahoo Messenger
gabefarkas



Joined: 31 Dec 2004
Posts: 1313
Location: Durham, NC

PostPosted: Mon Apr 04, 2005 5:57 pm Post subject: Reply with quote
any thoughts to who the mystery man can be?
Back to top
View user's profile Send private message Send e-mail AIM Address
jambalaya



Joined: 30 Jan 2005
Posts: 282


PostPosted: Mon Apr 04, 2005 6:10 pm Post subject: search Reply with quote
when i heard bay area and planning on doing even bigger projects than working with the blazers, i had an immediate guess. but then dan's comment made me wonder that maybe it wasnt the most obvious possibility, the comment about being already connected to paul allen's companies work didnt fit what i know in that case, but thinking about some of the companies paul allen has owned or still does (sports and sports information related) it still could be that person. out of respect for the desire for temporary anonymity, which probably has a reason, i won't float the name.


in case it is not, just out of curiousity i also researched a few other leads of folks this mystery person might be, including some others registered here. i also looked at a few that worked on advanced scout in the past, thinking maybe one is going out of their own. i didnt get a strong feeling on any of these.

by the way, could anyone clarify does advancescout still live (outside of GM office use of prior software)? i heard ibm gave it up and i couldnt tell looking at the successor owner, virtual gold company site if it was still alive or just a legacy listing of their past work & property. there is some talk of a new version there, but no date to clarify whether this listing was recent. i saw a lot of references to it in college courses on the web but they all seem pretty surface. anyone have or previously seen/used the software? any comments about what it did good& bad in its day?


Back to top
View user's profile Send private message
gabefarkas



Joined: 31 Dec 2004
Posts: 1313
Location: Durham, NC

PostPosted: Mon Apr 04, 2005 7:11 pm Post subject: Reply with quote
what's advancescout?
Back to top
View user's profile Send private message Send e-mail AIM Address
bchaikin



Joined: 27 Jan 2005
Posts: 685
Location: cleveland, ohio

PostPosted: Mon Apr 04, 2005 7:25 pm Post subject: Reply with quote
inderpal bhandari was the former IBM research fellow who did the advance scout stuff:

www.virtualgold.com/company_founder.html

don't know if he still does this but he was trying to sell it to nba teams a decade ago...
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Kevin Pelton
Site Admin


Joined: 30 Dec 2004
Posts: 978
Location: Seattle

PostPosted: Mon Apr 04, 2005 9:37 pm Post subject: Reply with quote
If the date wasn't wrong, I would have thought this an April Fool's joke, what with all the bizarre secrecy. It will be fascinating to see what eventually comes of this.

Some more background on Advanced Scout:

Official site: http://www.virtualgold.com/solutions_sports.html

http://citeseer.nj.nec.com/bhandari97advanced.html

NBA.com puts it to use: http://www.nba.com/playoffs2001/ibm/ibm ... rticleList

A nice example of how the Magic used the program:
http://www.cnnsi.com/features/1998/week ... uting.html

A key quote from that one:
Quote:
Sometimes the numbers lie. A couple of lucky bounces can convert two turnovers into baskets and cloud the data on the point guard you are evaluating. The use of video—as well as common sense—can help a team check the statistics.


Essentially, the program data mines to find really abnormal plus-minus-type statistics for given players/teams, usually against a given opponent. That does make one wonder how much of the findings are in reality sample-size based flukes. ...
Back to top
View user's profile Send private message Send e-mail Visit poster's website
KD



Joined: 30 Jan 2005
Posts: 163


PostPosted: Tue Apr 05, 2005 7:00 am Post subject: Reply with quote
Sorry to bring the discourse down a notch, but does Portland's particular equation include and/or account for any arrests for handgun possesion?

Obviously, any step up from the Whitsitt/Nash scouting theory would be a plus, but so would scouting players who eclipsed 600 on their SATs.
Back to top
View user's profile Send private message
jambalaya



Joined: 30 Jan 2005
Posts: 282


PostPosted: Tue Apr 05, 2005 10:04 am Post subject: advance scout Reply with quote
yeah i had seen these references and more. got the impression that the company might not have active running maintenance contracts today.
Back to top
View user's profile Send private message
kjb



Joined: 03 Jan 2005
Posts: 864
Location: Washington, DC

PostPosted: Tue Apr 05, 2005 10:26 am Post subject: Reply with quote
Meant to post this here, but I accidentally did it in another thread:

Back at Hoopsworld, we created a kind of character formula.

Something like:

(charitable donations + non-court ordered hospital visits + .5 * acquittals / arrests + fights with teammates + insubordination + illegitimate children + std infections + positive drug tests) * (approximate value / salary)
Back to top
View user's profile Send private message AIM Address Yahoo Messenger
Eli W



Joined: 01 Feb 2005
Posts: 402


PostPosted: Tue Mar 06, 2007 1:31 pm Post subject: Reply with quote
And the mystery millionaire is...Jeff Ma from PROTRADE (and the MIT conference technology panel).

Quote:
Two years ago, PROTRADE sought to create a way to project success in the NBA based on college statistics. Armed with data from the last 20 years, PROTRADE ran analysis on many different statistical categories like assists per minute, steals per minute, blocks per minute, assist-to-turnover ratio, and body-type information like age, height and weight, to determine what best predicted stardom in the NBA.

We took our results to a few NBA teams and eventually were hired by the Portland Trail Blazers to assist in their draft. For the last two seasons, PROTRADE has played a role in the Blazers' decisions of whom to select.


The article looks at the ten rookies that were projected to do the best by their system:

http://www.protrade.com/content/Display ... 19e0af0ea0
Back to top
View user's profile Send private message Send e-mail Visit poster's website
HeatherA



Joined: 03 Aug 2006
Posts: 55


PostPosted: Tue Mar 06, 2007 2:06 pm Post subject: Reply with quote
I did a bunch of research on Advanced Scout last summer. It appears that they were fabulously successful just as long as they were free (IBM initially offered the program for free to all of the NBA teams). Lots of positive press, GMs lauding their capabilities, etc.

In about 2001 Virtual Gold spun off and started charging (in the ballpark of $20,000 I think) for the program. It looks like essentially all (if not all) of the teams that were using it stopped at that point. Other than that Virtual Gold website mentioned above, there's nothing out on the web about it after about 2001.
Back to top
View user's profile Send private message
mateo82



Joined: 06 Aug 2005
Posts: 211


PostPosted: Tue Mar 06, 2007 2:10 pm Post subject: Reply with quote
edit: didn't realize this post was so old!
Back to top
View user's profile Send private message
holymoly



Joined: 30 Jul 2005
Posts: 63


PostPosted: Wed Mar 07, 2007 9:33 am Post subject: Reply with quote
Generally, I like their approach, however a few comments;

1. Are they simply looking at the percentle10 value of ORtg and DRtg, surely then their 'success' rating would be skewed to players playing for better teams (teams with higher ORtg). I would suggest that a net ratio from the individuals ORtg against the team and perhaps league averages would be better.
2. Looking at Portlands Draft picks for 2006, it doesnt seem that they have taken these 'forecasts' that seriously. Joel Freeland was a huge surprise for the first round and they wouldnt have had any stat forecasts for him (plays in Spain). Plus they traded away Thomas (2nd in ProTrades forecast) for Aldridge (not in top 10). They also got sergio rodriguez, another guy who wouldnt have been run in the model. I'm not saying that you simply build a model and pick in the order the numbers tell you too BUT what these numbers are telling Portland dont seem to have been explained very well by ProTrade.

On another note I 100% agree with Dan's comments. Our challenge is to get basketball people to realise how the numbers help them make better decisions. There is no substitute for a great coach but combining a great coach and analytical insight can only help you win.
_________________
.....
60% of the time, it works every time

Last edited by holymoly on Thu Mar 08, 2007 4:52 am; edited 1 time in total
Back to top
View user's profile Send private message Send e-mail MSN Messenger
tpryan



Joined: 11 Feb 2005
Posts: 100


PostPosted: Wed Mar 07, 2007 9:51 pm Post subject: Reply with quote
Holymoly:

Is any of the work on logistic regression models that you mentioned in the public domain? I would guess not but, if so, I would like to look at it. The reason I ask is because I published a paper five years ago with one of my students (see http://www.ingentaconnect.com/content/a ... 3/art00001 ) that showed problems with the method of maximum likelihood under certain conditions. (This is the only estimation method available in almost all software for logistic regression.)

Since then I have done additional joint research on the subject with one of the co-founders of a well-known software company and we will submit our research for publication if he can ever find time to return to our paper.

It's possible that the logistic regression data that are being used might have "near separation", which cannot be easily explained here but which can create havoc with maximum likelihood coefficients.

If any of you who work for NBA teams are involved in such work, I would be interested in looking at any data that are available, suitably disguised if necessary (or I could sign a Statement of Confidentiality, etc.)

Tom

Post Reply