It's spelled with an r, not an n

[Bobbofitos]

The thing is there's still a strong line in the sand between teams who understand this and teams who don't. I mentioned Cleveland, but take Toronto, who is a good comparison because they were in a similar position in the summer of 2010 (their superstar left and they had to rebuild). The Raptors under Bryan Colangelo as a whole have been a trade value accumulation apocalypse. To give an example Toronto traded TJ Ford, Rasho and a 1st round pick (that Indy used on Hibbert) for Jermaine O'Neal. This was a terrible trade value move because O'Neal, even disregarding his injury history, was so overpaid that he didn't have much value on the market, and he was to be a UFA after 2 years, while being old. It was essentially emptying the bank account to buy a stock before it's about to fall off a cliff. Ford, despite his injury, was still a youngish PG, and a 1st round pick speaks for itself. It also killed the team's capspace. Half a season they traded another 1st just to get rid of Jermaine O'Neal, so they could make a run at Odom, Hedo, Ariza in the offseason, none of which were likely to have eventual trade value. Colangelo also killed the team's capspace repeatedly with other moves like spending all the team's capspace on veterans in Parker/Garbojosa/Rasho in 06, Kapono the year after, Turkoglu in 2009. He does not understand capspace is a trade value building asset, whereas Cleveland's entire gameplan the last 2 years has been to use capspace as a trade asset. The summer of 2010 was particularly stupid for Toronto though. For one thing Bosh was allowed to leave in UFA for virtually no value (a couple 1sts in a S&T) despite everyone and their mother knowing he wanted out. We actually turned down Michael Beasley (coming off his 2nd season, still had value/upside) in the S&T for Bosh, again another strike for "doesn't understand trade value". The team then signed Linas Kleiza and Amir Johnson to a combined 50 million over the next 4-5 years, despite both being veterans on a team that didn't need to win for years. While Cleveland has spent the last 2 years trading capspace for picks, Toronto spent it on veterans and overpaid MLE guys. This offseason was more of the same. Lowry is a good player but a bad team should be wary of giving up a lotto 1st for him, because Lowry is a UFA in 2 years and could very well leave for nothing. Landry Fields was yet another disregard of capspace as a trade asset and he, Kleiza and Amir will be making 15 mil+ a year combined for virtually dispensible play, all locked up for multiple seasons. The team now Jerryd Bayless was renounced, giving the team no value in return for a player that likely could've fetched some at the TD. None of this indicates a team who realizes the concept of stockpiling trade assets - like at all.

This isn't quite Billy Beane's A's, but it's an indication of a strategy one team is taking while half the NBA or more, are sort of dinosaur-ing it. I have a feeling that if teams started to make decisions based on the idea of "maximizing the efficiency of every asset", many moves would be different

Fields, Amir, and Lowry were/are fantastic deals.

I think they're a team very much on the rise, and that's large part because of BC.

[Bobbofitos]

Valanciunas hyped up as "#2 if he had been in 2012" despite having horrible feel for the game

How have you come to the conclusion that he has "no feel for the game"? (What does that even mean?)

I would actually say the opposite - for a young big man, he has incredible skill and polish.

Ekpe Udoh picked ahead of Greg Monroe, Al-Farouq Aminu picked ahead of Hayward and George, Thabeet picked ahead of Harden. Udoh/Aminu/Thabeet all have serious feel for the game/skill problems, Monroe/Harden/Hayward/George skilled with feel but not as flashy physically.

RAPM, at a minimum, would argue that Ekpe has been much better than Monroe. Monroe can't guard a chair while Udoh seems to be a hidden gem.

Most of the draft busts fit a particular profile. They are physically gifted (athletic, long, etc.) but have weak skill or feel. So they end up journeyman bench players.

Sorry, but this is just throwing words on paper. The reason there are a lot of physically gifted flameouts is because almost every single prospect is physically gifted.

[Dr Positivity]

Physically gifted comparatively. Andrew Nicholson and Tyler Zeller are great athletes compared to human beings, but Thomas Robinson is an elite athlete for an NBA player. Thus Robinson is called a plus NBA athlete, Zeller and Nicholson aren't drafted for their athleticism. Players are drafted on what they can separate themselves from others in the NBA with

In regards to feel for the game, what I look for:

- How 'easy' does the game look for them.
- How 'natural' does their basketball playing acumen look. Often their game is "aesthetically pleasing"
- How much TIME does it seem like they have. Players with great feel make the game look slower than it is. Players with weak feel are in a rush.
- How spatially aware is the player of their surroundings? Players with great feel IMO seem to know where they are compared to their teammates. This might be what feel for the game is period, tangibly (spatial awareness)

Valanciunas IMO is an awkward player, usually plays like he is in a rush, and has a poor awareness of the court, particularly defensively. He gets lost on pick and rolls a lot, picks up bad fouls, etc. I see very little spatial and positional recognition of the players around him. He is "raw" from a mental perspective. The biggest thing he has going for him is skill. But IMO skill is almost entirely separable from feel for the game. If you watch Motiejunas, there's a player who's "feel for the game" is on the opposite end. He always seems to be performing moves smoothly and in control and as if he knows where players are spatially. He is much more slick and polished

One of the reasons I say skill is separable from feel is I am rating all the players in the NBA and have found plenty of players with high skill/poor feel for the game or vice versa. To give some examples of players that IMO have high feel for the game for their position but average/weak skill: Avery Bradley, Gerald Wallace, Ronnie Brewer, Tony Allen, Bruce Bowen, Nick Collision, Tyson Chandler, Andrew Bogut. Now high skill, weak feel for the game: Michael Beasley, Andray Blatche, Al Harrington, Chris Kaman, Leandro Barbosa, Kevin Martin, Jamal Crawford, Aaron Brooks, JR Smith. But really only a few are needed, if Beasley can be an extremely skilled PF with a weak feel for playing basketball and Brewer can be an extremely unskilled SG with a strong one, clearly skill is not a requirement for feel for the game, or basketball intelligence, or whatever it may be called. Of course there is also historical examples, players like Magic, Bird and Duncan have something going on awareness wise that goes beyond their skills, and Hasheem Thabeet and Yi Jianlian are extremely missing something in his feel that goes beyond their skills or lack thereof

[CrackersPhinn]

First, their general strategy seemed to be to trade away draft picks, recognizing that the bottom twenty slots aren't going to produce much value. (Implicitly, I suppose that this means that they didn't think their special analytic sauce could be applied to the draft. Perhaps someone here has some expertise on the Mavs philosophy on drafting).

I'm not sure you've quite landed on their draft philosophy. I can't help but notice they chose not to follow the strategy you've outlined in the most recent draft, instead bringing 3 rookies into the fold. Mark Cuban has made many comments about the severity of the "repeat offender" tax mandated by the new CBA and how even his team will have to change their habits now. I think these things are connected.

The best teams usually have high payrolls, and so it is important to use the late 1st round as a source of cheap talent to fill out a roster. Of course we all know the bust probability of late 1sts, which means there is significant risk involved. Much better to simply buy a proven commodity if you have the resources to do so. Before the new CBA Cuban did, but this is not the case anymore.

Teams often use the latter part of round 1 to swing for the fences and take some seriously flawed prospect with "high upside". They figure, "Hey, we're already good, we don't need help immediately. And imagine if he pans out!" This basically never works. But teams, even smart ones (look at OKC and Chicago in the 2012 draft), do this all the time. And it is a mistake. Miami traded their pick this year, and I felt that was a very savvy move. More good teams should do that if they think they won't need the rookies. Right now there are bad teams run by smart guys who can make clever deals to siphon assets from better, more desperate teams. We have seen this a lot lately. The good teams can return the favor if they exploit the bad teams' zeal for youth and trade supernumerary picks for futures ones and additional assets. But even if you don't trade the pick, I think Dallas' 3 picks this year show what they think they can extract from the late draft: defensive talent. Which is why they picked 2 steal machines and a shot blocker.

[CrackersPhinn]

wilq basically nails it about the difficulty of assigning credit for the kind of group decision making you see in front offices. There is also a related issue here, which is that analytics only really promises that you will know more with it than without it. If, for whatever reason, you choose not to act on this information then you can't really say that analytics failed you, even if that's how it appears to an outsider.

Take the Phoenix Suns for example. Let's say after the '06 season the Phoenix brass began to suspect that Amare was not the superstar he was billed as given their success without him. And let's also say they had plenty of statistical evidence to back up this intuition. That means they were burning resources by paying so much to Amare, and that since he was still broadly perceived as a superstar it would make a lot of sense to trade him. They could have probably recouped most of the wins they lost with his departure while also cutting costs. Maybe to the point that Phoenix would not have had to sell off it's picks every year.

Would the Phoenix GM have ever made a move like that, even if the numbers told him it was the right one? Not a chance. He has a career to look after, and only a title could vindicate trading a player like Amare. That's a very risky move as winning a title is really damn hard.

I just saw the movie "Moneyball", and what I found most amazing was Billy Beane's willingness to make himself virtually unemployable in the eyes of MLB. He had just come off a 102 win season and lost his best 3 players. Everyone would have understood a lousy follow-up season, so Beane's back was hardly up against the wall. But he took on enormous risk anyways. Most people would have acted like Art Howe, knowing that since convention is communal property it's failures can't be attributed to you and in fact you'll be credited because your employer will feel he always knows what he's getting with you.

I guess what I'm saying is that even with the Mavs, an org seemingly fully supportive of analytics from the top down, there are still PR issues involved with going the full Moneyball route.

[mystic]

According to Jeremias, between 2002 and 2011, Dirk when playing provided +5.6 points per 100 possessions, and he played 75% of the minutes on average.

Just want to point out that those longterm values aren't pretty good, because without a proper aging/development curve, you will basically have two overlapping waves with amplified or reduced values. Look at Steve Nash and Dirk Nowitzki for example. In the 10yr ranking Nash is ahead of Nowitzki while he is below Nowitzki, if you take the average yearly values. Nash was basically a +/-0 player from 2002 to 2004, while in the 10yr calculation he is now assumed to have +5.7 in average for that same timespan.

Now, I don't know if Daniel's formula is exactly fair for the counterfactual at hand. What we need to know is the price for talent that is available on the market. As such, the RAPM/salary relationship should properly take into account the realities of the CBA, which puts a ceiling on payment to rookies and stars. I don't know if taking these effects into account would fully account for this "overpayment", but it is an important caveat.

The NBA has not just a CBA with salary restriction for rookies or top-level stars, but also things like the Draft, Luxury Tax, shared merchandising, etc. pp. We are not talking about a free market here, but a cartel of 30 teams trying to make an overall profit. That is important and makes basically all tries to assess the value of analytics by looking at the salary and the success rate useless. Those 30 teams do not have the same ressources or same opportunities, there are restricted to operate within a certain city for example. Take the Knicks, with their own TV broadcast having basically unlimited ressources and thus can easily pay more than other teams for the players. AND they are supposed to do that in order to share the revenue among the teams via luxury tax. Being a bigger market might also be more attractive for players. Given the salary restrictions a smaller market team might just need to pay more for a specific player in order to snatch them away from a bigger market.

And if it didn't pay off for the Mavs - over a ten year period - who began when the low branches were laden with the largest, ripest fruit and for whom there was the maximum institutional support for thorough implementation of the program, what rational belief is there that it will pay off for others?

How do you know that this didn't pay off for the Mavericks? They built 3 completely different contenders, they just didn't have the necessary luck in order to end up with 3 championships. In 2003 Nowitzki suffers a knee injury during the WCF against the Spurs. If the Mavericks advance further, they very likely kill the Nets. In 2006 a better judgement by one referee (if they make the correct call against Wade (offensive foul and palming violation), and the Mavericks go back home being 3-2 ahead against the Heat.
The games are also decided simply by variance (luck or whatever you want to call it), making a judgement whether something is paying off or not based simply on championships is wrong.

[schtevie]

According to Jeremias, between 2002 and 2011, Dirk when playing provided +5.6 points per 100 possessions, and he played 75% of the minutes on average.

Just want to point out that those longterm values aren't pretty good, because without a proper aging/development curve, you will basically have two overlapping waves with amplified or reduced values. Look at Steve Nash and Dirk Nowitzki for example. In the 10yr ranking Nash is ahead of Nowitzki while he is below Nowitzki, if you take the average yearly values. Nash was basically a +/-0 player from 2002 to 2004, while in the 10yr calculation he is now assumed to have +5.7 in average for that same timespan.

I do believe that your objection to my argument falls a bit flat in this instance. I am not saying that this 10 year RAPM average is the best possible 10 year estimate of Dirk Nowitzki's value. To the extent that it is biased, however, I expect that it is biased downwards.

First, regarding aging effects, the years in question comprise a Dirk aged 23 through 32: in expectation, an upside approximately equalled by the down. So, this factor isn't highly relevant, if this intuition is correct. What does matter is that the series starts with DN presumed to have a value of 0 (I believe that's the initial prior). So, chances are the early years of the average are biased downward. Then there is the general bias of RAPM against elite performance. So, consider the +5.6 to be a lower bound.

And as for Steve Nash's numbers, what does that have to do with the matter?

Now, I don't know if Daniel's formula is exactly fair for the counterfactual at hand. What we need to know is the price for talent that is available on the market. As such, the RAPM/salary relationship should properly take into account the realities of the CBA, which puts a ceiling on payment to rookies and stars. I don't know if taking these effects into account would fully account for this "overpayment", but it is an important caveat.

The NBA has not just a CBA with salary restriction for rookies or top-level stars, but also things like the Draft, Luxury Tax, shared merchandising, etc. pp. We are not talking about a free market here, but a cartel of 30 teams trying to make an overall profit. That is important and makes basically all tries to assess the value of analytics by looking at the salary and the success rate useless. Those 30 teams do not have the same ressources or same opportunities, there are restricted to operate within a certain city for example. Take the Knicks, with their own TV broadcast having basically unlimited ressources and thus can easily pay more than other teams for the players. AND they are supposed to do that in order to share the revenue among the teams via luxury tax. Being a bigger market might also be more attractive for players. Given the salary restrictions a smaller market team might just need to pay more for a specific player in order to snatch them away from a bigger market.

I do not understand the relevance of your remarks for the point at hand. I am not claiming that any CBA has ever represented a free market for labor. The issue at hand is far more specific: did the Mavericks spend their salary dollars "efficiently"? Toward that end, one needs to establish a dollar/(R)APM tradeoff for contracts that would have been feasible over the time period in question. Throwing max. and rookie contracts into the mix, without any sort of adjustment, should bias that ratio downwards. (And additional account should be taken of veteran's minimum contracts, but that is probably less important.) I am trying to be fair and was simply making the point.

And if it didn't pay off for the Mavs - over a ten year period - who began when the low branches were laden with the largest, ripest fruit and for whom there was the maximum institutional support for thorough implementation of the program, what rational belief is there that it will pay off for others?

How do you know that this didn't pay off for the Mavericks? They built 3 completely different contenders, they just didn't have the necessary luck in order to end up with 3 championships. In 2003 Nowitzki suffers a knee injury during the WCF against the Spurs. If the Mavericks advance further, they very likely kill the Nets. In 2006 a better judgement by one referee (if they make the correct call against Wade (offensive foul and palming violation), and the Mavericks go back home being 3-2 ahead against the Heat.
The games are also decided simply by variance (luck or whatever you want to call it), making a judgement whether something is paying off or not based simply on championships is wrong.

And here, I don't know what to say. I "know" their efforts didn't pay off because they apparently spent a lot more money each year than they need have spent. I have not denigrated the Mavericks' accomplishments. The issue is the degree to which is was gained by being more clever or having spent more money than the competition. And I certainly didn't make a fetish of championships.

But the issue of variance and luck is an important point to emphasize: one should not expect that over a 10 year span that this should be relevant in terms of overall winning percentage. And it is here (again, pending a better formula for understanding salary) that the evidence presented suggests that there wasn't much (overall) return to the Mavericks in terms of the contribution of "analytics". Though - to be clear - I'd be quite happy to be proved wrong. I would like to think that there was at least one golden age of analytics.

[mystic]

I do believe that your objection to my argument falls a bit flat in this instance. I am not saying that this 10 year RAPM average is the best possible 10 year estimate of Dirk Nowitzki's value.

My only intention was to point out the issues with longterm studies without proper aging/development adjustment. Nothing more. It had nothing really to do with your argumentation.

I do not understand the relevance of your remarks for the point at hand. I am not claiming that any CBA has ever represented a free market for labor. The issue at hand is far more specific: did the Mavericks spend their salary dollars "efficiently"?

The Mavericks COULDN'T spend their salary more efficiently, because of the restrictions. It is pretty simple, once a team is over the cap, the only way to improve is by either improving the signed players or make trades. In order to improve a team must give up something which they don't need to get something which they want. That goes for both teams involved in a trade. The Mavericks wanted talent, other teams wanted salary relief. That's what the Mavericks did. In order to improve and get a better chance to win a championship, they had to take on higher salary players. They couldn't do anything else, because they had no other assets.
Now we can assess each of the trades in order to see whether they made the absolute best decision, but in essence just the fact that the restrictions forced them to take on more salary makes your whole argumentation pretty pointless. The only way to use the overall salary as an estimator for "good work" would be a free market in which each player can be signed each year to a new contract. Otherwise you have to account for multiple rules and regulations that will throw off everything you want to see. The question in place would be: Could another team get the same players for less money?

I am trying to be fair and was simply making the point.

I understand that, I'm just pointing out that the underlying system is introducing problems which can't be ignored.

And here, I don't know what to say. I "know" their efforts didn't pay off because they apparently spent a lot more money each year than they need have spent.

Sorry, I misread a comment which made me believe that you wanted to say additionally to your statement about the salary that the 1 championship would be a sign of failure here.

But the issue of variance and luck is an important point to emphasize: one should not expect that over a 10 year span that this should be relevant in terms of overall winning percentage.

In terms of regular season success? It may be more relevant as you think. Injuries to players at the "wrong" time or just simple variance can throw the results off, otherwise we would see a 100% correlation between win% and MOV over the time for example, which we don't see. The larger sample makes the issue smaller, that is true, but hardly completely irrelevant.
For playoff success your statement is really, really off. You should always expect that variance is playing a role, no matter what.

I would like to think that there was at least one golden age of analytics.

As I said, your assessment via salary is the problem here. You haven't showed us anything which would lead to the conclusion that the analytic job of the Mavericks didn't pay off. Or to make a slightly different point: Do you think the Mavericks analytic department made a mistake, because they thought Odom would be helpful?

[schtevie]

Mystic, let's be very clear:

(1) I am gratified that the "shortcomings" of the 10 year RAPM are no longer considered an issue for the matter at hand. And the 5.6 RAPM of DN might even be considered a lower bound, no?

(2) It is certainly not the case that the Mavericks couldn't have spent their money more efficiently. Are you saying that, ex post, Mark Cuban believes every transaction he has ever made have been the best possible? Something tells me he might not have been willing to let Steve Nash go, were he to have known the future. (How much did he spend the following year for point guard play?) The issue is simple and plain: there is an available market for talent, and one must be judged accordingly. Again, again, again, the issue is what were the true trade-offs available? If they are/were, in fact, represented by the formula Daniel offers, there is demonstrable underperformance. But there are (listed) reasons to believe that available talent was to be more expensively gained.

(3) None of my remarks in this string invoked championships as a measure of success.

(4) None of my remarks in this string invoked playoff performance as a measure of success.

(5) If you have any evidence that the regular season success of the Mavericks over the 10 year period in question was disproportionately (negatively) influenced by "extra" and unanticipated injuries, then say so. I don't think you will provide such evidence.

(6) Regarding your concluding comment, what I believe I have shown is that if the salary relation that Daniel provided is approximately correct (one that I am not leaning hard on) the Mavericks, given the inheritance of Dirk Nowitzki (and drafting of Josh Howard), on average, over the 2002-2011 seasons, spent much more money than necessary to achieve the level of regular season success that they actually achieved.

That you introduce the Odom transaction as a Parthian shot, suggests to me that you are not addressing the argument seriously, first because this didn't occur within the 2002-2011 timeframe in question and second because it betrays an unwillingness to engage. Any given transaction is a hit or miss; you make your best guess given the information at hand, then realize the consequences. Acquiring Odom may or may not have been rational, ex ante. Just as letting Nash go may have been. The point, again, again, again, is that on average, over ten years, one should expect a positive return to analytics.

Where is the evidence that this exists?

[mystic]

1. As I said, I just wanted to point the shortcomings out, also, because it influences the calculation. But overall it doesn't matter that much, because the whole argumenation is based on a flawed concept.

2. First, the Mavericks offered Nash the same amount of money over the same amount of years, with just the difference that the last year was only half guaranteed while it would have become fully guaranteed, if Nash had played 1200 min in the second to last season of the contract. Nash choose to leave, the Mavericks did NOT want him to leave, they were just not willing to overpay for Nash.

Second, you are making the mistake to take your hindsight knowledge of the events to assume how good the decision was at the time. It makes no sense to do such thing, because at that time, everyone was saying that the Suns overpaid to get Nash, an aging player who showed clear lack of endurance coming playoff time. Nash couldn't keep his performance level in the playoffs 4 years in a row, it was not to expected that this will change when Nash becomes older. The Odom-Example should have showed you that point. Nobody was arguing that Odom for a TPE and a pick was some sort of bad decision back in December 2011, but now we know it was. Circumstances are influencing the results A LOT. Nash is an outlier, there is no data suggesting that a player will improve that drastical at the age of 30.

Third, you don't understand the market in the NBA at all. Your whole idea would only make sense in a free market situation. Just the simple fact that a rookie can perform above average while being much cheaper, throws the thing completely off. Players will not just be paid for future expectations, but also for their past performances. The system is just set to be like that. So, for an average player being on a rookie scale contract a team has to pay much less than for an average player being a free agent. And the selection of the rookies is based on a draft. There is simple no free choice for the teams to get certain players. Be assured, the Mavericks would have selected different players, if they had higher draft picks. But that also means that the Mavericks would have been performed worse during the regular season.

Fourth, the simple reason I said that the Mavericks couldn't get the players cheaper is in fact the they likely couldn't. That doesn't mean that every decision was perfect, as I pointed out we would need to evaluate every made trade in order to come up with such conclusion anyway. So, no idea, how you can read that and can think I meant every decision was perfect. What I meant is that they couldn't get those players for cheaper salaries, because the only way the Mavericks could have gotten those players was via trades. The Mavericks done that, because they valued the talent more than the salary. Your whole idea doesn't take those restrictions into account and also falsly assume that a different set of players for less money could have achieved the same. The matter of fact is that the Mavericks had the 2nd highest Win% during that timespan, we can easily assume that they did something correct. The issue at hand is, whether they could have achieved the same result with a different set of players while spending less money. I say that the fact that the NBA is a cartel with specific rules and regulations does make the assumption false, no matter what. But this does not mean that I think the Mavericks did everything right.

Fifth, the Mavericks probably don't trade for Terry, if he is on a longterm contract. He was 27 with two years left on his contract. The length of the contracts is influencing the trade value of the players as well. To get Terry, they had to take on Alan Henderson as well to make the salaries match. They wanted Terry, but had to take on a bad contract in order to get him. Was there a player available who could have done the same as Terry did for the Mavericks for a lower price?

3. Acknowledged.

4. Acknowledged.

5. Well, Shawn Bradley's injury for example. Raef LaFrentz, Keith Van Horn, Erick Dampier, Finley, Stackhouse missing games. Look at 2005 for example, don't you think having a fully healthy Dampier, Finley, Stackhouse and Howard would have net a better result? If some of the players are not injured, the Mavericks are not making certain trades while aquiring a higher salary by those trades. If you want to stay on top, you have to pay for that, or you need an incredible amount of luck with rookies and injuries. Look at OKC for example, they have a low salary, because the rookies panned out while they had basically no major injuries to important players. Do you think any team could get the services of Harden+Ibaka for just $8m combined next season on a free market?
In 2011 Butler missed 53 games, Nowitzki 9, they went 21-5 in games having Butler and Nowitzki starting. That suggests 66 wins for a 82 games season. Butler was with $10.6m on the payroll, but the Mavericks also got money from the insurance company to cover a big part of the costs. Your calculation doesn't account for the real costs, thus you get a misleading picture. Take the Blazers as an example, in the upcoming season they have a $58m payroll, while in reality they also have to pay $16m to Brandon Roy, money which is not covered by the insurance. Overall, the Mavericks paid in 2011 about the same amount of money (in real money, not on the payroll) for their roster as the Blazers will pay next season excluding luxury tax.
We would need to look at the average missed time for certain quality players and compare that to the Mavericks situation.
What you would need is the REAL costs for each team and then account for circumstances introduced by the CBA in order to come up with a conclusion whether analytics help or don't.

6. As I said, every way of calculating that will have no chance to account for the CBA problems. A team with capspace had different problems and opportunities than a team over the cap. The whole situation in the NBA does not allow such conclusion based just on salary. You would need to show that a different set of players 1. was available for a lower price and 2. that these player sets would have achieved the same. You did neither of that.

7. The Odom situation was simply brought up to show you how the hindsight bias is influencing your analysis, without accounting for all the issues we have seen like injuries or mental issues.

8. You are not asking a scientifically useful question, because you want to have proof here. In science we would also ask the opposite question, making the hypothesis that it helps and now disprove that. You did not do that at all and while using a tool which is not appropiate to answer either of those questions.

[schtevie]

Hooboy...

Where are we in this? We have tentative agreement that Dirk Nowitzki was worth at least +5.6 for the Mavs over the 10 year period in question. If his true value was just a little bit higher (never mind the contribution of Josh Howard) this would leave almost nothing to explain, in terms of my argument. But never mind this, we move on.

You are diverted by a few points that shouldn't trouble you.

(1) Regarding Steve Nash, go back and read. I dismissed his departure from the first, precisely because I wanted to avoid "hindsight bias". His is but one of all the Mavs transactions over the years that influenced (counterfactually and factually) their performance.

(2) Regarding what is possible CBA-wise, in terms of acquiring players post-draft, there seems to be confusion. I will try to make the general point again, but I am not sure that there is a willingness to understand it. My point is simply this: if you look at the correlation of metric X to salaries (what I understand that Daniel did), you get a result that isn't relevant for determining, ex post, what was the range of potential transactions. Structurally undervalued players (those with their contracts having imposed price ceilings) weren't readily available for Mark Cuban to spend his coin on. Hence, my point was that this salary relationship would need to be addressed before a more definitive conclusion could be reached. That said, for there to have been a positive return to analysis, as indicated on the salary dimension, the Mavs would have to have spent relatively less for talent, not more. And my guess is that the data can not be manipulated to deliver this result, but it remains an open question.

(3) Regarding injuries, here is a relevant statistic you did not provide. Over the 10 year period in question (again, please don't forget, the whole point of this exercise is to look at long-term performance, as this is the only way to get rid of the noise) the Mavs' top 3 players, as determined by MPG, missed on average, 5.53 games per season. Not such an injury-riddled team. Now, it might be the case that a more robust analysis, digging deeper in the line-up (of what are less valuable players, however) will show the opposite. But it just might be the case that the bias cuts the other way: that the Mavs were healthier than the average team, on average.

(4) I have no idea how you come to the conclusion that the approach taken and the tentative conclusion reached involves and is tainted by "hindsight bias". Except for "assume Dirk Nowitzki", which is the up front stipulation and a completely reasonable one, there is none. The 10 year period in question was determined by Jeremias' rating, what is a fair representation of the Mavs +50 game winning run. And the salary data is what was. End of story.

[mystic]

That said, for there to have been a positive return to analysis, as indicated on the salary dimension, the Mavs would have to have spent relatively less for talent, not more.

That is your WRONG assumption. This hypothesis can ONLY be true, if there is a free market without any kind of restrictions, otherwise the salary to success rate tells you nothing at all without adjusting for all the implemented restriction. Matter of fact is that the Mavericks could not have aquired the players differently, because those players were under contract with other teams. You are basically assuming a linear relationship between success and salary, while there is no such thing. The rookie scale and the max salary alone will destroy that assumption. A player like LeBron James would costs MUCH more than his current salary on a free market, and a player like Harden or Ibaka wouldn't be available for about $8m next season. Those are things which completely destroy the relationship between salary and success rate.
Also, the cheaper talent must be willing to sign with the respective franchise for that price. Take Tyson Chandler for example, who wanted MORE money from the Mavericks than he agreed to sign for with the Knicks. The prices for players are distorted, thus, rendering your hypothesis that the advantage of the analytics must be seen in a salary advantage useless.
And all the other stuff is basically related to that point. If you don't get that the NBA cartel is governed by rules which are supposed to make it harder for higher spending and better teams to get better players for a cheaper price, it really makes no sense to discuss the rest of your post further.

Regarding the 2nd point, the relation between salary and metric is only useful for determining the supposed market value of a free agent. That does NOT help you in order to get a trade value for a specific player or tells you anything about the availability of needed players. You can argue that a player set X should just cost a certain amount of money when they just give you a certain WIN%, but that doesn't tell you anything whether those players are AVAILABLE for that price. When your analytical department tells you, that player x, y and z would be helpful for the team, the GM might answer that they are under contract and thus the team would need to give up assets in order to get those players. It is a question of opportunities and assets, and the assets of a better team with no capspace are limited by the CBA. If you give up talent in order to get talent, you might as well stay put and do nothing. The Mavericks aquired talent by taking on unwanted players additionally, adding to their payroll, but overall it still increased their chances to win a championship.
There is also a difference in prices for the size of the players. For an average center the price is higher than for an average point guard, just because there are less 7' people than 6'2'' people. When the Mavericks signed Erick Dampier to that contract, they not just signed for a slightly above average player here, they signed a center and those are less often available.

Regarding the injury problematic you completely missed the point. Whether the Mavericks top3 minutes getters were more or less injured is not the issue at hand, the question is how EXPANSIVE those injured players are. Keith Van Horn was with $15.5m on their payroll in 2006. Thus, he represents a more costly injury. Teams are effected differently by injuries, because the importance of a player is also determined by depth and the possibility to replace the injured player. In 2011 the Mavericks got lucky by getting Stojakovic for the minimum when they needed a player to fill in for some 15 mpg on SF on the other hand.

I'm not arguing that the Mavericks got always the worse end of the equation or made always the right decision, I'm arguing that all those rules, regulations, injuries and variance is throwing off your hypothesis that the value of analytics must be seen in a better salary to success rate. Can you understand that point?

[schtevie]

I think that you have epistemological problems with the approach taken. To the extent I understand your argument, it is that it is simply impossible to infer anything about the rationality of actual decisions from contemporaneous salary data. I think that is completely incorrect, so we shall agree to disagree.

Regarding the related point about injuries, pretty much same again. I would welcome a more comprehensive analysis of the totality of the Mavs injuries and the extent to which it biases the notional results, but I'm not going to do it, and referring to KvH in one year (who, perhaps, may have been overpaid?) - a single datum over a 10 year span - is not persuasive.

Finally, to needlessly reiterate, no, I do not understand the point that "all those rules, regulations, injuries and variance is throwing off your hypothesis that the value of analytics must be seen in a better salary to success rate." "Analytics" is an input into the win production function. If there is a positive return to analytics, it must show up in terms of a greater number of wins, all else equal. The Mavs did get more wins, but they spent a lot more money. This must be taken into account, and the preliminary results suggest that the return to analytics was not positive.

[mystic]

I think that you have epistemological problems with the approach taken.

If it makes it easier for you, you may want to look at this that way. But matter of fact is that you are making a hypothesis, that the advantage of the analytics has to be seen in a better salary to success rate, while you don't show at any point that this is the case for a non-free market with a lot of different rules which are specifially there to make it tougher and more expansive for better teams.

I think that is completely incorrect, so we shall agree to disagree.

We can, but that doesn't remove the CBA nor does it change the NBA environment.

Regarding the related point about injuries, pretty much same again. I would welcome a more comprehensive analysis of the totality of the Mavs injuries and the extent to which it biases the notional results, but I'm not going to do it, and referring to KvH in one year (who, perhaps, may have been overpaid?) - a single datum over a 10 year span - is not persuasive.

Seriously, at what point do you understand, that injuries are not affecting the costs for all teams the same way? I'm just used the Mavericks here as an example (and Van Horn specifically) to make that point. If all injured players would have the same costs to their respective teams, you can ignore that, but that is NOT the case. The injuries are effecting the costs-to-wins ratio significantly more for teams with more expansive players getting injured than less costly players on MANY different stages. The Mavericks did not sign uninsured players to long contracts like the Knicks done that with Amar'e Stoudemire for example, that is a difference which might be related to the Mavericks analytic department. Not taking those things into account will again throw off your conclusion.

If there is a positive return to analytics, it must show up in terms of a greater number of wins, all else equal. The Mavs did get more wins, but they spent a lot more money. This must be taken into account, and the preliminary results suggest that the return to analytics was not positive.

I understand your point, but you seem to fail to see WHY the Mavericks paid more. Again, the CBA is designed to make those better teams pay more and get less chances at better rookies. That is the whole purpose of that thing. Just the simple fact that the NBA is a cartel with specific rules which are favoring worse and less spending teams to improve while making it tougher for better and more spending teams, makes your whole hypothesis wrong. The matter of fact is that there is no high linear correlation between salary and success rate from 2002 to 2011, you get a better fit with a non-linear regression. And that is something we can expect based on the CBA.

[Bobbofitos]

Stevie, yes or no: The CBA is structured in such a way that your straight dollars-to-wins approach is not the most appropriate measuring stick.

Mystic says no. (I say no as well) If you say yes, all the other arguments are kinda irrelevant, since this is a pretty big deal.