Every once awhile a sports fan might notice that there is a world outside the games we watch and play. And when we look we quickly notice one important difference between the “real world” and the wide world of sports. In the “real world” it’s hard to figure out who is responsible for the outcomes we observe. For example, who should we blame for the government’s slow response to Katrina? Every person asked claimed it was someone else’s responsibility. Or consider the problems in the U.S. automobile industry? Again, ask anyone and you are sure to learn it was someone else’s fault.
In sports, though, a person can’t so easily hide. We know which team won and lost each contest. And player statistics allow us to move from these team outcomes to the individual players. As has been noted before, player statistics are specifically tracked so teams can assign responsibility (or blame, if you will) for outcomes to the individuals.
To illustrate, consider the Cleveland Cavaliers, a team that has a 4-3 record after seven games in 2009-10. Before the season started Cleveland was considered a serious contender for the 2010 NBA title. And if the Cavaliers finish with four wins in their final seven games, that dream can come true. Such a mark at the start of the regular season, though, suggests trouble.
Cleveland’s efficiency differential (offensive efficiency minus defensive efficiency) of 4.9 is consistent with a team that will win about 53 games. This is hardly the record of championship contender. And when we look at the individual players we can see who is responsible for Cleveland’s struggles.
Let’s start this exercise with how Cleveland’s players performed last year:
First String
PG: Maurice Williams [7.1Wins Produced, 0.119 WP48]
SG: Delonte West [7.3 Wins Produced, 0.163 WP48]
SF: LeBron James [27.8 Wins Produced, 0.436 WP48]
PF: Anderson Varejao [8.1 Wins Produced, 0.168 WP48]
C: Shaquille O’Neal [7.9 Wins Produced, 0.167 WP48]
Second String
PG: Daniel Gibson [0.6 Wins Produced, 0.015 WP48]
SG: Anthony Parker [6.3 Wins Produced, 0.114 WP48]
SF: Jamario Moon [8.4 Wins Produced, 0.194 WP48]
PF: J.J. Hickson [1.0 Wins Produced, 0.070 WP48]
C: Zydrunas Ilgauskas [3.4 Wins Produced, 0.093 WP48]
As noted in the last post, this starting line-up has the potential to rank among best line-ups since 1981. And the bench isn’t too bad, either. At least, that’s what we would think looking at last year’s numbers.
Now let’s look at what these ten players have done this season.
First String
PG: Maurice Williams [241 minutes, 0.023 WP48]
SG: Delonte West [94 minutes, 0.022 WP48]
SF: LeBron James [265 minutes, 0.448 WP48]
PF: Anderson Varejao [227 minutes, 0.286 WP48]
C: Shaquille O’Neal [178 minutes, 0.145 WP48]
Second String
PG: Daniel Gibson [155 minutes, 0.085 WP48]
SG: Anthony Parker [230 minutes, 0.026 WP48]
SF: Jamario Moon [71 minutes, 0.061 WP48]
PF: J.J. Hickson [54 minutes, -0.283 WP48]
C: Zydrunas Ilgauskas [154 minutes,-0.020 WP48]
With numbers in hand we can now assign responsibility. Although this team is off to a less than thrilling start, it doesn’t look like we can blame LeBron James, Anderson Varejao, or Daniel Gibson. Each of these players is actually doing more than he did last season. And Shaquille O’Neal – who I think some people have blamed – is actually pretty close to what he did last season. Given his age, that might be the best Cleveland fans can hope for.
When we look at the other six players, though, we see some players who have not produced at the level we saw last year. Hickson and Ilgauskas – who were each below average last year – are now in the negative range. Williams, West, Parker, and Moon – who were each above average last year – are now below average.
It’s very important to note that three of these players have yet to play 100 minutes. So forecasting out of this sample is not something we should do. The point of this exercise is to simply identify the specific players who have started slowly. And that exercise says LeBron, Varejao, Gibson, and Shaq (at least, given his age) are doing fine. The other six players, though, need to step it up if the Cavaliers are going to truly contend this season.
Let me close by noting a problem with a record from the start of the season. Had the Cavaliers started 29-5, and then won only four of their next seven games, Cleveland fans wouldn’t notice much. The team would be 33-8 and still on pace to win 66 games. But when the team starts 4-3, it becomes hard to imagine that a 29-5 run is just around the corner. Given what these players did last year, though, it’s possible for Cleveland to start winning much more frequently. In sum, in a few months this November record might be a very distant memory. And all those players we are blaming today might once again be celebrated in Cleveland.
– DJ
The WoW Journal Comments Policy
Our research on the NBA was summarized HERE.
The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score
Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:
Simple Models of Player Performance
What Wins Produced Says and What It Does Not Say
Introducing PAWSmin — and a Defense of Box Score Statistics
Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.
Wages of Wins Journal 
Chicago Tim
November 10, 2009
Slight typo in the last paragraph: “wining” for “winning.” Freudian slip? Great article, though — everyone seems to be piling on Shaq, unfairly, I judge.
I can’t resist another comment about Bill Simmons’ book, The Book of Basketball. He finally does acknowledge wages of wins in his entry on Wes Unseld, a little more than halfway throught the book (55% to be exact — sorry, my Kindle version doesn’t tell me the page number). But he lumps it in with Hollinger’s PER, then proceeds to blast PER for undervaluing Sean Battier, even though wages of wins doesn’t have the same flaw. So I’m still not sure he’s read your articles or book, or whether he was just looking for something to lump in with PER. And he seems completely oblivious to your criticism of PER!
To be fair, his larger point is that there are other statistical categories he would like to see measured during games, categories some teams may be privately measuring but not sharing with the public. I think that’s a valid point. But in the process he slams PER and implies that wages of wins has the same problems.
dberri
November 10, 2009
Thanks. Fixed the typo.
Maybe I should have you write the review of the Bill Simmons book. Not sure when I would get to this.
Jeff
November 10, 2009
re: Bill Simmons
He references Beverly Hills 90210 and MTV reality shows. Not that ad hominem criticisms are legit, but still . . .
brgulker
November 10, 2009
Chicago Tim and Dr. Berri,
It seems to me that any critique of Win Score that doesn’t first address how Win Score is derived is simply going to fall flat on its face.
Lumping it in with PER does, in my view, demonstrate that Simmons hasn’t engaged with Dr. Berri’s written material, published or otherwise.
That doesn’t seem credible.
horsecow
November 10, 2009
I seem to recall a post about the Suns last year that made a similar argument re Shaq, i.e. that he was doing fine and Suns’ struggles were caused by a decline in Amare Stoudemire’s performance. Now Shaq is gone, the Suns are back to playing well and the Cavs are struggling. Could it be that Shaq has a unique externality -style effect whereby his performance is unaffected but the rest of the team suffers? I haven’t watched the Cavs much this year, so I can’t say if that’s the case, but one common-sense argument to explain the dip in Williams’, West’s and Parker’s performance would go like this: having two non-shooters on the court at once (O’Neal and Varejao) makes it possible for defenses to switch easily onto perimeter shooters with no cost and hurts their shooting percentages. It happened in Phoenix and it could be happening again. One reason Ilgauskas was so good last year was that he was a constant threat to make a 19-foot jump shot. He opened up the whole offense. Shaq does the opposite. Plus he’s terrible on defense. Perhaps a good WP trade, but not a good NBA trade.
TRad
November 10, 2009
Dave
I think you shouldn’t review Simmons book. What for? Your job is analysis (OK, I know you decide to waste a lot of your time on economics, teaching etc, but we are talking about serious things, not hobbies), Simmons works in entertainment. Two different fields.
Of course it could be good for your book if you provoke Simmons to reply, but from pure cognitive point of view – who cares about Simmons?
khandor
November 11, 2009
The combination of players that Cleveland has right now will NOT go 29-5 in their next 34 games this season, as the other top teams across the NBA are simply too strong.
Number crunching will only get you so far, when the mix of players isn’t “right” according to “The Laws of Basketball Acumen”.
As long as Mike Brown doggedly insists on using Z and Shaq together … the Cavs will continue to be a slightly better than middle of the pack team in the EC.
reservoirgod
November 11, 2009
I don’t know if a review of Bill Simmons’ book is necessary but I would definitely be interested in some statistical analysis of some of the stories he tells, e.g. What wouldthe Nets have looked like if they signed Kareem?
123
November 11, 2009
I think the biggest problem is the absense of Ben Wallace. What made Cleveland so good last year was 50% defense. Wallace sure was/is a better defender than Shaq.
For shits&giggles: He’s also posting the highest OReb% of his career this year
Italian Stallion
November 11, 2009
It’s very hard to prove that one player (Shaq in this case) is having a negative impact on others despite achieving satisfactory stats on his own.
Some attempt to do it with player combination stats, but I think this is an area where someone that understands “the game” very well and also has strong powers of observation might be able to see those interactions better than the numbers (at least at this stage).
There tends to be two camps in player and team analysis – number crunchers and observers.
IMHO, they both have things to offer and can strengthen the overall analysis.
The trick is to leave the math to guys like D Berri because he is clearly superior to most, but to also recognize that not everyone is equally gifted in their understanding of the game, powers of observation, and intuition. So searching out the unique talents in that area should be the goal. Then you package it together and become super analyst. :)
todd2
November 11, 2009
For what it’s worth, Shaq’s last two teams exited the playoff’s in the first round. And ORB’s are a misleading stat; the team with the most in any given game loses the majority of the time. Poor shooting leads to more ORB’s.
123
November 11, 2009
what made you post such a useless post?
Shaqs’ last team didn’t even make the playoffs and ORB != OREB%
ilikeflowers
November 11, 2009
At this point there’s a good chance that this is just some combination of the following: small sample size, Z’s age, West’s ongoing personal issues.
ilikeflowers
November 11, 2009
And had they played some team other than the Celtics, they’d likely be 5-2 now and no one would be concerned.
khandor
November 11, 2009
Who the Cavs have played thus far is of no concern to me.
Strictly from a “Basketball Acumen” perspective using Z and Shaq together is a noxious combination for Cleveland that will habitually come up short on the scoreboard when they tussle with the other top dawgs in the NBA this season.
As long as Mike Brown proceeds on this current course, the Cavs will not be making a return trip to the NBA Finals this season.
ilikeflowers
November 11, 2009
Good for you. Alas your ‘Basketball Acumen Model’ won’t be testable on this point since strictly from a WOW perspective, given Z’s and Shaq’s ages and likely declines in performance the pairing may be unproductive and detrimental to a return finals trip. Even if the pair retains their levels from last year, the Celtics and Magic might have a say in this as well. I look forward to more predictions from your strategic model.
Tommy_Grand
November 11, 2009
“One reason Ilgauskas was so good last year…”
Um, let’s not go nuts.
eek
November 11, 2009
They can win 4 of their last 7 and still not win the championship.
todd2
November 13, 2009
Thanks for the correction. Point was he hasn’t made much of an impact for his teams over the last few years. Still making a positive contribution but not playing up to hype/expectations.
khandor
November 17, 2009
The Cavs have begun to play better as they’ve shifted away from using line-ups that include Shaq and Z together. :-)
29-5 is no longer beyond the realm of possibility, since they’ve made this adjustment.