Assigning Blame in Phoenix

Posted on February 5, 2009 by


The Suns are setting.  Across the past four seasons the Phoenix Suns have averaged 58 regular season wins a year.  After 47 games in 2008-09 the team has only won 26 times.  This puts the Sun on pace to only win 45 games.  All NBA observers can see this record and I think all can agree that Phoenix is not what it was last year.  The question, though, is why.

Avoiding Blame

When an organization declines outside the world of sports it’s easy for everyone to make excuses. The workers blame management.  Management blames workers.  And both might blame the economy.  The lack of data on performance, though, allows everyone to deny responsibility.

It’s a different story in sports.  At least, it can be if we pay attention to performance.

Currently the Phoenix Suns are acting like performance data, though, doesn’t exist.  Explanations for this team’s decline tend to focus on team chemistry.  Although such an explanation may be valid (probably not, but we can pretend), it avoids assigning responsibility to any specific individual.  Such an approach certainly avoids making any individual unhappy, but it likely prevents Phoenix from finding a solution.  After all, how does one fix “chemistry”?  And when does one know it’s fixed?

Assigning Blame

Fortunately for the Suns, performance data does exist in basketball.  And that data clearly points the finger of blame at one specific individual.

Table One: The Phoenix Suns after 47 games in 2008-09

Table One reports two forecasts of the Phoenix Suns.  The first looks at the number of wins this team could expect given what their players did last year.  The second looks at the number of wins given what the players are doing this year.  As one can see, the difference in these two perspectives is about 17 wins.  And more than ten of these lost victories can be tied to changes in the play of Amare Stoudemire.  In other words, if Stoudemire was performing as he did last year, the Suns would be on pace to win 55 games.   Such a mark would be fall short of where the Lakers are, but certainly could be good enough to challenge for second in the conference (and with the injury to Andrew Bynum, second might be good place to be in the West).

Stoudemire, though, is not performing as he did last year.  Last year he posted a 0.291 WP48 [Wins Produced per 48 minute].  Of the 129 players who played at least 2,000 minutes in 2007-08, Stoudemire ranked 12th in WP48.  So Stoudemire was an elite player.

This year his WP48 is only 0.123.  This mark would have ranked 73rd last year.  Again, the population of players with more than 2,000 minutes is 129.  So Stoudemire’s performance this year is below average (average in that sample is 0.142).  At least, below average for the players who get the most minutes.

Athletes are taught that teams wins and lose together.  Such a focus avoids the assignment of blame and treats all players as equally important.  But clearly this is a myth.   Some players are just much more productive than others.

Last season Stoudemire was one of those players who was “much more productive than others.”  And consequently, he deserved a great deal of credit for the team’s success. This year, though, he’s not offering the same level of production.  Consequently, he deserves much of the blame.

Explaining the Blame or Blaming the Vain

People have argued that part of the problem in Phoenix has been the change in coaches.  The new coach, Terry Porter, has emphasized defense (and therefore, de-emphasized offense).  Certainly both Steve Nash and Stoudemire are offering less under Porter.  But looking back at Table One we see the other players on the team are offering about the same.  In other words, I am not convinced this is coaching. 

No, I think this is mostly about Stoudemire.  And when we look at Table Two, we can see where Stoudemire has declined.

Table Two: Evaluating Amare Stoudemire

A few days ago I noted the specific issues with Stoudemire.  First of all – as Table Two notes — his scoring has declined.   This decline is caused by drop-offs in both shooting efficiency and shot attempts.  Certainly the decline in scoring is part of the problem.  Apparently, though, another big part of the problem is how Stoudemire has reacted to his declining point totals.  The story people tell is that Stoudemire is motivated by his point production.  When he gets as many chances to score as he likes, he’s motivated to rebound, blocked shots, and avoid turnovers.  When he doesn’t score as much as he likes, these other aspects of the game get ignored.

In sum, people are arguing the vanity of Stoudemire is causing him to choose to offer less.  And this choice is a big reason why Phoenix is not contending.

As a consequence, it’s now reported (by Adrian Wojnarowski of Yahoo!Sports) that Phoenix is contemplating sending Stoudemire to another destination.  If Stoudemire was offering what he did last year, such a trade would seem like a bad idea.  After all, it’s difficult to find a player who can rank in the top 15 in the league.  But if the Stoudemire of last year is lost to Phoenix – and that looks to be the case – a trade might be a very good idea.  Remember, Stoudemire is still an above average scorer.  So he should be able to fetch more production than he’s offering the Suns right now. 

Although such a trade may not be welcomed by fans of this team, it may be the best move for this franchise to make.  After all, once you we have assigned blame for an outcome, the next step is to impose consequences.  And when a player puts his scoring ahead of winning, it’s probably time for the team to move on.

– DJ

The WoW Journal Comments Policy

Our research on the NBA was summarized HERE.

The Technical Notes at 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

Wins Produced vs. Win Score

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.