Basketball Prospectus: Unfiltered Everything Else is Fluff.

December 9, 2009

Steve Nash and Quantifying the Subjective

Filed under: Uncategorized — Kevin Pelton @ 3:10 pm

Today is Steve Nash Internet Day, with the Bright Side of the Sun blog inviting writers around the Web to share why the Phoenix Suns’ point guard is so special.

There is a tendency in the statistical community, especially in baseball, for analysts to take glee in being able to disprove the sport’s received knowledge. As an APBRmetrician, I’ve never really taken that approach. Instead, one of the most exciting things for me is being able to do the opposite–use the numbers to quantify something that has been merely suspected in the past.

One such notion is the concept of a player “making his teammates better.” Conventional wisdom has it that this effect is real and important, and no player has benefited more from it than Steve Nash, who won back-to-back MVPs in large part thanks to his supposed ability to improve his fellow players.

Certainly, there are statistically-inclined skeptics. John Hollinger devoted an essay in the first edition of his version of the Pro Basketball Prospectus series to showing that Jason Kidd, Michael Jordan and John Stockton did not make their teammates better. In The Wages of Wins, David Berri too argues against the concept.

With due respect to their position, I disagree with Berri, Hollinger and the other doubters. One of the fortunate aspects of the advances the statistical community has made in the past five years or so is that we now have more of the data we need to show how Nash in particular gets more out of his teammates.

During the 2005-06 season, I first took a look at this issue for by comparing the offensive performance of Nash’s teammates with him on the floor to when he was on the bench. While the effect was not uniform–Shawn Marion, whose drop-off since being traded by the Suns has been attributed in part to being separated from Nash, actually scored slightly better with the MVP on the bench–in general Nash’s teammates shot the ball better and committed far fewer turnovers when playing alongside him, an effect that swamped their tendency to get to the free throw line somewhat less frequently.

What that study lacked was a way to measure Nash alongside his peers. It was possible that Nash’s passing did in fact make his teammates better, but that this was common for point guards. Now, thanks to the regularized adjusted plus-minus offered by, we can make this comparison. In addition to the standard overall, offensive and defensive plus-minus numbers we’ve seen in the past, offers breakdowns of how players impacted their team’s performance in each of Dean Oliver‘s Four Factors using the 2006-07 through 2008-09 seasons. When it comes to shooting (as measured by effective field-goal percentage), the leader by a mile is Nash.

Player            eFG%  StdErr

Steve Nash       +2.96   0.40
Ray Allen        +1.57   0.39
Rashard Lewis    +1.52   0.38
Kobe Bryant      +1.37   0.40
LeBron James     +1.37   0.39

What this stat means is that, adjusted for teammates and opponents, Nash’s teams shoot the equivalent nearly three percent better from the field when he is on the floor. What is remarkable about this is how much better Nash rates than anyone else in the league. His effect on shooting is nearly twice that of second-place Ray Allen, and Allen and Rashard Lewis are the lone players in the league who have even half as positive an effect on their team’s shooting as Nash. (The difference between them also dwarfs the uncertainty in the statistic, as measured above by its Standard Error.)

Naturally, part of this is in fact Nash’s own shooting, which is phenomenal. However, the ratings of other players seem to put an upper bound on how much of a factor it could be. Is Nash’s shooting that much more valuable than Allen’s or LeBron James‘? His advantage, then, seems to capture how Nash truly does make his teammates better shooters with his passing. Let’s say that we can measure this by the difference between Nash and Allen, which is about 1.4 percent. Given that Nash is on the floor for a little over 50 Phoenix shot attempts a night, his ability to make teammates better is worth nearly 1.6 points per game. Over the course of a full season, that’s a little bit more than four wins we can credit to Nash’s ability to make his teammates shoot better.

While other adjusted plus-minus metrics have not been as positive on Nash, RAPM rates him the best offensive player in the league from 2006-07 through 2008-09 and third overall behind Kevin Garnett and James despite a poor rating at the defensive end of the floor. This certainly doesn’t come as a surprise to anyone who has seen Nash play or knows his history of leading elite offenses–adjusted for league, the five best post-merger offenses all featured Nash at the helm–but sometimes it’s nice to be able to quantify what we already know to be true.

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress