Official Racism

A friend of mine sent me a link to the following article which concerns a study done on the supposed racial biases of NBA officials. Please read it.

http://www.nytimes.com/2007/05/02/sports/basketball/02refs.html
Interesting.

What do you think?

I mulled over this for a bit and decided to offer a few words of caution to anyone who's now eager to take this study as proof of racism. It's a rebuttal of sorts.

It sounds to me like the Wolfers-Price study is pretty sound. I have no doubt that the data is accurate and their results genuine. I'm not even entirely surprised by the findings, I just don't agree that these results necessarily indicate a racial prejudice. That is, I'm gonna go out on a limb and suggest that black players get called for more fouls because black players do in fact commit more fouls. Before calling me a racist, at least hear me out. The most compelling aspect of the study to me (and why the NBA's counter-study means nothing) is that W.P. managed to adjust for potentially problematic variables - that centers are disproportionately white, veterans get more favorable calls than rookies, home v. road biases, influence of certain coaches, and finally, "player assertiveness". These variables are crucial as far as I'm concerned, and if I were to question this study at all this is exactly where I'd focus my energy.

The first question you have to ask is, the study claims to have accounted for all of these other factors, beyond race, which would (I agree) affect the way fouls are called, but what methods exactly were used to correct for these effects? In any kind of multivariable regression corrections like this are to some extent arbitrary, or at least less than exact. Since we're dealing with such a thin statistical margin those external adjustments have a major affect on the final analysis. So, how much of this is really just the numbers speaking for themselves? Another question to ask is, are the variables W.P. accounted for (even assuming corrections for these variables are accurate) enough, or are there other more subtle variables that may have been overlooked? How attuned to the pro-game does a researcher have to be to understand the particular scenario in which a bench player, for instance, might be put into a game with the specific intention of committing a foul? Or to be able to recognize a hard foul that's committed to disrupt a fast-break vs. a nickel-dimer hand check at the top of the key? Can we assume that an academic researcher understands those subtleties?

The most interesting possible oversight of the W.P. study to me revolves around the influx of white/European players into the NBA over the last 5 years (and it's unclear if European and Latin players are a distinct "race" for the purposes of this study). This influx has given rise, or occurred simultaneously to a major shift in the way basketball is played in the NBA. During the mid-90's (when this study began) the NBA game was lower scoring and by all accounts a more defensive one. Without having any data available to me, I would guess that more fouls were committed on average during these years than at any other time in the game's history. As it happens, the mid-90's also was the apex of black predomination in the NBA, all to the effect that there were more black players committing more fouls during this era. There then began a steady transformation. As the early 00's rolled on the NBA was more and more liberally opening its doors to white Europe. Concomitantly, after embarrassing losses in the Olympics and World Games, NBA teams like the Suns and Mavericks began adopting the faster, more free-flowing, and less physical style that's played in the international arena. Over the last couple of seasons this style has continued to take hold throughout the league. Also, as you know, the NBA has gone through a number of rule changes (the zone defensive most notably, and also the oversight of moving screens which a lot of tall white centers are the benefactors of) that have reduced the amount of fouls called in the NBA. And there's a few major implications here: the NBA goes from a foul-happy, grind-it-out league in the 90's that's almost entirely black, to a league that over the last 5 years is less physical, more fluid, and has seen an unprecedented proliferation of white players. Does the W.P. study acknowledge this very real and dramatic shift?

Lastly, what jumped out at me from the study was the last variable included in their calculus: "player assertiveness". The obvious: what is "player assertiveness"? How is it defined? How is it quantified? Not only is "player assertiveness" a quality that is exceedingly difficult (if even possible) to measure but also in my mind, for this study, the most significant variable of them all. To anyone being honest with himself there is a very recognizable difference between the way black players play basketball and the way white players do. There are, of course, exceptions on both sides but this study is no more concerned with exceptions than I am. The most glaring difference between white and black players is that black players are more physically imposing - stronger, faster, bigger. To me these characteristics don't necessarily translate into aggressiveness, or assertiveness (whatever the difference there), but do translate into presence, by which I mean that a stronger, faster, bigger athlete is more likely to be involved in more of the action than his slower, more diminutive counterpart. In the same way that a more athletic shortstop will attempt more groundballs, and therefore be prone to more errors, a more athletic basketball player will be involved in more defensive plays and more prone to committing fouls. I also think there is a very real argument that exists in examining where these players come from before they get to the NBA. Is Ron Artest's New York upbringing a coincidence or a predicate of his basketball persona? Does hip-hop culture's lionizing of thugishness and posturing influence the play of black players? It is no secret that the NBA (unofficially) embraces these credos and that its players mimic them with very inconspicuous gestures (look no further than Matt Barnes, as a pre-game ritual, patting down Stephen Jackson like a cop performing a weapons search on a suspected assailant). Is it such a stretch that this belligerent attitude manifests during play? Isn't it also true that white players (especially European ones) don't exhibit, at least to the same extent, the attitudes of that culture? And why wouldn't that be reflected in how "aggressively" or "assertively" a player plays defense.

Obviously there is a lot to be teased out of any analysis of this kind. The W.P. study claims to have teased out potential complicating issues to the best of their statistical ability (one that galactically exceeds mine) but I wonder just how far they went, and even if they accounted for everything, the method by which they did it. Ultimately, I tend to think that this study confirms not what I would call "subconscious racism", as did the article, but instead exposes very real and intractable racial differences (maybe more accurately just cultural differences that for now take hold along racial lines). Maybe black players do commit more fouls. Unfortunately anyone who suggested as much publicly would be harangued. "That's just racist, man, that's ignorant." Good thing we can't even talk about it, heaven forbid we might learn something.