Tuesday, May 29, 2007

My Annual Post On Sports


It isn't very often that I read an article in Jet - or even the New York Times - and find myself reminded of my stats classes in grad school. There's a reason for that. Such articles are few and far between because most ordinary folks can't understand what statistical analysis really means, as a rule. This is one of the reasons people are so quick to say flatly, "You can prove anything using statistics." What they mean is that statistics can be used to "prove" anything - right or wrong. But the reality is that "statistics" can only "prove" "anything" to them. If you have a clue, statistics can only bear out the facts.

When I first read about the study conducted by Justin Wolfers, an assistant professor of business and public policy at the Wharton School at University of Pennsylvania, and Joseph Price, a Cornell graduate student in economics, I knew there wouldn't be much flap over it, even though there ought to be. Then, when I saw the article in the Jet Magazine, I really winced. It was virtually unintelligible. That is, I read it twice, and I couldn't make heads nor tails of what it was supposed to be saying. I would have ignored it, if I hadn't already understood the first piece.

Wolfers and Price found that, during the course of the thirteen N.B.A. seasons from 1991 through 2004, White referees called fouls at a greater rate against Black players than they did against White players. In fact, they went so far as to say that the difference in the rates "is large enough that the probability of a team winning is noticeably affected by the racial composition of the refereeing crew assigned to the game." Whoa!

The crunchy part read something like: "Across all specifications, we find that Black players receive around 0.12-0.20 more fouls per 48 minutes played (an increase of 2-1/2 to 4-1/2 per cent) when the number of White referees officiating a game increases from zero to three." The bottom line: a difference of about two victories in an 82-game season.

The N.B.A. went ball-istic, of course (sorry I couldn't resist). The Commissioners rushed right out and had their own study done, which "found," needless to say, no such problem. Three independent experts hired by the Times to examine both studies, however, declared the Wolfers-Price work far more credible, though they may have been ham-strung somewhat by the fact that the N.B.A. refused to provide the underlying data they claimed to use to make their alternate conclusions.

The N.B.A. shouldn't take it all so hard. First of all, the fact that the whole society is rampantly racist is easy to document. Are the N.B.A. Commissioners trying to claim that it and it alone stands above the socially institutionalized norms? I would hope not. Secondly, it's not as if someone is accusing them of intentional oppression. When they said the league does not consider race in the hiring process for referees, nobody (and certainly not Wolfers or Price) claimed otherwise.

When the referees are taken out of the equation, however, it's hard not to catch the differences. "Player-performance appears to deteriorate at every margin when officiated by a larger fraction of opposite-race referees," write Wolfers and Price, except for shooting free throws (when the referees are irrelevant). And there you have it.

The researchers have tried to reassure the N.B.A. Commissioners that they're not calling anyone racist, per se, but bias based on skin tone can be subconscious, they say. "You see two players [collide] on the floor and you have to call a block or a charge. Does the skin color of the players somehow shape how you interpret the signals your brain gives you?"

The numbers say they do. "Basically, [our research] suggests," says Wolfers, "that if you spray-painted one of your starters White, you'd win a few more games." Damn! No wonder the playing fields are still not level. Folks that look like me gotta win part of the time.

No comments: