Implicit bias and the ‘eye test’

The guys at Boxscore Geeks have been killing it lately. Patrick Minton had a great post last week, titled “Let’s Face It: We’re All Racists.” And Dre Alvarez (along with Patrick) have been devoting time in the site’s two most recent podcasts to implicit bias, a concept that’s getting considerable attention inside and outside of academia.

These threads all point in the same direction – our biases, and our eyes, deceive us all the time. When Patrick says we’re all racists, he’s not talking about the dyed-in-the-wool devotees of racial supremacy. He’s talking about ordinary folks – you and me – and how our unconscious biases influence the way we see the world. Dre has talked about the power of unconscious, or implicit bias, to shape our views of reality and has noted that one antidote for that enduring problem is to make those implicit biases explicit, to draw them from the shadows of the unconscious into the light of the conscious brain. Doing that might allow us to act to change behaviors, if we so choose, that were driven by the unconscious bias.

One instructive example was the study that Joseph Price and Justin Wolfers did a few years ago on referee bias in the NBA. Their study found that officials – black and white – tended to make calls in favor of their own kind. Though the bias ran in both directions, the greater number of white refs in the NBA than black ones had a disparate impact on black players. But the bias ran in both directions. The NBA angrily denied that the zebras displayed anything but the height of professionalism in their work. But an interesting thing happened: in follow up work, the evidence suggests that awareness of the possibility of bias on the part of the refs actually reduced their bias in officiating (Dave Berri recently discussed the original research and the follow-up here). These were not snarling, hood-wearing individuals. They were merely human, governed by the same unconscious forces that influence us all. When made aware of these in a specific context, they appear to have had the capacity to change their behavior.

One of my favorite books in political science is Tali Mendelberg’s The Race Card. In it, the Princeton scholar examined the 1988 presidential election, best known for the so-called Willie Horton ad (and Michael Dukakis’ ridiculously dorky tank ride). She showed that the power of the Lee Atwater-led Bush campaign was to capitalize on unconscious bias. Atwater himself summed up this approach to using race in politics in the most stark and powerful terms. But the point was this – in an age when most individuals don’t think of themselves as racist and don’t want to be seen as racist, racialized messaging works only when it is implicit, not explicit. Talk about blacks as savages or intellectually inferior and you’re going to turn almost everybody off. Talk about crime and welfare and you’re messaging is going to be palatable to much a larger number of people. As in the Price-Wolfers study, Mendelberg found evidence that when implicit appeals to race were made explicit, people recoiled from their own prejudices and saw the objects of their bias differently.

All this is of profounder social import than evaluating athletic talent. But there is a connection (actually, there are several connections). Specifically, our eyes, memories, hearts and minds do deceive us. It’s an inescapable fact of our humanity. And it’s why the cherished “eye test,” a phrase that did not exist in the sports lexicon when I was a kid, is so problematic.

To explain why, I am going to reproduce here something I wrote almost a decade ago, about the criticisms of Michael Lewis’ Moneyball and sabermetrics more broadly (I also happened to be writing at the time about Dave Berri’s then newly-published Wages of Wins and Bill Gray, a well known hurricane expert who was a global warming skeptic. Gray insisted he could discern more about climate phenomena by flying into hurricanes than by consulting computer models and sounded like one of the angry scouts in Lewis’ book describing the young climate scientists).

Here’s me, circa 2006:

One of things that struck me about the criticisms of Moneyball when that book came out three years ago was the anger directed toward the author and, by extension, Beane’s methods themselves, for displaying a kind of pointy-headed arrogance about sabermetrics’ understanding of baseball. But…the charge of arrogance arguably better applies not to the data crunchers, but to their critics:

There are interesting parallels between [the global warming] debate and the one in Michael Lewis’ Moneyball, the book that explicated Oakland A’s General Manager Billy Beane’s philosophy for evaluating talent and putting together a consistently competitive baseball team on a limited budget. On one side of the divide that Lewis sets up in the book (and over-states to some extent) are the traditionalists, that is, the scouts, who observe young players in person and assess their future potential based on what their eyes and experience tell them. These are blue-collar, beer drinking guys who’ve spent a life time in the baseball equivalent of the trenches – in broken down hotels and half-empty ballparks in the middle of nowhere. On the other side are the so-called sabermetricians, the new wave of baseball analysts and personnel people who rely on sophisticated computer analysis of a player’s previous record to project their likely future performance. Lewis portrays the scouts as resentful at the young whipper-snappers who don’t actually know anything about the sound and feel and smell of the game, so wrapped up are they in their ivory tower models.

The scouts and [global-warming skeptic Bill] Gray see the computer driven intellectuals as elitist and arrogant and out of touch with how things really work. But, the striking thing about the scouts and about Gray is this: the powerful and unstated arrogance in their outlook. Their eyes tell them everything they need to know about how the world works. That humans can deceive themselves, or miss crucial details of a complex picture, or maintain perceptions colored by bias of one sort or another is implicitly ruled out here….

Neither global warming scientists nor sabermetrics claim to have God-like powers of foresight. Instead, they utilize probability theory to argue that, over the long run, the patterns and trends they see have increasing relevance. They can’t know what will happen tomorrow. But, they can make plausible guesses about what, on average, is likely to happen over time. That’s a very different kind of knowledge claim than the one that Bill Gray imputes to his ideological opponents. Gray can’t predict the future either. But, he wants us to trust that he can, because he’s flown into hurricanes and if anyone fails to see what he’s been through, what he knows in his gut, well, they can’t be acting in good faith.

The performance analysts and stat guys (and global warming scientists) can, of course, be arrogant, not to mention wrong. But, The Wages of Wins, like the more rigorous performance-based turn in analyzing sports more generally, is serving an important function: scrutinizing our own pre-conceptions (a worthwhile endeavor in all areas of human thought) and holding accountable the too common seat-of-their pants analysis of the sports commentariat who think that because they say so, no further exploration (or explanation) is required.

As the kids today like to say – Same.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s