Silver and Gold

Posted on February 23, 2009

I like Nate Silver, the adorkable math whiz with his square black frames and Lewis Skolnick haircut. Silver is a math prodigy, best known for perfecting something called logistical regression—which my two times taking high school geometry leaves me unable to explain.But the gist is this: In 2008, Silver used a math model he originally developed for baseball statistics to correctly choose the presidential winner in 49 of 50 states. On his website, FiveThirtyEight.com, Silver also correctly id’d every Senate race, making him quite the star in political circles, quoted by Bill Kristol, interviewed by Keith Olbermann, and described by the New York Times as one of the breakout online stars of the year.Yet when New York Magazine asked him to use his sabermetric formulas to correctly predict the Oscars, he should have known it was a bridge too far. Sabermetrics might be a good fit for baseball, where empirical data and algorithms make day-to-day sense. Baseball’s a stat sport where performance averages are fluidly tracked. The same, in some sense, holds true for politics, where voter performance state-by-state, and even city by city, is trackable and therefore trendable.The Oscars, however, seemed far-fetched. Can art be mathematically predicted?New York was willing to give it a go. Of the hullabaloo, they wrote, “Informally, it involved building a huge database of the past 30 years of Oscar history. Categories included genre, MPAA classification, the release date, opening-weekend box office (adjusted for inflation), and whether the film won any other awards. We also looked at whether being nominated in one category predicts success in another…”Here’s the problem: Art ain’t logical. Creative expression isn’t something that can be broken into empirical optimization algorithms like Derek Jeter against Curt Schilling. Nor are the Oscars a choice between two. While gamesmanship naturally enters the equation, the politics of Oscar aren’t the equation as a whole. Many don’t vote by loyalty, classification or box office. They use fuzzier logic: personal taste.(And that’s assuming there’s logic in taste at all.)In the end, Silver stumbled. His computer chose Slumdog Millionaire for best picture, but so did everyone else. The same for Kate Winslet and Heath Ledger. On the losing end, he predicted Mickey Rourke as a 71 percent favorite to win Best Actor, but the Oscar went to Sean Penn. (Silver gave Penn less than a one in five chance of winning.) Silver gave only slightly better odds to Penelope Cruz winning Best Supporting Actress, going instead with Taraji P. Henson as the odds on favorite. All told, Silver’s batting average wasn’t much different than the average schlep in an Oscar pool.But that’s alright. Nate remains on General Manager speed dial and will be handsomely paid as a political consultant. Art, on the other hand, Silver doesn’t have the mind for. There’s something else required, something which goes beyond when a film opens, or whether an actor’s won before. A set of numbers he hasn’t found.One stat gap was all too obvious. In an interview last week, Silver sheepishly admitted he hadn’t seen most of the films. C’mon, Nate. You might not need to watch a pitcher to know how effectively he throws, but that’s the difference between pitchers and pictures. Two hours on a mound and two hours in a theater are two vastly different worlds.

© Copyright Walkabout Jones • Powered by Wordpress • Inspired by Detour theme Edited by Cynthia Enciso.