Nate Silver had a good year. After correctly predicting the winner in 50 of 50 states in the presidential election and 31 of 33 Senate races, he was trumpeted by breathless bloggers as the “real winner” of the election. He released a book, The Signal and the Noise, that’s currently at #7 on the New York Times Nonfiction Best Sellers list. He has been called a witch, and even the “Lord and God of the Algorithm.” And in the true test of modern celebrity, he even has his own meme: Drunk Nate Silver.
But is he 'Person of the Year' material? TIME awards the Person of the Year to the “the person or persons who most affected the news and our lives, for good or ill, and embodied what was important about the year.” TIME already showed Silver some love in 2009, when he made their 100 Most Influential People. Still, the top slot is usually reserved for political leaders and pathbreaking scientists, so the case for a data-crunching blogger seems pretty slim. But Nate Silver has become representative of something greater. As he himself said on The Daily Show, for critics and fans alike, his blog “FiveThirtyEight became invested with symbolic power.” Many have lionized Silver not only for his personal acumen but as the champion of some wave of data-fuelled pragmatism they hope will wash away the poison of inane punditry forever. Nate Silver’s success became not just a personal victory or a victory for his algorithm, but a victory for science over ignorance.
It is tempting to try to slot Silver into a narrative of the triumphant march of truth, like William Saletan’s assertion that in this election “reality overwhelmed pretense, gamesmanship, and self-deception.” The implications of Silver’s thrashing of empty-headed pundits are clear. If Americans are willing to give scientific approaches their due over gut feelings, the end of the Tea Party is at hand. We can finally get over phony debates about global warming, evolution, and any number of issues where liberals feel that conservatives ignore the science or the numbers.
The predictions of a realist revolution are a bit premature. I’m not actually a Nate Silver acolyte. The coverage that has suggested “science wins” versus “gut feelings” has been overly simplistic, falling prey to the notion that if you just look hard enough, you can actually “know” who’s going to win. This is a victory of smart and sophisticated use of data in making predictions. Nate Silver didn’t “know” who was going to win. He predicted a 90.9% chance of an Obama victory based on aggregating and weighing a wealth of polling data. A 90.9% chance means that Romney still had a 9.1% chance, but pundits would have lambasted Nate Silver the day after the election if Romney had won, saying the science was wrong. Some of the celebrations of Nate Silver’s glory reveal a similar fuzziness on statistics.
Illustrating how this kind of thinking would play out among the public, Dylan Byers at Politico wrote, “should Mitt Romney win on Nov. 6, it's difficult to see how people can continue to put faith in the predictions of someone who has never given that candidate anything higher than a 41 percent chance of winning.” This highlights the problem that most people do not understand (or perhaps trust) statistics on a basic level. If you put three red apples and one green apple in a bag and tell me to pick one blindly, I’ll say I think I’ll get a red apple based on a 75% probability. The problem is the people who would laugh if I draw the green apple and say “Ha! You’re wrong! Your math is nonsense!”
So let’s not make Nate Silver into God or the devil but hope his popularity encourages a bit more intelligent use of data in public discourse. Silver’s vindication in a bananas media environment is certainly a good thing. To the extent that his time in the spotlight encourages more responsible analysis both by shaming the pundits and by proving the huge audience for it, he has been an important figure.
Of course, I don’t truly believe Nate Silver should be Person of the Year, which would privilege the obsessions of the American blogosphere over the reality of life and death across the world. But when my ballot comes in for Popular Statistician of the Year 2012, I know whose name I’ll be putting down.