Nate Silver is leaving The New York Times for ESPN when his contract is up in August, and his fancy new home will have him performing in a vast array of new roles. Like a part in the Oscar broadcast, an event that has eluded him in the past. Because, well, Nate Silver isn't very good at guessing the annual award show.
We'll preface this with a necessary disclaimer: one of the most interesting and detail filled reports about the battle between The New York Times and ESPN/ABC to land Silver comes from Politico's Mike Allen in Sunday morning's Playbook. In case you don't know, Silver and Allen are old foes who have sparred openly over the last year. Politico hates Silver's level-headed statistical analysis and Silver thinks Allen's Politico is covering politics all wrong. So, there's a history here.
Allen reports Silver began courting serious offers from the Times, ESPN/ABC, and NBC/MSNBC shortly after the election. (The election that he called with deadly precision.) The Times and ESPN quickly became the two major players; executive editor Jill Abramson, publisher and chairman Arthur Ochs Sulzberger Jr. and CEO Mark Thompson working for the Times, while ESPN president John Skipper, senior vice president Marie Donoghue and Grantland publisher David Cho worked for the Worldwide Leader. On Saturday, The New Republic's Marc Tracy reported ESPN were trying to land Silver for the better part of five years, and he downplayed the "tensions" between Silver and the Times that may have led to his departure. But Cho seems like the behind-the-scenes hero here considering the new model for FiveThiryEight.com resembles exactly what ESPN ended up doing with Grantland. Per Allen:
The business model mirrors Grantland’s: a strong, independent brand that ladders up to the bigger brand of ESPN (and, in this case, ABC News). Nate will appear on the air on ESPN and ABC, and will get “verticals,” or web hubs, devoted to a variety of new topics. He’s very interested in education, so there’s been a lot of conversation about that. And, of course, weather and economics. His Oscars predictions did well for The Times, and now he’ll work for the TV home of the Oscars.
So that's what will meet Silver when his contract with the Times is finally up and he makes the move to ESPN in August: a new, dedicated website where he can crunch numbers on weather, education, or economics. Whatever he likes, really.
But one of the most interesting parts of Allen's report is that detail about Silver's role in the Oscar broadcast, because the normally lights-out statistician is kind of terrible at calling the Academy Awards. His first notable stab at calling Oscar predications happened in 2009 for New York magazine. His model factored in everything from Oscar history to genre to box office performance. But it didn't work; Silver went four for six guessing the major categories that year. (He had Mickey Rourke beating Sean Penn for Best Actor and Taraji P. Henson beating Penelope Cruz for Supporting Actress.) After a two year hiatus, Silver went five for six in 2011 with a refined prediction model. (He had David Fincher beating Tom Hooper for Best Director.) So he seemed to be on the right track. But after another year off, Silver returned to the Oscars in 2013 and went four for six again. (He had Steven Spielberg beating Ang Lee for Best Director and Tommy Lee Jones beating Christoph Waltz for Supporting Actor.)
The Oscars, like the Super Bowl, continues to evade Silver's usually perfect prediction model. His predictions would be pretty good for a layman like you or me, but the expectations are that Silver can always game the system and that hasn't been the case so far. But now he wants to put himself on the line, and on television, to finally conquer the movies. If he does it, curse the guy in your Oscar pool who copies Silver's picks every year. If he continues to fail, let the sheep follow him. Bet against his picks, especially in the supporting acting and directing categories, and let your wallet reap the rewards.
This article is from the archive of our partner The Wire.