Last night's Oscars were a chore to watch, and not just because the broadcast dragged on forever or because James Franco never warmed to his role as co-host. The larger problem was that there were no surprises; all the favorites won. Christian Bale? Check. Colin Firth? Check. "The King's Speech"? Check. Maybe Oscar punditry has gotten good enough that actually watching the show seems superfluous.
I blame Nate Silver, or at least people who think like Nate Silver. He's the baseball statistician who revolutionized election punditry in 2008 with his nearly spot-on forecast of that year's electoral college outcome. (He was wrong only about Indiana, which went narrowly for Obama.) His key insight was that election results follow predictable patterns — just not the patterns that candidates or the media typically worry about. Rather than looking at a single surprising poll out of, say, North Carolina as evidence of a sudden bad trend for Obama there, Silver looked at various underlying demographic factors, the pollster's history of reliability, an average of other polls, and polls in neighboring states — all of which provide more useful information than one potentially misleading poll.
In the same spirit, Oscar prognosticators — including Silver — have figured out which awards are good predictors of the Oscar results, and which aren't. Hint: The Golden Globes aren't, and those awarded by film-critic societies are often worse. Not surprisingly, the best predictors are awards given by groups, such as the Director's Guild of America, with members who actually vote on the Oscars. Meanwhile, online betting markets, which have a decent record of predicting winners, can help round out the picture.
These signals aren't 100 percent reliable. For one thing, some plausible-sounding patterns are entirely speculative. One theory that floated the Oscar-viewing party I attended was that the Best Picture winner wins the best cinematography and best art direction awards about 98 percent of the time. In fact, the two lesser awards don't even track each other.
Furthermore, Silver himself didn't have a perfect record. Best Director was his Indiana; that Oscar went to Tom Hooper of "The King's Speech," not David Fincher of "The Social Network." Then again, that was a risky pick; Hooper was the favorite among pundits — as were all the other major winners.
This will be a quandary as Oscar punditry keeps getting better. Do viewers want to invest hours watching a sprawling, often rudderless broadcast, just on the off chance that one of the pre-show predictions happens to be wrong?