There are three ways to predict the Oscars. The first is to go strictly on facts, stats and precedents. The second is to use a mixture of sentiment and psychological projection. The third, and best, combines a bit of both with good old-fashioned gut instinct.
Statistician extraordinaire Nate Silver, unsurprisingly, opts for the first method. His election-style Oscar predictions are based purely on how nominees have fared in previous awards: a system that poses some problems this year, when the Academy’s earlier-than-usual deadline for nominations voting resulted in less correlation than usual with several major precursors — the Actors’ and Directors’ Guilds in particular. (Stat geeks love to tell us, or example, that Marcia Gay Harden is only person in the 19-year history of the SAG Awards to win the Oscar without a Guild nod — but there’s a strong possibility that number could triple on Sunday.)
Silver’s justification for factoring only precursor awards into his predictions is other external factors — industry sentiment, campaign savvy, potential backlash, and so on — often amount to “more noise than signal.” He writes: “If a film is the cinematic equivalent of Tim Pawlenty – something that looks like a contender in the abstract, but which isn”t picking up much support from actual voters – we should be skeptical that it would suddenly turn things around.”
Fair enough. There’s certainly no arguing with Silver’s method when he predicts landslide wins for “Argo” in Best Picture, Daniel Day-Lewis in Best Actor and Anne Hathaway in Best Supporting Actress — when a contender wins that many of of the key precursor awards, it’s usually only outside observers, as opposed to actual voters, who get hungry for a momentum shift. Still, Silver’s rankings in these locked-and-loaded categories are open to question: does anyone really think “Zero Dark Thirty” is currently running second to “Argo” simply because of its early run of critics’ prizes? Is “Silver Linings Playbook” really coming in sixth? Surely not.
Silver has trouble with Best Director, where the fact that the most significant precursors have been won by a non-nominee leads him into grayer territory. Taking previous nominations into account leads him to predict Steven Spielberg, by a hair, over Ang Lee, though he admits this is the same method that made him predict Taraji P. Henson over Penelope Cruz four years ago (ha!), and that it’ll be “blind luck” if he’s right on this occasion.
He also has Emmanuelle Riva — my prediction or Best Actress, as well is Kris’ — down in third place, claiming that the BAFTA is a less reliable indicator than SAG in the acting categories. True, though could as easily mention that the Brits have called six of the last seven Oscar winners in the category. And the reliance on past wins is highly questionable in Best Supporting Actor, which he’s currently calling, reasonably enough, for SAG winner Tommy Lee Jones — but with BAFTA and Globe winner Christoph Waltz in third, and Robert De Niro rather implausibly in last place.