Nate Silver at work. (Credit: CBS News)

With great precision FiveThirtyEight's Nate Silver predicted which states President Obama would win in his second-term election. Now he is applying his statistical-analysis methods to predicting who will take home the Oscars on Sunday.

Silver has made Oscar predictions in the past, but with only 75 percent accuracy. By comparison, he picked all 50 state winners in the 2012 presidential election. Silver refined his Oscar prediction models this year by assigning weights to the glut of movie awards that precede the Oscars, similar to how he weighted the various polls in calculating his election predictions. Silver explained:

These patterns aren't random: instead, the main reason that some awards perform better is because some of them are voted on by people who will also vote for the Oscars. For instance, many members of the Screen Actors Guild will vote both for the SAG Awards and for the Oscars. In contrast to these "insider" awards are those like the Golden Globes, which are voted upon by "outsiders" like journalists or critics; these tend to be less reliable.

Silver is giving the best-picture edge to "Argo," best actor to Daniel Day-Lewis, and best supporting actress to Anne Hathaway by wide margins. His results are in line with what the movie critics have been predicting. The competitions for best actress, best director, and best supporting actor, however, will be closer races, according to his calculations. Those awards will be the true test of Silver's Oscar prediction prowess.

(Credit: Nate Silver/New York Times)