If you only believe one set of Oscar predictions in the run-up to Sunday’s Academy Awards, Nate Silver’s may just be the most reliable bet. The respected American statistician and founder of the website FiveThirtyEight.com today published his list of the likely winners in the six main categories. And his methodology points to a good night for both Argo and Lincoln.
Silver’s findings largely support the current bookies’ odds. He tips Argo to win the best picture prize and Daniel Day-Lewis to be named best actor for his role in Lincoln. He goes on to suggest that Silver Linings Playbook‘s Jennifer Lawrence has the slight edge over Jessica Chastain in the best actress race. Silver has Les Miserables chanteuse Anne Hathaway winning the best supporting actress Oscar by a country mile, while he tips Lincoln’s Tommy Lee Jones to inch ahead of Philip Seymour Hoffman and Christoph Waltz to triumph in a tight three-way race for the best supporting actor prize.
The statistician identifies the best director category as the site of this year’s closest battles. He tips Steven Spielberg to win by the most slender of margins over Life of Pi director Ang Lee.
Writing in the New York Times, Silver outlines the formula behind the predictions. “Our forecasts are based on which candidates have won other awards in their category.” he says. “We give more weight to awards that have frequently corresponded with the Oscar winn ers in the past, and which are voted on by people who will also vote for the Oscars. We don’t consider any statistical factors beyond that, and we doubt that doing so would provide all that much insight.”
Silver first caught the public’s attention when he correctly predicted the outcome of 49 of the 50 states in the 2008 presidential election. In last year’s contest between Barack Obama and Mitt Romney he went one better, producing a clean sweep of electoral college forecasts. However, he adds that his previous attempts to predict the Oscars – in 2009 and 2011 – was only 75% successful. His new, improved methodology, he admits, “won’t be foolproof”.
guardian.co.uk © Guardian News & Media Limited 2010