In a recent New York Times Magazine story, Nate Silver put forward a model for forecasting the 2012 presidential election. Political scientists Brendan Nyhan and Jacob Montgomery put it through its paces and conclude that it predicts elections pretty well, but not quite as well as existing forecasting models.
In particular, they say, it overestimates the effect of candidate ideology on election outcomes. Statistics wizard Andrew Gelman agrees with them. “The difference in vote, comparing a centrist candidate to an extreme candidate, is probably on the order of 1-2%, not the 4% that has been posited by some,” he writes.