FiveThirtyEight and RealClearPolitics are two sites that have garnered acclaim and traffic during the election season as voters try to get a leg up of the latest prognostications. While we can't know exactly how the election will turn out, these sites have forecasting down to a science. But there's another website that could give us a peek into who will win this year's hotly contested presidential election: Google. The megabeheamoth search engine is actually a source of tremendous - and entertaining - data and some it of can actually correlate with election outcomes as Seth Stephens-Davidowitz explored at the New York Times.
While so much of the information gleaned from the search data is either ridiculous - the popularity of "Paul Ryan shirtless" - or superfluous - how the number of searches of a candidate in a region corresponds to their popularity in said area - there are some useful tidbits to be pulled for the campaigns, particularly in terms of voter turn-out. Says Stephens-Davidowitz:
If search rates for voting information were higher in the first half of October 2008 than in the first half of October 2004, voting rates tended to be higher in 2008 than in 2004. It's true for midterm elections, too. If search rates for voting information were higher in the first half of October 2010 than in the first half of October 2006, voting rates tended to be higher in 2010 than in 2006.
This predictive power was significantly stronger than that of other variables we might use to predict area-level turnout, like changes in registration rates or movement in early voting.
Of course, there's plenty of garbage to sift through as well, as Stephens-Davidowitz notes, including searches about Romney and Mormon underwear as well as Obama and racist epithets. Still, once you look past the noise, there may just be some patterns worth teasing out, adding to the stacks of numbers already being crunched for November 6.