By Rebecca Goldin, Ph.D.
The Chronicle of Higher Education posted a very interesting blog on several people using quantitative models to predict who will win the presidency. According to them, television pundits will go the way of predictors like the home games of the Washington Redskins. If the Washington Redskins last home games before the election are a good predictor (having gotten it “right” 18 out of the past 19 elections), then Romney will be elected the next President of the United States. The quants, on the other hand, are predicting that Obama will win. These include Sam Wang at Princeton, Drew Linzer at Emory, Nate Silver at the New York Times, and Simon Jackson of The Huffington Post. The question is whether Big Data and models using this data in systematic ways will be better predictors than the pundits, economic indicators, or, well, the Redskins.
