Forecasting the US Midterm Elections

Since early September, Alfred Cuzán has been applying the skills of Polly, the election-forecasting parrot, to predicting the outcome of the U.S. midterm elections. Polly is a colorful and vocal ambassador for the important evidence-based principle that one should combine forecasts from diverse methods and independent forecasters in order to improve accuracy.

The PollyVote Project uses the high-profile application of election forecasting to demonstrate advances in forecasting methodology. Her focus is on the principle of combining by averaging forecasts from different component methods. This approach has lead to highly accurate forecasts for the past five U.S. presidential elections. Polly's caretakers are Scott Armstrong, Alfred Cuzan, Andreas Graefe and Randall Jones.

For several weeks, Polly has been combining predictions about the U.S. midterm elections for the House of Representatives obtained with several methods (statistical models, judgment methods, and a betting market). Across methods, the mean forecast is for the Democrats to lose 43 seats relative to their 2008 total, three more than needed for the Republicans to retake control of the chamber.

Check the latest forecasts…

Now Polly has added the U.S. Senate to its reports. Taking the median of 10 predictions lumped together, the forecast is for the Democrats to end up with 51 seats, for a net loss of eight. That would still leave them in control of the upper house. The end result will be a divided congress, with the Republicans controlling the House and the Democrats the Senate.