Tim Wigmore

How the forecasters could fail for the 2012 presidential election

The really curious thing about this year’s US presidential election is that it looks set to defy all political forecasts. While the most respected political science models have predicted victory for Mitt Romney, polls have consistently suggested otherwise.

Political science and predictive models seldom receive much attention in the UK but they enjoy a strong tradition in the US. And they normally get it right. Most spectacularly, the University of Colorado’s forecasting model in 2000 successfully predicted that Al Gore would win the popular vote but lose the electoral college. Since its inception in 1980, the University of Colorado’s forecasting model has successfully predicted the results of the last eight US presidential elections.

Colorado’s model is based on forensic analysis of state-by-state factors, especially unemployment rates and real per capita income. Contrary to conventional wisdom, the model finds that there is no statistical significance in ‘the location of a party’s national convention; the home state of the vice president; or the partisanship of state governors’.

Comments

Join the debate for just $5 for 3 months

Be part of the conversation with other Spectator readers by getting your first three months for $5.

Already a subscriber? Log in