Biographic model predicts Obama would win against Palin in 2012

Palin’s book won’t do it: a new forecasting model that analyzed their biographies predicts that she has little chance to defeat President Obama in a potential 2012 show-down. Wharton Professor J. Scott Armstrong and Dr. Andreas Graefe, researcher at the Karlsruhe Institute of Technology in Germany, developed a model to predict the winner of U.S. Presidential Elections based on candidates’ biographies.

Their model, named PollyBio, uses 62 biographical cues that are expected to have an impact on the chances of a candidate on being elected. For example, they look at whether a candidate comes from a political family, is first-born, single child, lost a parent in childhood, is married but not divorced, has children, graduated from a prestigious college, held political offices, has authored a book, has military experience, and is tall and good-looking. The candidate who achieves the higher overall score is predicted as the election winner.

Armstrong and Graefe tested their model for the 29 U.S. Presidential Elections from 1900 to 2008. Their model failed only twice: In 1992, it did not predict Bill Clinton to succeed George Bush and it wrongly predicted Gerald Ford to win against Jimmy Carter in 1980. For the remaining 27 elections, the model correctly predicted the winner, a performance that compares favorably to other models as well as polls and prediction markets.

The forecasts can be made as soon as the candidates are known; they can be issued even before, conditional on who is expected to be in the race. Thus, the model can help the Republican Party in who they should nominate to run against President Obama in 2012.

Based on her biographic data – which includes her new book – Sarah Palin does not seem to be a good choice. While Obama achieves a score of 25, Palin's score is 16. This translates to a predicted two-party vote share of 57.7% for Obama, with a 95% prediction interval of +/- 6.7%. If one removes predictor variables that might favor males in inter-gender races (such as the candidates' height and weight), Obama would still be predicted to receive 56.3% of the vote. Even if Palin would be perceived as more competent, intelligent, and attractive than Obama – three cues for which no data are yet available – the model gives her little chance of defeating the incumbent president.

This suggests that the Republican Party further extends its search. The PollyBio model can help with that. Evaluations of other potential Republican candidates will be posted here in due course.