"Professors Ken Bickers and Michael Berry, of the University of Colorado, have a system for predicting the Electoral College outcomes of presidential races. Their model has accurately forecast the winner of every presidential race since 1980. According to an article published by UC-Boulder, they even got the Perot-flavored election of 1992, and the Bush-Gore photo finish in 2000, right.
This year, the Bickers-Berry model shows Mitt Romney winning with 320 electoral votes to Obama’s 218, with a 20-vote margin of error. A popular vote margin of 53-47 percent in Romney’s favor is predicted.
The Bickers-Berry model draws upon a wide range of state and national economic data, rather than collating public opinion polls. It anticipates little lasting effect from factors such as the location of the party conventions, the vice-president’s home state, the party affiliation of state governors, or – according to Bickers – “gaffes, political commercials, or day-to-day campaign tactics.” He finds the focus of voters upon big issues “heartening for our democracy.”
The Associated Press notes that “the model does not account for sudden changes in the economy or unexpected developments in states split 50-50.” There appear to be quite a few states fitting that definition at the moment. The Bickers-Berry model has Obama losing almost every swing state, including Pennsylvania, Ohio, Wisconsin, and Florida.
Interestingly, the model predicts different partisan effects for two key economic factors: “Voters hold Democrats more responsible for unemployment rates while Republicans are held more responsible for per capita income.” That’s obviously not good news for President Obama, who has made double-digit unemployment a permanent feature of the American landscape.
The forecast that has Romney winning with 320 electoral votes is based on five-month-old economic data, with an update planned for late September. Maybe Romney will do even better, when even more dismal Obama economic data is plugged into the Bickers-Berry model.
On the other hand, the professors note that it’s hard to predict if the public will judge the economy in “absolute” or “relative” terms – in other words, will they consider the totality of the Obama record, or will they accept a possible uptick in a few key indicators during October as encouraging signs that the President is turning around?"