DENVER - A University of Colorado analysis of election factors that's accurately predicted the winner of every presidential election since 1980 revealed Wednesday that the 2012 winner will be Republican Mitt Romney.
"Based on our forecasting model, it becomes clear that the president is in electoral trouble," said Kenneth Bickers, a political science professor at CU-Boulder.
According to an analysis that looks at economic data from all 50 states and the District of Columbia, President Barack Obama will win 218 electoral votes, well short of the 270 benchmark he must reach in order to win a second term.
That calculation paints an awfully different picture than the current electoral map eleven weeks from Election Day, where RCP averages show Obama with a slight lead in nine of the 10 swing states that will decide the election: Colorado, Florida, Iowa, Michigan, Nevada, New Hampshire, Ohio, Virginia and Wisconsin.
Only North Carolina shows Romney ahead based on the average of recent polls.
Although the polling averages show Obama's lead on those swing states is mostly within the margin of error, the president would have to lose all 10 of those states to come in at 218 electoral votes, as the CU study predicts.
Looking at per capita income and state and national unemployment rates, the study's authors conclude that Obama's incumbency advantage is offset by stubbornly high unemployment -- and that other factors, like the home state of the top candidates, support from swing state governors and the locations of the party conventions, don't mean much at the end of the day.
According to Bickers, election prediction models "suggest that presidential elections are about big things and the stewardship of the national economy. It's not about gaffes, political commercials or day-to-day campaign tactics."
But, Professor Michael Berry, who conducted the study along with Bickers, cautioned that the model isn't a sure thing just because it has been for 32 years now.
"As scholars and pundits well know, each election has unique elements that could lead one or more states to behave in ways in a particular election that the model is unable to predict," Berry said.