research News

Campbell predicts close race; Obama likely to win popular vote


Published September 20, 2012

Jim Campbell
“The prediction is not so definite that a Romney win is impossible, but an Obama win is more likely.”
James Campbell, UB Distinguished Professor of Political Science

A UB political scientist internationally recognized for highly accurate election prediction models says President Obama is likely to receive 51.3 percent of votes cast in the November election.

James E. Campbell, UB Distinguished Professor in the Department of Political Science, notes that while the forecast does not predict the electoral vote winner, it is quite rare for a candidate to win a plurality of the vote and not a plurality of electoral votes. Of course, the 2000 Bush-Gore presidential race proved that it is possible.

“I estimate that there is a 67 percent chance that President Obama’s vote will be over 50 percent,” he says, “so the forecast is for a close race tilted to Obama.

“The prediction is not so definite that a Romney win is impossible,” he says, “but an Obama win is more likely.”

Campbell says his presidential-election forecasts predict the national, two-party, presidential-vote percentage for the in-party candidate. He published his trial-heat-and-economy forecasting equation in 1990 and its companion convention-bump equation in 2004.

His forecasts and how he arrived at them are reported in detail in his article “Forecasting the Presidential and Congressional Elections of 2012: The Trial-Heat and Seats-in-Trouble Models.” Campbell’s article is included in a symposium of 13 forecasting articles that he edited for the journal PS: Political Science and Politics, a publication of the American Political Science Association.

The symposium will appear in the journal’s October issue, but will be online in the next few weeks.

The 13 forecasting models vary in their forecasts from a modest plurality for Barack Obama to a modest plurality for Mitt Romney, but most see the election as being much closer than Obama’s 2008 win.

Five of the models predict that Obama will receive a plurality, though three of these are on the cusp of predicting the election to be a toss-up. Five of the models predict a Romney plurality. The final three models see the election as being a toss-up, though each sees a slight tilt to Obama. The median forecast is of a 50.6 percent Obama-vote squeaker.

The American Political Science Association has scheduled a panel meeting on Oct. 16 in the National Press Club in Washington, D.C., to discuss the forecasts in the symposium.

In his PS article, Campbell also predicts congressional net-seat change of between three to 14 seats for the Democratic Party this year, an outcome derived from two versions of his “seats-in-trouble” forecasting equation.