The 2012 presidential election is currently too close to call, says Emory University political scientist Alan Abramowitz in his first "time for change" election forecast this year. The model has correctly predicted the last five winners of the national popular vote with an average error of two percentage points.

However, the "time for change" model does predict that modest economic growth heading into the election will lead to a narrow victory for President Obama.

Abramowitz, one of the foremost authorities on election forecasting and modeling, tweaked his model this year to reflect the growing partisan divide among voters and improve its accuracy.  He says fewer voters are willing to cross party lines, which cuts President Obama's incumbent advantage nearly in half, to 2.5 percentage points instead of 5.2 percentage points.

"The last four presidential elections have produced the closest victory margins and the smallest inter-election vote swings of any four consecutive elections in the past century," he explains.

Abramowitz's basic "time for change" model uses three factors to predict the winner of the national popular vote: the incumbent president's net approval rating at the end of June, the change in real GDP in the second quarter of the election year and a first-term incumbency advantage.

Abramowitz says the election could come down to one or two battleground states.  "Be prepared to stay up late on election night," he says.

The full forecast can be found on Sabato's Crystal Ball.