The Obama Effect
The big story of the past few weeks has been the fact that Barack Obama is not running away with the presidential race. In a political climate in which the GOP is very unpopular (generic congressional ballot polls indicate that Democrats lead Republicans by double digit margins), most pundits think that Obama should have a huge lead over McCain. Instead, the race is tight—the RealClearPolitics poll average puts Obama up by only 3.6 points. This turn of events has Republicans feeling a bit of hope, while Democrats wonder what is wrong with the Obama campaign.
Still, Obama is considered the favorite—InTrade gives him a 60-40 edge over John McCain. He has kept his lead in the polls, however narrow (since May, only poll has put McCain on top), he leads in the RCP electoral count 238-163 (without counting toss up states), and his favorable ratings hover a few points above those of McCain. Liberals worry about Obama’s poll numbers, but console themselves with the thought that if the election were held today, the polls indicate Obama would win, and win by a reasonably comfortable margin.
But the polls always indicate the Obama will win. The Dinkins Effect is one of the most notorious phenomenons in politics—black candidates almost always do much better in the polls than in the actual election. And Obama will almost certainly be affected by this trend.
The Dinkins Effect has already had a significant effect on the Democrat primaries. After the Iowa caucuses, Hillary Clinton was doomed. She had finished third in Iowa and trailed Obama by massive margins (Rasmussen had her down by seven) in New Hampshire. If she lost two consecutive contests, she was finished—and given Obama’s impressive poll lead, her chances of pulling out a victory were next to none. Some pundits speculated that if she could keep her margin of defeat to a few points, she would have done well.
Clinton won New Hampshire by three points.
Rasmussen is an accurate pollster, and virtually every other polling organization gave Obama a similar lead. The fact that he lost was a staggering and wholly unforeseen defeat. But it was not the last such defeat.
After Super Tuesday (which Obama did win decisively, though he did manage to turn a one point Rasmussen lead in California into a nine point loss) and the resulting Obama tsunami (Obama won twelve straight primaries), the next important states were Ohio and Texas. Winning either state would give Obama the nomination. Rasmussen gave Clinton a six point lead in Ohio, and put Obama up by one in Texas. Clinton trounced Obama by ten in Ohio, and won Texas by four. Once again, Obama had underperformed expectations.
Still, Obama could seal the nomination by winning Pennsylvania. Rasmussen (why do I always use Rasmussen? Because heir polls are considered very accurate, and they have an easily searchable website) had Clinton up by five, which put Obama in striking distance. But Hillary blew him out by ten points.
New Hampshire, Ohio, Texas, California, Pennsylvania—all are important states. And Obama underperformed in all of them—and his dismal performances all fell either out of or at the outer reaches of the margin of error.
Of course, polling primaries (particularly a primaries where the candidates have so much in common; about the only that separated Obama and Hillary is race and gender) is different from polling a general election. But still, the disparity between Obama’s polling results and election results can’t be good news for the Obama campaign. If we assume that the RCP average of state polls gives Obama three points per state more than he will actually get, his electoral count plummets from the 317 electoral votes to 255.
Perhaps Obama’s general election performance will be entirely different from his results in the primaries, and maybe something will happen between now and November which will make all the current polls wholly obsolete. But if the November political landscape resembles the present one in the least, Democrats should not be overconfident in November. If fact, perhaps they should not be very confident at all.