You know the presidential campaign has taken a turn against John McCain when everyone stops talking about how close the elections will be and starts talking about Barack Obamaâ€™s big lead and the potential for a rout. So, should Obama supporters start ordering the cake and balloons for their November election-night parties? Maybe not yet. Not only can overconfidence lead to lower voter turnout for Obama, but we still donâ€™t know if there will be a Bradley effect.
Unfortunately, American electoral politics has consistently shown that, when a black candidate is running against a white candidate, the black candidateâ€™s pre-election poll numbers end up being higher than their actual vote tally. The theory is that certain white voters, for fear of seeming racist, will tell pollsters they are voting for the black candidate when they actually intend to vote for the white candidate. Or, they change their vote at the last minute.
Whatever the cause of the Bradley effect, we canâ€™t assume it wonâ€™t apply this November. Already in this election cycle, some observers questioned if the effect played a role in the New Hampshire primary where polls showed Obama ahead of Hillary Clinton by as much as 13% before he ended up losing the state by 3%. There are plenty of other reasonable explanations for the discrepancy (not the least of which is bad polling) but, certainly, the Bradley effect is a real concern.
I do believe we are a less racist nation now than we were even ten years ago and Iâ€™d be surprised if a significant number of voters are intentionally misleading pollsters. I expect, barring one heck of an October surprise, that Obama will win. But itâ€™ll be interesting to see if thereâ€™s any Bradley effect in what is the biggest election ever between a black candidate and a white one.