Close race might actually be good for Dems

Conventional wisdom says Obama will need a big lead come election time, because as was shown in the Democratic primary polling months ago, his actual vote tallies tended to be lower than the polls predicted. The reason given for this is that some people say they plan to vote for Obama, ashamed to admit their racial bias, but actually vote the other way. This is extrapolated to the presidential election by some folks, and there may be some truth to that, to be sure.

But on the other hand, I could see it working this way come election time: If Obama has a big lead in the polls, disgruntled Clinton backers may want to register their protest by voting for McCain, figuring the Dems will get the White House anyway. In this case, the polls could be off, and the protest votes could tip the election to McCain.

But if polls show a neck-and-neck race, those same people might be more motivated by the fear of the reality of a McCain presidency. They might still tell pollsters they won't vote for Obama, but when they get in that booth, visions of Roe v Wade overturned, far-right Supreme Court justices on the bench for a generation, extended war and more of the Bush years should be enough to bring them to their senses.

The same could be true for people who normally vote Republican. Many professed Republicans are angry with their party, and rightly see that traditional conservative values are not being well-represented by the GOP of late. Therefore, many have said they will vote for Obama. There may be many more that feel this way, but don't want to be seen as disloyal to their party, and are keeping mum about it. Come election time, if the polls are close, they might secretly vote for Obama, knowing the Dems will be better for working people like themselves. However, if polls show Obama with a comfortable lead, they might feel they should vote GOP out of loyalty.

I've got no hard data to back this theory up, but hey, this is blogger-land, where everyone can spout their little predictions and possible scenarios - we don't need no stinkin' data! But seriously, I could see this effect panning out. What do you think? Leave a comment!