Every four years I predict who is going to win the presidential race and every four years I get it wrong. So this year shall not be different!

My prediction is as follows:

While I don’t think that all of the polls are wrong, I suspect that the results will look more like the shows that are showing a close race rather than the ones predicting a blow-out. At least I think so as it pertains to the popular vote. I think that the race will be a little bit closer than it now appears for a few reasons. Even though its effect has been diminished, I think we’re more likely to see a Bradley Effect in a presidential race than anywhere else. We’re on the cusp of electing our first black president, and I think publicly people want to be seen as being on the right side of that whereas in the privacy of the voting booth I think they may go with their gut. I think that there also may be a Shy Tory Factor where conservatives are less likely to talk to pollsters in general. Mostly, though, I think that a lot of the undecided voters are people that voted for Bush in 2004 and not having been completely sold on Obama will fall into line for McCain.

I don’t think that this will be enough, though, and so I am predicting an Obama win of about 4% with Obama getting a 51% and McCain getting 47%. Basically, that’s a little better for Obama this year than Bush did in 2004. However, on the Electoral College front I see this different translating into a substantial Obama victory. By and large it seems that Obama has run the tighter and more organized campaign. I see Obama’s people getting the vote out where it counts the most. Of course, even there my predictions are modest compared to a lot of the polls, but it’ll certainly be a lot better for Obama in 2008 than it was for Bush in 2004 as far as the EC goes.

I also predict that the media is going to be very quick to call states that Obama wins and slower to call McCain states. I don’t see any reversals as I think that the states that they call for Obama will ultimately go to Obama, but I think (for instance) that you’ll get an early verdict on Obama’s victory in Virginia and only later will they decide that McCain won North Carolina, even though both will win their their respective states by about the same margin. We saw something very much like that in 2004 where the press could not call Pennsylvania fast enough even though in the final tally it was nearly as close as Ohio and closer than Florida. I think that the main reason that the press will do this is not to “fix” the election or anything like that, but because the press has clearly believed that Obama would win this from the start and anything that steps in the way of that belief will be due greater scrutiny than would anything that confirms the belief.

The Democrats will make gains in the House as well as the Senate, but will fall a couple seats short of the 60 that they need for a filibuster-proof majority. Despite what the polls say, I predict that Proposition 8 (banning gay marriage) in California will fail.

Anyway, below are my predictions for the 2000 and 2004 race, where you can see how wrong I was. My 2000 prediction doesn’t look terrible until you consider that I predicted that Bush would win the popular vote by a substantial margin and underperform in the Electoral College. My 2004 popular vote prediction was reversed, with Kerry winning by 3% instead of Bush.

Note that the EC tally for the 2000 election isn’t right because it uses post-census numbers.

-{I am not even going to try to moderate the comment section on a post like this. I am putting this up for posterity rather than to spark debate, so the comment section is closed.}-

-{After further consideration and encouragement from Abel, I’ve decided to open comments up on this post for predictions and predictions only. No candidate advocacy or denigration, please.}-

Addendum: As I mentioned in this post and last, I get an intuition on election day that can contradict my previous predictions and is usually more accurate. This morning, the intuition was: Landslide.