OK, here we go: the prediction column. Yes, I am predicting that Barack Obama will win, but since I need to get a whole column out of this, let me build slowly and methodically toward the dramatic conclusion.
First, let me start with this somewhat astonishing (I think) observation. The numbers I’m about to give you? I could have given you this exact same Electoral College prediction in June. Or March. Or January. Or probably even last year. For all the twists and turns, all the debates, all the news developments expected and unexpected, the basic math of this election hasn’t changed barely a whit to my eye. It was always close and still is. The electoral map always favored Obama and still does. It’s almost as if we could have had this election months ago, sparing the nation billions of dollars and a lot of agita.
Another point I want to make before starting. It’s worth bearing in mind in several cases, as I show below, that Obama won by some pretty big margins in some important states in 2008, so he can afford drop-offs of 7 or 8 percent in some states, or at least 5 percent, and still win them. You might want to look back over some of those 2008 margins before you settle into your easy chair tonight.
Now, I begin with the states that are automatic Democratic states—states where neither candidate is even contesting matters. Those states add up to 201 electoral votes (yes, George Will, this includes Minnesota).
Then let’s take three states that are being contested this time but have been voting Democratic at least since Bill Clinton. These are Pennsylvania (Democratic since 1992), Michigan (also since 1992), and Wisconsin (since 1988). Some Republicans are hoping to snare at least one of these states.
Are these states really in play? Granting that nobody knows anything until the votes are counted, I’d have to say it’s highly unlikely. States that have been voting one way for nearly a generation could flip from this column to that. But generally speaking they do so only in a real wave election. This is not a wave election, or at least it certainly doesn’t feel and smell like it is. There is no strong momentum right now in one direction or the other. This race now is just two big lumbering dinosaurs pushing against each other.
Now, if unemployment were still 9 percent and the most recent jobs number was 50,000, then I could totally see it. It would be a wave election, and that wave would sweep Obama away. There’s no question that these states will be a lot closer this time—in 2008, Obama won Michigan by more than 16 points, Wisconsin by around 14, and Pennsylvania by 10. We’re more likely looking at margins of less than half those. They may all be within five points. But it doesn’t smell like we’re in an election that actually shifts them.
So put these two categories, the automatic and the pretty strong leaners, together. They lift Obama to 247.
Romney’s automatic total is 191. Then I guess we throw him North Carolina. I actually don’t believe it’s at all insane to think Obama could pull an upset there. All depends on turnout. But the Democratic Party in the state is in pretty weak condition right now. I’m being a tad generous here since many polls show this within the margin of error, but generous is the kind of guy I am, as you well know, so I’ll put Romney’s automatic plus lean total at 206.
That leaves seven states that combine to 85 electoral votes: Florida (29), Ohio (18), Virginia (13), Colorado (9), Iowa (6), Nevada (6), and New Hampshire (4). These are where the game will be decided. From the math, you can see that Obama needs just 23 of these 85, while Romney needs 64 of them.
Let’s start with the easier ones. It appears from most of what I’m seeing that Nevada is banked in Obama’s column, and that Iowa is close to banked. So that gets Obama up to 259, just 11 electoral votes away.
No Floridas, no hanging chads, no Supreme Courts. A lot of states will be close, but I think the result will be clear enough.
I will give Florida to Romney. It’s totally up in the air, I think, and could go either way. But until it’s proven otherwise, I say that the weight of most polls is probably pretty accurate, and the aggregate of Florida polls seems to me to lean very slightly toward Romney. So that gets him to 235. Four states to go.
I’ll toss New Hampshire into Obama’s column. It’s not so different from Pennsylvania, Michigan, and Wisconsin, except that it did go for George W. Bush in 2000. Again, I look at 2008 margins. Obama won it by 10 percent, so he has a cushion here as well. Polls suggests he should be able to eke out a win of three points or so. That’s four more EVs, or 263.
We are left with Colorado, Virginia, and Ohio. I’m going to give Colorado to Romney by a very slim margin. Obama has tended to lead there, but I’m just going to do it, for two reasons. First, I think Colorado swing voters are probably a little more libertarian than Eastern swing voters, which means a few more Romney votes from them. Second, I just don’t think Obama has quite enough momentum that everything is going to tumble his way. That gets Romney to 244.
And that’s where he stops. Obama takes Ohio and Virginia. In Ohio, I think he’s pretty clearly ahead. It’s what the polls say. Virginia is more mixed, but it’s my hunch.
So: Obama 294, Romney 244. No Floridas, no hanging chads, no Supreme Courts. A lot of states will be close, but I think it will be clear enough. Popular vote will be Obama by maybe 1.8 percent, something a shade under 2 percent.
Mind you I lowballed Obama in 2008, for whatever that’s worth. I didn’t give him North Carolina or Indiana, so I have a history of humility on these matters, you could say.
You can watch my Guardian video of my 2008 prediction here. And here you can watch my predictions about the 2012 race that were made in April 2011, when I left the Guardian and came to the Beast. This was my last Guardian video. In it, I said Obama would win reelection with “just above” 300 electoral votes, which will end up being right if he wins Colorado. I confess that I also said the GOP ticket would be Tim Pawlenty and Marco Rubio. Win some, lose some.