Reductionism has its charms. You can eliminate a lot of clutter and complexity by believing that the future will hinge entirely upon one dis­cern­ible and measurable factor, element or thingamajig. It’s like knowing a code word, or a secret handshake.

Remember this:

“Ben — I just want to say one word to you — just one word.”

“Yes, sir.”

“Are you listening?”

“Yes I am.”


So what’s going to decide the 2012 presidential election? “The economy, stupid,” is the officially sanctioned response. And the economy will depend on oil prices. Or something happening in Greece, maybe.

Carter Eskew picks the unemployment rate. That’s the statistic that matters, he think. I’m not sure about that, but in his blog post he makes a good point that everyone should heed: “presidential elections are increasingly a series of state-wide elections” His message is that the national unemployment rate doesn’t matter as much as the unemployment rate in, say, Ohio, or Virginia, or Iowa, or Missouri — swing states that have lower-than-average unemployment and where Obama could well pick up the electoral votes he needs to get over the top.

Of course, under the Constitution, presidential elections have always been a collection of statewide contests. The national popular vote doesn’t matter a jot. So it’s hard to know what to make of national polls, particularly ones that don’t sort out the likely voters. This morning, the estimable Nate Silver of the Times warns that we’re still in the general-election preseason, and haven’t seen the best and sharpest arguments from the campaigns about why their guy is the right choice and the other guy is a doofus. Silver focuses on Obama’s national approval rating. Talk about a country that’s divided on Obama: “Right now, in the RealClearPolitics average, 48.3 percent of Americans approve of the job that Mr. Obama is doing, and 48.6 percent disapprove.” So, close election ahead. According to Silver, approval ratings are good predictors of elections, particularly for incumbents.

So what causes a voter to approve of the job a president is doing? There’s no easy answer. I’ve interviewed a lot of voters over the years and you can’t predict what they’re going to say or what they really care about. Americans are a diverse bunch. There’s all types. It’s hell on reductionists.

Obama’s approval numbers are pretty steady. He remains the favorite to win, but it’ll be close in all likelihood. As I wrote last August, Obama goes into the fall with some wiggle room, because he won a large Electoral College majority four years ago. But structurally the country is in a phase in which all elections will be fairly close if not excruciatingly so. We remember 2000, obviously, and we should remember that John Kerry needed merely to convert about 60,000 votes in Ohio to win the White House. Even four years ago, Obama piled up his large EC majority with a series of fairly narrow wins in battleground states.

Barring something unexpected popping up on the campaign trail, it’s hard to see how it won’t be another close election. The usual drill: Ohio and Florida and a very late night.