Over at TNR, various smart folks are worried about the election. Reading their concerns, though, the anxiety seems like a habit of mind in search of a compelling rationale. And because this is the Glorious Ezra Klein blog, let's start with a chart:
What you're looking at is the Pollster.com tracker starting on September 1st. As you can see, the late phase of this election, the phase in which Obama is winning, begins around September 10th, when McCain loses his lead and sees his numbers enter a prolonged period of decline. During this period, McCain goes from a high point of about 47 percent on 9/11 to a low point of about 42 percent on 10/12. Since then, he's rebounded slightly. That's a loss of about a sixth of a percent a day. Meanwhile, Obama went from a low point of around 44 percent to a high point of around 51 percent. That's a gain of about a fifth of a percent a day. It's true that there's been some late-stage tightening, but how much? McCain's final number on the chart is 44 percent. That Obama dip you see at the end has put him a shade over 50 percent. All this happened over the course of an incredibly volatile month-and-a-half, in which the stock market bottomed out, three debates happened, Palin went from steamroller to laughingstock, and Obama began flooding the airwaves with unprecedented advertising. When worrying about the election, you have to not only argue that the fundamentals are somehow changing, but that they're changing much faster than at any other point. McCain would have to make up in a week what he lost in a month -- and in an incredibly bad, event-driven month to boot. It's not impossible, but it's unlikely. And I'm just not seeing the evidence. Noam Scheiber and John Judis both worry that there will be a white backlash driven by a) images of black early voters and b) an economic message that speaks of socialism but dog whistles racial redistribution. Maybe, but there's no real evidence of that. Jon Cohn thinks that McCain is finally focusing on the economy, and his new message "is bound to improve his poll numbers." Maybe. But my sense is we're past the point in the election when a slightly retooled message can make up that much ground, that quickly. This election has broadly been driven by events, not strategies, and the only potentially predictable event left is a racial backlash that hasn't manifested in any surveys or polls or other data points. It's probably a safe bet that the polls will tighten as some Republicans return home to McCain, but not by much. My prediction -- which may look very dumb in a week -- is that the two-party vote share will end with McCain at 46% and Obama at 54%. Events or undetectable social trends could, of course, flip this election. As the saying goes, in politics, anything can happen. But it usually doesn't. Update Judis e-mails to note that he's not pessimistic. In fact, in the post I linked, he says, "I think that...Obama supporters can feel pretty confident that he will win by three or four points." Meanwhile, some in comments are wondering why I'd even write a post like this. The answer, simply, is that I think it's correct. I could be wrong about that. But my sense is that the media wants this to be a volatile race, liberals can't believe it's not a volatile race, and conservatives have to believe it is a volatile race. But though it could probably become a close race, I can't find any data that really shows McCain on the verge of a sufficient comeback. But maybe others will see things differently, or have data points I'm not taking into account. Thus, a blog post!
