Goofy post by the usually-acute Noam Scheiber, arguing that Obama's position in the polls is more tenuous than it appears.

I can imagine a losing scenario that doesn't involve outside events. It goes something like this: Obama wins all the Kerry states plus Iowa and New Mexico, giving him 264 electoral votes, then narrowly loses the rest of the red states where he's currently competitive. According to the latest RCP averages, the next most competitive red states, in descending order of favorability to Obama, are: Virginia, Colorado, Ohio, Missouri, Nevada, Florida. Let's focus on Virginia, since it represents the knife-edge between winning and losing--the potentially decisive red state where Obama's currently got the biggest lead.
And then Scheiber lists a bunch of reasons why Virginia might diverge from the rosy national trend: McCain could throw a lot of resources at it; state polls are a few days behind national polls; maybe there's a Bradley effect. "There's no reason to think [McCain] couldn't lose the popular vote by 2-3 points but still win Virginia by 1." He calls this list "the caveats that keep me up at night."

Noam, dude, get some sleep. It's true that Virginia -- or any of the six borderline states that Scheiber lists -- could be a few points off the national average in either direction. Polling is an inexact science, and things change in two weeks. But Scheiber acts as though, once Virginia falls, every other swing state must fall in the same direction. But of course, if Virginia is a few points off the national average, that makes it the exception. And if Virginia goes to McCain, Obama only needs one of the other states on Scheiber's list. (Nevada makes it a tie, which likely goes to the Democrats.)

Put it another way: given Scheiber's starting scenario (Obama wins the Kerry states plus Iowa and New Mexico), McCain needs to win Virginia, Colorado, Ohio, Missouri, Nevada, and Florida. If he does that, it won't be thanks to a flukey dice roll and some bad polling; it'll be a palpable, dramatic, nationwide turnaround.