Looking in on the "I'll show you nerds how to math" theme, James Fallows points out the wide gap between what pollsters, poli-sci types, and poll aggregators are thinking, and what Republicans are expecting. Fallows concludes:
Someone is out of touch with reality here, and in a more fundamental way than I can recall.The point is not that an Obama win would "prove" Rove and Gingrich wrong, or a Romney win would "disprove" the state-poll models. We're talking about probabilities, not certainty. (To spell it out: A tossed coin has a 50 percent probability of coming up heads. The fact that it comes up tails doesn't "disprove" that probability.) But this is not like the normal closely-run election, in which both sides are saying, "It will be close, but I think it will turn out our way." Nor is it like the normal impending landslide, in which one side maintains a brave face but knows how things are headed.