The Winning McCain unleashes an invective or two to remind us that we need to run through the tape at the close of polling places today (that’s right, there’s a language warning on a McCain post, and it’s not from Meggie Mac):
Polls don’t win elections, and regression analysis sure as hell doesn’t win elections. Politics is not a science, and trying to reduce elections to trends, polls and mathematical formulae is one of those situations where when the only tool you’ve got is a hammer, every problem looks like a nail.
If you’re a thousand miles away from a district and don’t actually know anything about what the candidates and their campaigns are doing, it is tempting to look at poll numbers, examine past voting trends, and start making assumptions about what the result will be. But when we yield to that temptation, we ignore the Hayekian insight: Information is diffused throughout society in such a way that no one — not even the best-informed “expert” — can know everything.
So Jay Cost doesn’t know everything, Charlie Cook doesn’t know everything and Michael Barone doesn’t know everything, either. Yet their status as political experts requires them to make predictions and we mere mortals . . . well, we don’t know nothing about winning no elections.
In short – until and unless you, and those you know, vote, the Democrats are still in charge.
