Monday, November 05, 2012

What if Gallup is right?

I'm a "science" guy. I tend to accept polls as being the distilled magic of statistics. The battleground state polls have shown Obama leading throughout this campaign, so, all things being equal, I rate Obama the favorite.

On the other hand, the polls are screwy. On the third hand, there is typically a word for people who spend their time complaining about the polls - "losers."

On yet another hand, there is this perspective offered by Bob Krumm:

Meanwhile, in what should have sent up alarm bells across the land, Barack Obama’s campaign was diminished to arguing that it would win on the strength of its get-out-the-vote efforts with youth and minorities. Second to mentioning Harry Truman, this is the last refuge of the losing side. Almost never is depending on sporadic voters a recipe for success. Even in 2008, Obama’s leads with those demographics only padded the margin he had already won because of his eight point lead with independents. Falling precipitously from his earlier indy numbers, he and his acolytes should have known that calamity lay ahead.

Most pollsters took a beating in predicting the race. But that should have been expected too. The last two times that a Republican challenged a Democratic incumbent (1996 and 1980) the polls overestimated Democratic support by 5.1 and 7.2 points. And ‘96 was not even in bad economic times.

The biggest gap was not one of gender, but between the opinions of those who voted and those who did not. Those most likely to vote–whites, property owners, investor class, church-goers, married, and the elderly–despised where Obama had taken the country by margins of sometimes more than 20 points. Singles, minorities, and the youth still backed the President, but with diminished support and turnout from four years before.

The entire outcome hinged on who turned out to vote, and when only nine percent of the electorate wished to talk to pollsters, projecting who that was likely to be, became a fool’s errand. But statisticians are not above being fooled. A look at any one of those demographics–even the ones that still supported him–would have showed the President’s downward shift from four years before. But it was too easy to weight bad samples toward a guesstimate of a turnout model instead. Only Gallup got it right, and the result was nasty for Democrats. They went from a 12-point advantage in party identification among likely voters in 2008, to a one-point disadvantage in just four years.

That may be the model that is informing George Will and Michael Barone's predictions.

No comments:

Who links to me?