Thursday, October 29, 2009

HORSERACE

Ah, Those Perennial Late-Breaking Undecideds . . .

A political science professor writes in:

There is a well-known principle in election polling — to the best of my knowledge, it was first enunciated in an article by pollster Nick Panagakis in 1989 — that in any election involving an incumbent, the odds are very strong that the undecided vote will break against the incumbent.  Specifically, Panagakis analyzed 155 races, and found that the plurality of undecideds voted against the incumbent 82 percent of the time, versus only 12 percent where the incumbent won a plurality.  The logic is that the incumbent tends to be much better known than the challenger, so if you're still undecided in the closing days of an election, you actually lean against the incumbent.  Assuming this to be true, it suggests that if the polls still show a dead heat by election day, Christie stands a very good chance of winning.

I'm familiar with that rule of thumb. But I recall back in the Kerry Spot days, a lot of Democrats insisting that Kerry would get 2/3 of the remaining undecided and only 1/3 would go to Bush, ensuring the Massachusetts senator's victory. But things didn't turn out that way: "John Kerry and George Bush split the undecideds evenly, 1.1 points each, from the average of the final polls."

I guess ever since then, I've been a little wary of attributing the undecideds one way or another. Does it make sense to me that the remaining undecided would prefer Christie or Daggett to Corzine? Sure. These voters know what they have gotten with Corzine: the worst state and local tax burden in the nation, higher tolls, residents fleeing, higher unemployment, a deep-rooted culture of corruption. If he hasn't closed the deal by now, it's not clear what he could do in the next four days to change people's minds.

There's also the wrinkle that if most remaining undecideds jump on the Chris Daggett bandwagon, Corzine could eke out a narrow win.

But there's a school of thought that undecided voters in late October are among the least informed, least aware, and least thoughtful voters out there; people who have tuned out the entirety of the campaigning for months, who didn't watch the debates, don't read or watch coverage, and can have their decision influenced by the most unlikely and random factors.

If the undecideds break heavily for Christie, it will be welcome news to the GOP. But I wouldn't quite count on it.

10/29 11:07 AMShare