Anti-Climacus

WELL: I decided not to be a wuss and wait for the latest ARG tracking poll to make my prediction on Hew Hampshire. Here goes nothing:

Kerry: 41%
Clark: 19%
Edwards: 17%*
Dean: 16%
Lieberman: 6%
Kucinich: 1%

1. Methodology: look at the polls and guess what people are gonna do. Note that Clark has held strong at 18-20%, and Kerry's rise can be explained without recourse to his attracting undecideds, so there's no reason to give him an undue advantage there.

I feel reasonably enough confident in my model to suggest that the margin of error is +/-3% for all the candidates.

2. Assumptions: Dean's slid a lot lately, but he also hasn't done anything to damage himself since Monday night, so I'm assuming his floor is 15%. I don't think he's set to collapse entirely (he's going to finish ahead of Lieberman), but he could drop down as low as 11-12%. Those votes would mostly go to Kerry.

Edwards is going to end up with a reasonable portion of the currently undecideds. They're obviously not swayed by anything they've seen thusfar. Kerry will get about half because he's the frontrunner and they'll pile on, but most of the rest will go to Edwards because he's optimistic, and people like optimism.

Clark, Dean and Edwards are all within the margin of error of each other, so the potential for chaos abounds.

3. Big winner: Kerry. Not because he'll win going away (he will) but because Dean, Edwards and Clark will finish so close to each other that there'll be no clear second-placer, and none of them will drop out.

*my model predicted Edwards and Dean at 16%, but as ties are ways of wimping out of saying who'll win, I went with my gut and thought that Edwards would probably win the undecideds I gave to Clark and gave him an extra point.