(Yesterdays numbers in parenthesis. MoE - 100)
Dean: 100.0 (100.0)
Clark: 60.9 (61.3)
Lieberman: 56.9 (54.3)
Gephardt: 50.4 (48.6)
Kerry: 49 (46.9)
Edwards: 26.9 (28.6)
Sharpton: 14.1 (13.6)
Kucinich: 9.3 (9.7)
Moseley-Braun: 9.1 (8.6)
New info: Zogby NH poll, ARG NH, new OK poll, media results for 12/3
Old info: previous ARG NH, previous OK, media results for 11/26
As often happens when a significant amount of new info is added into the ECC, little changes. Kerry seemed destined to fall off the face of the earth after the NH polls, but the OK helped him back up and pushed Clark down from the mid-60's. Also, the media results were not favorable to Clark yesterday.
There will be no ECC tomorrow or Saturday, as I will be in CT.
My theory behind the ECC is to take as many factors into account as possible, thus producing the broadest view of the campaign at any given time. Many other predictive models have failed in the past because they fixate on one or two causes behind election results. The way I figure it, by trying to avoid elevating specific causes (NH results, Gallup results, economic fugures, whether or not the candidate is from the south), I can some what sidestep attributing cause altogether. Rather than having one or two causes for campaign success, I have chosen to combine four generally accepted models (early state / momentum, national polls, money and delegate counts) and then combine them. Although I give preference to the early state / momentum theory, (my version of that theory, specific to this campaign, constitutes half of the ECC), largely because it has so much currency here on dkos, hopefully this broad will prevent me from thinking I have the most accurate model around. Equally important, hopefully it will allow me to offer a very good guess at any given point as to how the campaign is progressing.