LOL At Chris Cilliza-Random Walk-Obama Has A 98.35% Chance Of Winning OH
infernoman wrote on Nov 01, 2012 23:13:56 #
Continuing my own model of Ohio:
----
I include all live polls of LV's (using python+Mathematica), trim the top and bottom poll (in terms of O-R differential) for each target day, and add a random walk model on top
The most recent days incorporate about 13-14 polls each; basically, all live polls with sampling started & completed in the 12 days before the given day.
With 5.2 days left, current diff of 3.25%, and a daily standard deviation of 0.66%, a random walk model shows only a 1.65% chance of Obama losing OH.
No assumptions about ground game relative strength, LV models being too restrictive, any Bayesian priors, etc. Just the dirt-simple random walk.
Romney needs the OH state pollsters (all 10+ of them) to be systematically biased/wrong to win Ohio.
= new reply since forum marked as read
