Jay Cost's poll calculation and EV analysis today is VERY interesting.
Nationwide
Bush: 49.57%
Kerry: 46.28%
MOE: +/- 0.9%
(Respondents: 8,954; Polls Used: ABC News, Fox News, LA Times, Gallup, Newsweek, Battleground)
Based on these results, we can be 99.9997% confident that Bush presently has a lead.
Present Probability that Bush will win the Electoral College: 96.36% (This is the probability that Bush wins FL and IA and WI or OH. Thus, we can be 96.36% confident that Bush would receive a minimum of either 271 EVs or 281 EVs).
Evaluation: Bush looks to be in very strong shape in IA and FL. At this point, his numbers are inching closer and closer to 50%. His numbers in Ohio are coming back to their levels from mid-October. I believe this is due mostly to bad polls replacing good polls and then being replaced in turn by good polls. He still likely retains a lead in WI, but the Mason-Dixon numbers are somewhat unsettling (though they are mitigated by the MN numbers, which in this post are skewed by an outlying survey from St. Cloud State).
Bush also seems to be inching upward in PA, due to a string of polls calling the race a tie (Quinippiac even has him up), which might result in a big pro-Bush surprise on Election Night. Gallup should release another PA poll sometime before 11-02. That should be the one to look for.
Right now the EV math is looking awfully tough for Kerry. He is definitely behind in FL, IA and, though I do not cover it here, NM. This gives Bush a minimum of 266 EVs. Plus, Bush is likely leading in OH and WI -- and I think Kerry will be unable to hold MN when all is said and done. The word on the ground is that BC04's organization in MN is a sight to behold. The big question on my mind right now is not whether Bush gets to 269, but whether he breaks 300 (which he would do if he carries FL, IA, NM, WI, OH and MN -- that would be 306).
http://jaycost.blogspot.com/
That's very interesting. My calculation of the accumualtion of state-by-state polls (mostly using Mason-Dixon, SurveyUSA, Strategic Vision, and ARG) is as follows:
Bush 48.90%
Kerry 45.73%
Turnout 110,000,000
Margin 3,500,000 for Bush
(I projected turnout in 2004 by increasing the total of voters by the growth in population in each state, then multiplying the state polls by that number.)
That implies that about 0.6% of each candidates support level in the national polls is not being caught yet in the state polls. But otherwise, it shows an identical 3.25% race.