http://synapse.princeton.edu/~sam/pollcalc.htmlThis fellow, Prof. Sam Wang of Princeton University, does an interesting electoral-college meta analysis, which is now showing Bush ahead by 0.3% and a much-improved electoral map.
Below is a meta-analysis directed at the question of who would win the Electoral College if the election were held today. Meta-analysis provides more objectivity and precision than looking at one or a few polls, and in the case of election prediction gives a more accurate current snapshot. Many of you agree - this site gets thousands of visitors every day.
These calculations are based on all available state polls, with an emphasis on likely voter data that include Nader where he is on the ballot. The three most recent polls for each state are averaged and the standard error of the mean is used to calculate the probability of every combination of possible state results. See below for an explanation of the methods. Your polling-related comments are welcome.
<snip>
A strong predictor of victory is winning two or three of the following states: FL, OH, and PA. Pollsters know this and so poll there frequently. To improve the confidence of my estimate, in those states (and others where data are available) I will start averaging three polls or more, and include all polls completed in the last seven days. This improvement comes at minimal expense in time resolution. This is very different from the electoral-vote.com approach, which uses one poll per state and is therefore very noisy. Today we have 8 FL polls, 5 OH polls, and 6 PA polls. Kerry is ahead in PA, Bush is ahead in FL, and OH is tied