9/8 Election Model update: Kerry 296 EV, 88% win prob. ICR poll included.

[TruthIsAll]

I have included the ICR national poll in the National model (16 polls). ICR has Kerry ahead by 47-46 (RV). I use Registered Voters (RV) polls, rather than Likely Voter (LV).

The update fixes a typo: the latest Zogby WI poll was incorrect (48-48); it's now Kerry by 50-48. This increases his win probability slightly (1-2%).

Every little bit counts.

http://www.geocities.com/electionmodel/

[demosincebirth]

1. The media falling all over the Newsweek and Gallup poll.

Don't they do research? Some times I wonder

[TruthIsAll]

2. We have come to this truth: don't believe, don't trust the media.

Especially the cable whores. They lie, they misrepresent, they omit, they are lazy, they are bought, they are spineless, they are...traitors.

There are still a number of good old-fashioned print reporters out there (Krugman, Palast, Breslin, etc.). And there are even a few on cable. But most have gone over to the dark side.

My goal in developing this model is to cut through the media BS by using a large number of national and state polls for the respective models.

The national vote percentages (11/16 poll averages) are very close to the equivalent weighted state percentages which are added together. This gives me confidence in the models.

The variation in probabilities between the state and national models are due to the different methods used and definition of a "win".

The national probability is the probability of a national majority popular vote (exceeds 50%).

The state probability is the result of a Monte Carlo simulation to calculate the number of times Kerry gets over 269 EV in 5000 election trials. So the probability is the number of wins/5000.

[TruthIsAll]

3. AM kick

tia

[oasis]

4. Kerry 50-48. A reasonable figure. Keep it coming.