Electoral Vote Predictor 2004:   Kerry 238   Bush 296

Data in Excel format Battleground states Polling data     Key Previous report Next report News   Strong Kerry (112) Weak Kerry (63) Barely Kerry (63) Exactly tied (4) Barely Bush (46) Weak Bush (123) Strong Bush (127) Needed to win: 270

Oct. 5

New polls: AL FL MI MN NH NJ NV OH OK PA

Welcome	FAQ	Donations	Electoral college
Electoral college graph	Projected final map	Cartograms	Movie
State poll graphs	State poll tables	EVs as bar charts	More data
Political humor	Political websites	Articles and books	Kerry and Bush records
Americans abroad	Foreign nationals	About RSS feeds	Enhancements

News from the Votemaster

American Research Group has more data on the debate. Kerry gained in all three panels they used. In the random telephone interviews, Kerry now leads Bush 49% to 46%.

The Gallup Poll puts them even at 49% to 49%, a big change from previous Gallup polls.

Zogby gives Bush a 1% lead. He also says that according to his poll taken Oct. 1-3, people think Kerry won the first debate by a margin of 2 to 1.

Pew puts Bush ahead 48% to 41% among registered voters.

CBS says they are tied. In short, we are in a very volatile period.

Not a lot of change in the state polls right now. I installed new software yesterday to average recent polls per state going back 30 days. There was a create hue and cry that this was a bad thing because the race is in such a state a flux that polls that old are meaningless. In the middle of the day I changed the look-back parameter to 7 days, so now only polls within the past week are used. I agree with many readers that 30 days is too much. If I were a Republican, I would have said "I am going to stay the course, even though I am wrong." Fortunately, I am a Democrat so I can flip-flop all I want. Let's see how a 7-day look-back window works. The idea was to prevent the map from changing wildly every day. Averaging introduces some stability, but also introduces lag in responding to current events. Finding the right parameter might be tricky.

Unfortunately, the new averaging software had a bad interaction with the software that makes the state graphs and predicted map and being in a big hurry to get it out the door does not help. It is being worked on. I'm sorry (Democrats also apologize when they mess up.)

Thanks to all of your generous donations, I am now running ads on the websites of a number of college newspapers, mostly in swing states. The current ads are at universities with a total enrollment of over 500,000 students. At this point the goal is to get the students enthusiastic about the election so they vote. Here is the current list:

I am also running ads on many blogs. See the donations page for more details. Realistic suggestions about where else to advertise are welcome. Sites other than blogs and college papers are possibilities. However "realistic" means (1) They take ads, and (2) you have checked on the price. College websites are on the order of several hundred dollars per month. That is doable. Thousands of dollars for an ad is not doable. Donations are still welcome. All of them go to more ads.

-- The votemaster

WWW www.electoral-vote.com