Probability of Bush Losing: 98%

[Tempest]

http://synapse.princeton.edu/~sam/pollcalc.html

Likelihood analysis of all possible outcomes of election based on state-by-state polls

Calculations by Prof. Sam Wang, Princeton University. Your comments, including reports of possible errors, are welcome.

I originally did this calculation to help think about how to allocate my campaign contributions. I believe that one can make the biggest difference by donating at the margin, where probabilities for success are 20-80%. To read a discussion click here. Since trends are now strongly in Kerry's favor, I recommend that Democrats give to the Senate campaigns of Tony Knowles (D-AK) and Inez Tenenbaum (D-SC) and to the DSCC. For those of you still nervous about the national election (see the bias analysis below), I recommend the voter registration and turnout organization America Coming Together. For the optimists there is the DCCC.

This calculation is a snapshot in time. It takes a large number of state polls and calculates the likeliest number of electoral votes if the election were held today and reflected those polls.


Results as of Sunday, July 18, 2004 based on state polls:

Counting the last six polls the current probability of a Kerry win is 98 percent. Counting only the last three polls the probability is 99.98 percent.

270 electoral votes (EV) are needed to win. The current 50th percentile outcome for Kerry is 296 EV. The 95 percent confidence band is 270-322 EV.

This calculation would be affected by an overall poll bias. This could happen if sentiments have changed since the time of the polls or if a discrepancy exists between state polling methods and actual voting patterns. For the last three polls the effects of bias favoring Bush are as follows. 2 points: Kerry win 95%, 50th percentile 287 EVs (95% confidence 264-302). 4 points: 26%, 279 EVs (222-297). Note the strong threshold phenomenon between 2 and 4 points.

Current probabilities by state (rounded to the nearest percentage point): AR 15, AZ 0, FL 98, IA 100, ME 98, MI 100, MN 100, MO 11, NV 25, NH 96, NM 100, OH 26, OR 100, PA 89, WA 100, WV 18, WI 88. The calculation is currently dominated by the large states FL (98%), PA (89%), and OH (26%). In most high-likelihood scenarios, two of these are needed to get over 270 EV (my earlier posting left out PA).

Added comment: the current snapshot suggests that the threshold state is PA, where Kerry leads in 11 of the last 12 reputable polls since May 1. There is a rank ordering effect, in the sense that states tend to move together. The rank order (using the mean margin divided by the SEM) among big swing states is currently
D <-- MN - MI - FL - PA - WI - OH - MO --> R

[madrchsod]

1. 2%

diebold and their band of brothers-results 51% bush.

[goclark]

8. Diebold is the key to this election

We can be as confident as we want to be - the fix is in.
The Rethugs have Diebold and they will win unless we can get bold.

[cliss]

2. Bush is tired.

I've seen photos of him lately. The summer is wearing on, and all he really wants is to go home, pop open a cold one, and kick back. He's tired of all the bad news, the regression analyses, the t-factor statistical probabilities.

It's an uphill battle, but for what? He's not sure of the reasons anymore.

[drmom]

5. Do you think W even understands what a regression analysis is?

It's very hard to imagine any type of critical thinking coming from him. But, I do agree he would like to strip down to his underwear, grab a couple of beers and watch some NASCAR.

[kodi]

6. and we are tired of bush. he's old now and change is wanted

.

[shivaji]

3. Prof Wang has failed to take into account the most important

factor, DIEBOLD! During my 35 years of living in Chicago, no
matter what the polls said, mayor Daley's machine always won
by a comfortable margin. It is not who votes, it is who counts
the votes that determines the winner.

[Feanorcurufinwe]

4. And something else -- coattails

if there is a strong Kerry surge it will help all Democrats, everywhere.

[PROGRESSIVE1]

7. Great to hear!