Wisconsin Recall: The adjusted Final Exit Poll was forced to match an unlikely recorded vote

Richard Charnin
June 6, 2012
Updated: July 11

The media and the exit pollsters have done it again.

Before the first votes were posted, the media reported that based on the exit polls, the election was “too close to call”. But Walker won by 53.2-46.3%, a 173,000 vote margin. Assuming “too close to call” meant that the exit poll indicated a 50/50 split, then there was a significant 7% discrepancy between the unadjusted exit poll and the recorded vote. I believe that Barrett was actually leading the exit polls. Of course, we will never know until the unadjusted exit polls are released. In any case, what caused the unknown red shift?

According to the Wisconsin True Vote Model , Barrett was a likely 54-46% winner. Barrett should have won easily – assuming the caveat of a fair election. But the election was very likely stolen.

Forcing the exit poll to match the recorded vote

The Final Wisconsin adjusted exit poll (2547 respondents) indicated that Walker had 53.0% (see the NY Times link below). The 0.2% difference between the Final and the recorded vote was the result of the standard policy of forcing the unadjusted poll to match the vote.

The pollsters claim that the exit poll had a 4.0% margin of error. But they can’t mean the final adjusted poll because it is always forced to match the recorded vote within 0.5%.

Why did the media not provide the actual unadjusted exit poll demographics? Was it because they knew that they would have to adjust all the crosstabs to match a rigged recorded vote – and did not want the public to view the “adjustments”?

The Fraud Factor

And as is always the case, there was no mention of the fraud factor in the mainstream media. There never is. To the exit pollsters and the media, there is no such thing as election fraud.

The GOP employs overt voter disenfranchisement in plain sight by robocalling voters with false information and having election workers discourage voters from using paper ballots and vote on unverifiable touchscreen DREs. But we are supposed to believe that right-wing voting machine manufacturers would not stoop so low as to write malicious code to covertly flip votes in cyberspace.

In 2010, Walker “won” by 52.2-46.6%, supposedly due to low-Democratic turnout.
Was the election a prologue of the recall?

In the recall, Democrats turned out in droves, they wanted Walker gone. There was no way that the unpopular Governor would match, much less exceed, his 2010 vote – if the votes were counted as cast. But that is a quaint notion considering the overwhelming statistical evidence of systemic election fraud since 1988.

Implausible 2008 returning voters and 2012 vote shares

Obama had a 56.2% recorded share in Wisconsin and 63.3% in the unadjusted exit poll (2.4% margin of error). Assuming Obama had a 60% True Vote share, then to match the recall vote, Walker needed the following:
1) 81% of McCain and 71% of Obama voters turned out.
2) He needed to win 25% of Obama and 95% of McCain voters.
3) He needed 46% of new voters who did not vote in 2010. The 2012 exit poll indicates he had 45% and that new voters comprised 13% of the total vote.

In order to win by his recorded vote, Walker needed a 10% advantage in returning 2008 voters and a 20% advantage in net defections. That is highly implausible.

Exit poll oddities

1) A full 5% of voters were not white or black. But their vote is n/a.
2) Philosophy: 13% of liberals voted for Walker?
3) Party ID: 34% Democrat/ 35% Republican in a progressive state?
4) Labor: Just 62% voted for Barrett?
5) Obama preferred by 51-44%, yet Barrett lost the recall by 53.2-46.3%?
6) Barrett only got 81% of would-be Obama voters?
7)Turnout:47% of recall were returning Walker 2010 and 34% Barrett? That’s a 13% difference. In 2010 Walker “won” by 52.2-46.6%.
8) Urban vote: Barrett had just 62% in big cities?

Margin of error?

The pollsters indicate that there were 2547 exit poll respondents and that the margin of error (MoE) was +/-4%. Presumably, this includes a 30% cluster factor.

The adjusted poll had a zero MoE since it was forced to match the recorded vote. What is the point of mentioning a MoE if the exit poll is adjusted to match the recorded vote?

The pollsters must be referring to the unadjusted exit poll, but of course that is not for public viewing. In any case, the 4.0% MoE is too high, considering the number of respondents (n).

The simple formula is: MoE =.98/sqrt(n). Adding a 30% cluster factor, the theoretical MoE is 2.6%= 1.3*.98/sqrt(2547). So how did the pollsters come up with the 4.0% MoE?

If we had unadjusted exit poll data, the margin of error would be applied to determine the interval where the vote share would be expected to fall 95% of the time. That’s why unadjusted exit polls are necessary. The standard practice of forcing the exit poll to match the recorded vote implicitly assumes zero fraud, i.e. the recorded vote is identical to the True Vote. It never is.

The Ultimate Smoking Gun: Unadjusted state presidential exit polls (1988-2008)

July 11 Update: There are 274 state exit polls listed in the Roper archive for 1988-2008 (only 24 are listed for 1988). I originally used the True Vote Model to estimate the 26 missing 1988 exit polls. Dr. Bob Fitrakis, writing in The Free Press, referred to the earlier probability calculations in an excellent article: Wisconsin: None Dare Call it Vote Rigging

Here are the revised numbers, based on 274 exit polls:

– Republican recorded presidential vote shares exceeded the corresponding unadjusted exit poll shares in 232 (85%) of the 274 state elections for which there is exit poll data. One would normally expect approximately 137 (50%). The probability is virtually ZERO.

– The exit poll margin of error (described below) was exceeded in 135 (49%) of the 274 polls. The statistical expectation is that the margin of error (MoE) would be exceeded in 14 (5%). The probability is ZERO.

– 131 of the 135 exit polls in which the MoE was exceeded moved to the recorded vote in favor of the Republican (the “red shift”).  There is a ZERO probability that this one-sided shift was due to chance. It is powerful evidence beyond any doubt of pervasive systemic election fraud.

– The Republicans won the recorded vote in 55 states in which the Democrats won the exit poll. Conversely, the Republicans lost the recorded vote in just two states (Iowa and Minnesota in 2000) in which they won the exit poll. If the elections were fair, the number of vote flips would be nearly equal. The probability of this disparity is virtually ZERO.

Calculating the probabilties

The probability P that 55 of 57 exit polls would flip from the Democrats in the exit polls to the Republicans in the recorded vote is given by the Binomial distribution:

P= 1-Binomdist(54,57,.5,true)
P= 1.13E-14 = 0.000000000000011 or 1 in 88 trillion!

The probability that the exit poll margin of error would be exceeded is 5% or 1 in 20. Therefore, approximately 14 of the 274 exit polls would be expected to exceed the margin of error, 7 for the Republican and 7 for the Democrat.

Given the relationship between the exit poll, margin of error and corresponding win probability, we compare the 274 state exit polls to the corresponding recorded votes. The Republicans did better in the recorded vote than in the exit polls in 232 of the 274 elections. The probability of this one-sided red-shift is 9E-35 or 1 in 100 billion trillion trillion.

The MoE was exceeded in 131 exit polls in favor of the Republican – and just 4 for the Democrat. The simple Poisson spreadsheet function calculates the probability P:
P = 3.74E-116 = Poisson (131, .025*274, false)
P = .0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 0000000000 00000374

1988-2008 Red-shift Summary (274 exit polls)
The following table lists the
a) Number of states in which the exit poll red-shifted to the Republican,
b) Number of states which red-shifted beyond the margin of error,
c) Probability of n states red-shifting beyond the MoE,
d) Democratic unadjusted aggregate state exit poll share,
e) Democratic recorded share,
f) Difference between Democratic exit poll and recorded share.

Year RS >MoE Probability.... Exit Vote Diff
1988 21.. 12... 2.5E-12..... 50.3 45.7 4.6 Dukakis may have won
1992 45.. 27... 1.1E-26..... 47.6 43.0 4.6 Clinton landslide
1996 44.. 19... 2.5E-15..... 52.6 49.3 3.3 Clinton landslide
2000 34.. 17... 4.9E-13..... 50.8 48.4 2.4 Gore win stolen
2004 42.. 23... 3.5E-20..... 51.1 48.3 2.8 Kerry landslide stolen
2008 46.. 37... 2.4E-39..... 58.0 52.9 5.1 Obama landslide denied

Total 232. 135. 3.7E-116…. 51.7 47.9 3.8 Exact match to the Nat Exit Poll

Note: 274 exit polls from 1988-2008
(24 in 1988, 50 in each of the 1992-2008 elections)

The conventional wisdom is very conventional – and very misleading:
http://www.huffingtonpost.com/2012/06/05/wisconsin-recall-vote_n_1572662.html

The NY Times Election site has the FINAL, adjusted exit poll crosstabs.
http://www.nytimes.com/interactive/2012/06/05/us/politics/wisconsin-recall-exit-polls.html