Chances of Wisconsin Presidential Vote Results shown to be 1 in 850, and Worse for Other States

[Coyotl]

Chances of Wisconsin Presidential Vote Results shown to be 1 in 850, and Worse for Other States
November 28, 2016

The Columbia Free Press has published images showing the statistical likelihood of variance from the exit poll results that the reported vote counts represent. In the case of Wisconsin (Figure 1), Trump received 44.3% of the exit poll share, but won the state with 47.9%. The likelihood of such departure is 1 in 850, which is much less than 1% chance of happening, actually at 0.1% chance.



Pennsylvania is another state slate for a recount request ............

[jmg257]

1. What exit polls? The ones from CNN seem spot on.

[Land Shark]

3. Those are exit polls adjusted to match reported results (except votes counted after the election)

You need to use the unadjusted exit polls to make the comparison.

[jmg257]

4. Sheesh - thanks. How/why do they do that??? Where are the raw ones?

[Land Shark]

8. People screen capture them before adjusted. You can find election data here

Tdmsresearch.com

[jmg257]

12. Thanks - here is some onfo from Edison research who apparently does the actuall polling.

Still a bit confusing on them being updated after the fact.

http://www.edisonresearch.com/behind-numbers-2016-national-election-exit-poll/

[Land Shark]

16. You should be confused

Edison has some history of controversy in that they have argued that exit polls are not "intended" to.prove fraud therefore can't be used as such. But the US government regularly advises other countries to use exit polls this way.

And besides, if a.criminal doesn't intend to leave fingerprint evidence of a crime by grabbing a glass with their hand, does that mean the fingerprint can not be used as evidence of a crime?

[Igel]

28. Exit polls in other countries

that are used to expose fraud have a very large percentage of the voters surveyed.

In the US, they pick representative polling places and have a model as to who's voting. You take the numbers you got, plug them into the model, and out pops the prediction.

You poll one precinct, let's say. your "model" electorate" was
12% black (R-D split 10-90), 5% Lat. (20-80), 1% Asian (1-99), and the rest white (55-45). Your sample is
5% black (R-D split 1-99), 3% Lat. (5-95), 5% Asian (10-90), and the rest white (50-50% split).

Ooh, that's bad: your model is wrong.

At first you take the results for the voters assuming your model is correct, because the model isn't for "all those voting by noon" but "all those voting by the end of the day". But then you find that for 100 votes you assumed 10 of the blacks vote (D) and only 1 (R) when actually they accounted for only 5 (D) votes. You thought 3 (D) Latino votes, but instead get 1 (R) and 4 (D)--you're down one (R) vote but 4 (D) votes. Instead of one (D) vote you get effectively none from the Asians, but the white vote netted more (R) than you thought.

When you finally know what the electorate is like, you can either insist that reality is wrong and your fantasy model is right, or you can adjust your model for (a) turnout, (b) skew in the demographics, (c) wrong assumptions about how each demographic would vote. So, for example, in Texas the surprise was how many Latinos voted for Trump. He didn't get the percentage Romney got, but given the rhetoric the assumption was he'd be in the single digits for Latino support. He got something like 20% of the Latino vote.

This is a work in progress on election day, and the big question after recalibrating the model to match reality is always, "So, how can we improve our model next time?"

Note that the "catch election fraud" tends to get over 90% of the population polled, and they don't make predictions based on models. They actually count votes and fill in those areas missed *after* the fact based on reality.

You can run lots of valid tests based on those statistics. But when you run tests based on stats that result from a small sample with an assumed model, your tests say much more about the model used than the actual data. In other words, there's a very, very small likelihood that the model the exit pollers used was correct.

[Land Shark]

29. So you're thinking exit pollsters are pretty stupid then

Since knowing who the electorate isand how they voted is one of the key services they are trying to sell, and you're saying that none of this is reflected whatsoever in the margin of error and any adjustment to the same based on the clustering effect, which does seem to be an adjustment that reflects the concerns you're mentioning

The forced match to election results is just making things up though.

[tandem5]

9. Where are the raw ones? That question has been hovering in the air since at least 2000.

[Land Shark]

13. Election data and analysis without opinion published at tdmsresearch.com

[imaginary girl]

5. the exit polls you are looking at have been "corrected" to correlate with the "results"

These stats are regarding the original, uncorrected, poll results. I think someone mentioned that in another discussion already, but just wanted to correct the record for anyone reading your comment who may not know that.

[jmg257]

6. Yes I have seen that too...Who/why do they do that? Weird. nt

[Land Shark]

10. A gap between exit polls and Election results beyond margin of error means one of them is wrong

So, they simply assume the exit poll is wrong and adjust it. Not sound scientific or statistical practice.

[Land Shark]

11. The kicker is that since 2000 the exits are always red-shifted (election results better for Rs)

And the size of the red shift keeps getting larger.....

The states of WI, PA, FL and NC were all.well outside the margin of error. So we're a bunch of senate races, especially Missouri. The biggest red shift on the presidency was over 10 points in Utah, where trump.desperately needed to beat both McMullin and Clinton in a tight 3 way race.

[jmg257]

14. That is just crazy - what's the point then? They just want to get an idea of trends? Not

actual "results"? Crazy.

[Land Shark]

17. They want to create the fodder for those who read political tea leaves

All the endless often pointless analysis

[imaginary girl]

25. I don't know (n/t)

[Coyotl]

22. Source: CNN.com. Compiled by Jonathan D. Simon election night, unadjusted numbers.

https://groups.google.com/forum/#!topic/electionintegrity/-LpcH9z5tKg

2004 Kerry Exit Data Source: Dr. Steve Freeman, compiled election night, unadjusted numbers.

[gordianot]

2. I do not expect a different outcome.

It is heartening to see all of the Republican angst over the recount.

[Land Shark]

7. People protect themselves from disappointment. Nobody knows what the ballots really say

Until they are.counted with human witnesses.

And those paper ballots mean nothing when they fight so hard against recounts every time, regardless of margins. Until there is a hand recount, only machines have handled ballots and nobody really knows anything.

[Land Shark]

15. A few factors militate against successful recounts, like....

Nineteen counties in Wisconsin refused hand recounts and do machine recounts. That ought to be the same result under most conditions and is not a meaningful recount because it doesn't adjust for mistakes possible in the original method. There is also the amount of time that goes by. Last presidential recount was Ohio 2004 and three election officials were convicted of rigging it to match the original count. (Huge nonpartisan motive: avoiding embarrassment and avoiding being "the next Florida 2000.&quot That was actually a Democratic county of Cuyahoga rigging it to be the same, which helped Bush.

But aside from these non-ballot extraneous factors, anyone who says the recount could never change anything is either high or much worse. They simply can not know what the paper ballots say. Nobody knows until a recount by hand is done. And unlike hand counting errors that are rare and I very small portions, a machine can make huge errors and never "know" the difference. As long as the error doesn't reach the level of absurdity, election officials will just sign off on.it as the certified result.

[Coyotl]

20. The Cuyahoga Election Board chairman was the head of the Ohio Republican Party in 2004.

In other words, the Republican boss was the guy counting one of every eight Ohio Democrat votes. He was fired by the Sec. of State eventually, but not charged with any felonies. Several of his elections workers were tried and later got off on retrial.

[hueymahl]

18. Garbage in-garbage out

Flawed statistical analysis because the validity of the data is untrustworthy.

[triron]

19. Was a similar anylysis done for Pa, Mi, Fla., and N.C.?

[Coyotl]

21. Yes, same link.

[triron]

23. I found Pa and N. C.

but no other swing states.

[Coyotl]

30. We have the numbers of respondents for each state and the formula for MoE here:

https://groups.google.com/forum/#!topic/ElectionIntegrity/
Compiled by Jonathan D. Simon www.CodeRed2016.com
https://groups.google.com/forum/#!topic/electionintegrity/-LpcH9z5tKg


The battleground states have near the same robust polling so you can compare the %'s and have a good idea in relation to the three charts we have. The less contested the state, the wider the margin of errors are (fewer respondents).

Or, you can calculate them individually:
National Vote 21,753
AZ 1729
CA 2282
CO 1335
FL 3941
GA 2611
IA 2941
IL 802
IN 1753
KY 1070
ME 1371
MI 2774
MN 1583
MO 1648
NC 3967
NH 2702
NJ 1590
NM 1948
NV 2418
NV 1352
OH 3190
OR 1128
PA 2613
SC 876
TX 2610
UT 1171
VA 2866
WA 1024
WI 2981

[triron]

24. I did a very rough back-of-the-envelope

calculation on the chance of the deviation toward Trump for all the exit poll vs actual vote results and assuming only a binary value (toward Trump or toward Clinton) and the probability was less than 1 in 13000 that the vote turned out as lopsided as it did.

[Coyotl]

26. Indeed. Each exit poll is a separate poll. What are the odds they all will shift red?

And what are the odds they all shift red moire than twice as much as 2004, when we had the same Senate seats up for re-election?[center]

[Coyotl]

27. Zero states split their tickets for president and Senate for the first time in American history

Zero states split their tickets for president and Senate for the first time in American history
By Stephen Wolf Dec 01, 2016

MANCHESTER, NH - OCTOBER 24: Democratic presidential nominee former Secretary of State Hillary Clinton (R) and New Hampshire Gov. Maggie Hassan look on during a campaign rally at Saint Anselm College on October 24, 2016 in Manchester, New Hampshire. With just over two weeks to go until the election, Hillary Clinton is campaigning in

Americans in every state began directly electing their senators in 1914 following the passage of the 17th Amendment. The next 25 presidential elections, from 1916 to 2012, always saw at least some states vote for candidates of different parties for president and Senate. That long streak finally came to an end in 2016, when every single state voted for the same party for both the presidency and the Senate. Democrats won 12 Senate seats, all in states that Hillary Clinton carried, and Republicans won 21 Senate contests, all in states where Donald Trump prevailed. Republicans are also strongly favored to win a December runoff in one more Trump state, Louisiana.

You can see this remarkable set of results in sharp relief in the graph below:



.................

Exit polls are noteworthy in this regard also, particularly so because they deviate to greater red shifts in accord with Republican need to win Senate seats.



The 20 contested Senate states have an extra 1% red shift compared to the mean of the 29 states polled nationally. Those without contested Senate seats average 2.4%.

[triron]

31. knr