Odds of the Red Shift: Thanks to the DUers who checked the calculations.

[TruthIsAll]

Thanks to those Math DUers who responded to my post to confirm
or reject the use of the Binomial Distribution probability
calculation for the Red Shift from the Exit Polls. 

The Probability that at least 41 states would deviate from the
Exit Polls to Bush is given by:			

Probability	=1-BINOMDIST(N-1,51,0.5,TRUE)	
P = .000007			
The Odds: 1 in 135,706	
I originally had it as 1 in 1.7 million.  		
			
Here are the probabilities of N states moving from the exit
polls, without regard to the MOE, in favor of Bush. 
			
N	Probability	Odds: 1 out of	
26	0.500000	2	
28	0.287925	3	
30	0.131219	8	
35	0.005487	182	
38	0.000311	3,220	
41	0.000007	135,706	< Actual odds
42	0.000002	590,175	
44	0.000000	16,508,110	
45	0.000000	109,148,693	
49	0.000000	1,698,830,489,389	
50	0.000000	44,590,095,320,500	
51	0.000000	#DIV/0!	Cannot compute
			
***********************************************************	
For the Probability that at least 16 states would deviate
beyond the Exit Poll MOE to Bush: Originally, I had it as 1 in
4.5 Billion. Wrong. I entered 5% as the probability that in a
given state the MOE would be exceeded. It should have been
2.5%. 

Making this change, the odds exploded to 1 in 200 TRILLION.
This was was off also, but not by all that much.  As MathGuy
pointed out, we need to calculate the probability that the
Bush vote tallies would exceed the MOE in AT LEAST 16 states. 

This is just 1 - the complement probability: 
P = 1 - the probability that 15 or FEWER states would exceed
the MOE.

The Probability, assuming a 3% MOE, is 7.382983E-14.			
So the Odds are 1 in 13.5 TRILLION. That's it. No further
changes. I promise. 
			
If we consider that Exit polls are more accurate than public
polls, and assume a 2% MOE, then for 23 states to deviate
beyond 2% the Probability is ZERO. Excel can't compute a
probability when N reaches 18.			

Probability	=1-BINOMDIST(N-1,51,0.025,TRUE)		
			
N	Probability	Odds: 1 out of	
2	0.365526770	3	
4	0.038533663	26	
6	0.001676323	597	
8	0.000037195	26,885	
10	0.000000477	2,098,096	
12	3.816504E-09	262,019,924	
14	2.014000E-11	49,652,431,051	
16	7.382983E-14	13,544,660,533,445	 Actual odds
17	4.996004E-15	200,159,983,438,689	
18	0.000000E+00	#DIV/0!	Cannot compute
			
			

[cornermouse]

1. And in plain english for those of us who

don't have your ability in math and statistics?

[StClone]

2. If you accept it without knowing probability math

If we consider that Exit polls are more accurate than public polls, and assume a 2% MOE, then for 23 states to deviate beyond 2% the Probability is ZERO. Excel can't compute the odds when N reached 18.

[cornermouse]

3. Thank you.

[TruthIsAll]

4. In plain English, the deviations from the exit polls to the vote tallies

for Bush could not have been due to chance.

It was mistabulation or fraud.
Take your pick.

[cornermouse]

5. Thank you.

Some of us think that numbers are next door to meaningless, even though they aren't.

[Duncan]

6. Keep It Up TIA. Thanks. (n/t)

[platinumman]

7. Your assumptions are wrong

As has been said many times in past threads, your maths may be right, but your assumptions are not, particularly your assumptions of randomness, which make your maths irrelevant. You are diverting attention away from what may be real proof of fraud. By way of explanation, if you are interested, I will repeat here what I wrote in a previous thread started by you.

Any analysis from the figures that are available from the exit polls are meaningless if we do not know the design of the poll. Many people, understandably, have compared the US exit polls with those of the Ukraine. However, the Ukraine polls were designed to check the integrity of the election. I think there should have been such a poll in the USA, but there wasn't. The US Polls were not that at all, but a commercial undertaking to provide a degree of prediction to the media, post-election analysis, and, of course, profit to the pollsters. And they had to meet these aims within their budget.

Let's have a look at your own exit poll. One that I am sure you carried out. Let's say that among your acquaintances, you would expect 20% to vote for Bush, or maybe 20% voted for Bush last time round. (yes, I doubt if you even know any Bush voters, but never mind). Now, if you asked your acquaintances on the evening of the 2nd who they voted for, and you found out that 30% had voted for Bush, you should start worrying that Bush had won the entire election, even though, in your own little poll, Kerry was winning easily.

That is more or less how the Mitofsky poll might have worked. (I say 'might', as we don't know, but it is how such polls are conducted in the UK, and it is a straight, possible, explanation to the figures that we have). They are not going to send pollsters out into the countryside, as they would be standing around doing nothing most of the time. Instead it is far more efficient to keep them in the urban areas of most states. They know that these are mainly democratic voters, and the figures obtained would reflect that fact. However, the important figure is the 'swing': in other words, how much the measured lead of the Democrats is greater or less than the lead which would be expected in those areas to win the election. The measured lead may be 3%, but if one would expect 5% in that area, then that 3% would mean a prediction of a Republican win. However, it would take time to work out, and margin of error is necessarily far larger than that which would be obtained by pure one-stage random sampling, but it is enough for the temporary needs of that day. (If you think that such a predictive poll is a waste of time, I would not disagree with you: it serves commercial purposes only)

Therefore, as I said, any analysis without knowing the design is spurious. You may well ask why Mitofsky does not produce the design. They, or the networks who employed them certainly screwed up the presentation of the polls, and they should be professionally embarrassed. There is a lot of competition in the polling field, so it may be that he is keeping it secret as a matter of safe principle. However, a lot of the statements that pollsters have made by way of explanation are, as has been rightly noted, a bit ridiculous, and it would be best if they just came out and told us the design, but they can't because it is owned by who paid for it. Of course, if there has been fraud (and I have to think there was) one would hope that it does show up in the exit polls, but one cannot be sure of that, as the swing necessary to return the election to Kerry would probably be less than the margin of error.

However, there is a simple explanation for the odd figures, which they can trot out at any time and I fear that if you constantly bring up this red herring, then, in the eye of the public, the stronger claims of fraud proof will be damaged. Such important claims are the lack of voting facilities and the clear opportunities to corrupt the counting software. Given time, such claims are verifiable, through inspection of documents, regression analysis, and, over time, auditing and compulsory inspection of source codes. I'm sure all these things will happen, but it is a waste of time, for now at least, to try to claim proof from the exit polls.

[jkd]

8. The problem with your analysis

I made this post before on another thread, but nobody chose to answer. I'll try again?

Concerning BYU’s exit poll accuracy, Dr. Freeman said this:

“True to their word, predictions in this year’s contests were quite accurate. In the Utah presidential election, for example, they predicted Bush 70.8%, and Kerry 26.5%. The actual was Bush 71.1%, Kerry 26.4%. Consistently accurate exit poll predictions from student volunteers, including in this presidential election, gives us good reason to presume valid data from the world’s most professional exit poll enterprise.”

Mitofsky had Kerry at 30.5%. Isn’t Dr. Freeman contradicting his own argument?

New Hampshire had the largest MOE and the exit polls were wrong. I believe Utah's exit poll was wrong. Please figure the odds for me that both Utah and New Hampshire exceeded the MOE. I'm not a mathematician. When both sides of a question appear unlikely, there must be something wrong with the question.

[platinumman]

9. Dr. Freeman is watching you.

The last person I read who took a stand against this exit poll argument was told that such a great statistician as Dr Steven Freeman started the theory in a paper published shortly after the election, and who was he to disagree?

I was intrigued, as I had never heard of any Dr Steven Freeman, so I read his paper, then went to his departmental web page, at
<http://www.organizationaldynamics.upenn.edu/center/nav.cgi?page=freeman>
and also his own home page, at
<http://www.appliedresearch.us/sf/>.

I discovered that he is certainly not a statistician, and not even a mathematician. He does not appear to have any training or experience in either academic statistics or in commercial polling techniques. His field is management, business administration, and entrepreneurship. He's written some interesting stuff, but all of it is miles away from statistics and polling.

At the end of his paper on exit polls, he asks, like you, for peers to check his paper, but nobody with any academic position in statistics would bother with what is essentially a very basic entry level paper. It is something I might expect from first year undergraduate students. I might forgive them for overlooking facts that they could glean from a basic polling manual, but this guy is a PhD. This is most certainly not a post-doctoral paper.

You may wonder why he wrote it, as the paper looks distinctly out of place in his CV, and certainly won't get him any promotions. And I have to say, he doesn't really look like a political activist. Then, when you read his personal home page, you realise what he is up to. His special research areas are loss, change, and resilience. His JOB (not his interest, but his JOB) is "studying the effects of loss and adversity on organizations and individuals".

So, guys, he is studying you. You are his guinea pigs. You have suffered a loss, and he is studying how you respond to it. He has dropped a pebble in the water and watching the ripples. I think it is highly unethical, and methodologically unsound, but I suppose it is a rat race in academia.

And to preempt those flames:

No I'm not a Democrat. I'm a Socialist.
No I'm not American. I'm British.
If I were American, I would vote for anyone committed to reform.

Yes, I think you are quite correct to be crying fraud, as the whole system needs to be changed. My personal hatred (apart from the basic inability to count votes) is the inequitable distribution of voting machines. That is something from the nineteenth century, but what can we expect from an eighteenth century constitution?

Oh, and I think you are being undermined by people with unsustainable theories.

[TruthIsAll]

10. For whatever its worth....

Are you now resorting to bad-mouthing the messengers?

I have 10,000 posts on this forum.
How many do you have?
Why should anyone believe you?

I also have THREE degrees in applied mathematics and have been developing, designing, programming and implementing mathematical models for many years - in aerospace, engeering, communications, finance and investments.

What's your background? How many models have you developed?

My "customers" have always been satisfied with the results. I "walk-the-walk", while others who seek to belittle my efforts just "talk-the-talk". They never show their work. Because they haven't done any of their own.

And no one has been able to refute the analysis. Because they cannot.
In fact, they now agree the mathematics (don't call it "maths") is correct. So they need to raise bogus strawmen about exit poll sampling error and statistical independence and voter psychology and on and on and on..

And yet they ignore the documented incidents which all favor Bush.
"I PRESSED KERRY AND IT CAME UP BUSH" - 86 OUT OF 88
Documented.
"I PRESSED BUSH AND IT CAME UP KERRY" - 2 OUT OF 88
Documented.

How come, all of a sudden, tried and true polling methodology is being questioned? The methodology and science has been around for centuries. It's Probability and Statistics 101.

Sampling error? It's called The Central Limit Theorem. It's called the standard deviation. It's from Bernoulli. It's from Gauss. And countless others who have laid the groundwork.

Statistical independence? Exit polling 41 out of 50 states is as independent as you can get - unless you need to fudge the results after the fact. The fudging of the nationwide results is definitely NOT independent.

Those who ignore the facts of this election also ignore the facts of history. And those who choose to ignore are doomed to repeat them.

Yes, there has always been fraud.

BUT THE FRAUD IN THIS ELECTION AND 2000/2002 IS DIFFERENT BOTH IN KIND AND IN MAGNITUDE. IT'S NOT SUBTLE FRAUD. IT'S MASSIVE IN-YOUR-FACE FRAUD. ARE YOU FAMILIAR WITH FLORIDA 2000? OR GEORGIA 2002? OR OHIO 2004?

ALL DOCUMENTED.

And finally, the analysis is MY OWN. No one is telling this analyst what to think, what to say, what to spin - or how to compute.

No one is paying me a dime. I don't do this for the money.
I do it becomes no one else, except for a few, has.

I will not wait for the academics or reporters to do their jobs.
After all, they don't want to lose them.

[Vinnie From Indy]

11. Rock On TIA!

"I would have been a doctor, except for all that science stuff they wanted you to learn" Woody Harrelson

I guess they weren't lying to me in school when they said math might some day save my life (or in this case my country).

[RubyCat]

12. Your analysis of the election results is better than the Mystery Pollsters

or anyone else that I've seen. They can't attack you on the math so they find something else to attack you with. Any day now I'd expect them to say "Bernoulli? Gauss? What, is he practicing gay math?"

[Roger_Otip]

13. platinumman makes some good points

i don't think anyone doubts your motivation or mathematical ability, TruthIsAll, but i think most people - and not just people on DU - accept that the deviations of the exit polls from the vote were not down to random error. there was clearly something systematic going on - where people disagree is on what that was. the options i've heard are:

1) vote rigging
2) exit poll rigging
3) bush voters lied or refused to take part in polls
4) poll methodology - either deliberately oversampling democrats in order to weight results later, or a flawed oversampling of democrats

maybe there are others. i'd rather hear what people think about these various explanations, what evidence there is for and against each, whether the truth could perhaps be a combination of some or all of these.

[senseandsensibility]

14. Regarding "maths"

I believe the poster is British, and that's how they refer to mathmatics, or math. I've heard it before (From Fergie of all people), but it does sound very strange to American ears. However, I agree with your post, TIA. Everything you say is backed up with the mathmatics to support it, and, as you say, you are not being paid by anyone.

[platinumman]

15. In response to your questions and points

In answer to your points

a) Why does it matter how many messages you have posted?

b) I'm not asking anybody to 'believe me'. Just to read what I have to say and make their own judgements.

c) Are you seriously saying that I should believe you just because you say you have three mathematics degrees? Sorry, but my opinion would not change even if you said you were a Nobel prize winner. As I am not asking you to believe me because of my background, I do not intend to display my resume, as I haven't got time for all the email that would produce. All I am asking you to do is stop making unfounded assumptions and pick up a basic book on commercial polling.

d) Why do you keep saying that nobody can refute your analysis? It has been refuted many times by many people. To be irrefutable, your analysis should be able to justify its basic assumptions. You have not. How can you possibly justify the assumption that the Mitofsky exit poll used pure random sampling, contrary to normal commercial practice? Do you have some inside knowledge? Anyway, you haven't justified it, and therefore your analysis is easily refutable.
Perhaps you would like to comment on the points I made. Where do you think the flaw is? Why would it not be possible to predict a winner with selective sampling? Many polling companies would love to hear from you, because that is exactly what they do all the time. It is what we do in everyday life.

e) I am not raising any 'straw men' "about exit poll sampling error and statistical independence and voter psychology". It is you who are have brought up these issues. However now you come to mention it, how can you possibly expect us to respect your analysis if you believe that statistical independence and sampling error are irrelevant?

f) I'm surprised that somebody with three mathematics degrees is bothered by the fact that outside the USA, we study 'maths', not 'math'. However, as it does seem to bother you, I will say 'mathematics'.

g) Yes I am familiar with Florida, Georgia and Ohio. But what you insist on bringing up again and again has got little to do with the situations there, and is only damaging the drive for free and fair elections. Unfounded and badly researched theorising on one issue can damage the credibility of good steady field work on the obvious inequitable distribution of voting opportunity and possible invasion of the computer codes.

h) Yes, your analysis is yours if you want it. Nobody else wants it. I'm sure you can play with it as much as you like, as long as you realise that it is not worth very much unless you can justify your assumptions.

i) Look, I'm not criticising your motivations. But right now, your efforts are wasted. Why don't you just wait until we know the design of Mitofsky's exit polls? Even then, however, your analysis can only indicate fraud: it cannot prove it. There are clearer ways to prove it than that.