BASED ON EXIT POLLS, THE PROBABILITY KERRY WOULD LOSE OH AND FL = 0.15%

[TruthIsAll]

That's less than 1/6 of one percent!

To put it another way, the chances are 1 out of 667 that Bush would win BOTH states.

Assuming a 2% MoE for the exit polls (they are much more accurate than standard polls), the probabilities are:

Ohio Exit Poll:
Kerry 52 - Bush 48
Prob (Kerry wins OH)= 97.7%

Florida Exit Poll:
Kerry 51 - Bush 48
Prob (Kerry wins FL)= 93.5%

Then
Prob (Kerry loses FL and OH) =0.15% = (1-.977)*(1-.935)

Prob(Kerry wins OH or FL or Both)= 99.85% = 1 -.15%

How did I calculate the probs?
Simple. Feed the data into the Excel Normal Distribution function:

Since the MoE = .02, the Standard Deviation is .02/1.96 = .01

Prob (Kerry wins OH)= NORMDIST(0.52,0.5,0.01,TRUE) = 97.7%

Prob (Kerry wins FL)= NORMDIST(51/99,0.5,0.01,TRUE) = 93.5%

[BUSHOUT]

1. Source? n/t

[JohnnyCougar]

2. He did the statistical calculations himself.

[BUSHOUT]

6. Do you seriously know what you're talking about? Curious....n/t

[TruthIsAll]

13. Hey Bushout, why the nasty tone? If you have a question, ask me.

Do you have a problem with the Normal distribution?
Do you have a problem with the Standard Deviation?
Do you have a problem with the Joint Probability calculation?
Do you have a problem with the Exit Polls?

Do you use Excel?
Then try it yourself.

[TruthIsAll]

4. Just feed the exit poll and std dev into the Excel Normal Distribution

Since MoE = .02, the Std Dev = .01

Prob (OH)= NORMDIST(0.52,0.5,0.01,TRUE) =97.7%

Prob (FL)= NORMDIST(51/99,0.5,0.01,TRUE) = 93.5%

[BUSHOUT]

15. So, no source then?

I have a problem with unexplained and unsourced numbers.

Your "probability of Kerry wins" numbers annoy me because I don't see how they're arrived at. And you're not showing it either.

Sorry about my tone, but my patience is thin these days.

[TruthIsAll]

18. What is unexplained? I show the calculation. Are you math-challenged?

As for the source of the numbers, they have been posted on DU from the get-go. Have you been sleeping the last three days?
Where have you been?

These poll numbers were released on election day in the afternoon.

Are you sure you want to see BUSHOUT? Or IN?

[BUSHOUT]

30. You're just very bad at explaining what you're trying to do n/t

[BUSHOUT]

31. You don't explain how you come up with 97.7% and 93.5%

[TruthIsAll]

35. I told you it was the Normal Distribution function.

You obviously never had a course in statistics. If you did, wyou would understand this. If you didn't, I will try to explain:

The Normal Distribution is a mathematical function. It is the cumulative probability function (the bell curve) which is based on statistical sampling theory.

We need to determine the probability that Kerry will get over 50% of the vote in a given state. We do this by utilizing the exit poll split (52-48) and the standard devation based on the sampling error of the mean. The standard deviation used is 1% (about 1/2 the Margin of Error).

The result of the calculation for OH tells us that there is a 97% probability that Kerry would get over 50% of the vote in Ohio, based on the 52-48 exit poll. That is, in 97 elections out of 100, he would win the state.

The result of the calculation tells us that there is a 93% probability that Kerry would get over 50% of the vote in FL, based on the 51-48 exit poll.

[BUSHOUT]

32. AND you're wrong about the margin of error also! This is a WASTE!!

[TruthIsAll]

36. I am not WRONG about anything. If you don't like the results, ignore them

I don't need to waste anymore time on you.

BUSHOUT, you're out.

[BUSHOUT]

38. State exit polls from VNS are 4% MOE. You know what "MOE" means? :)

[DELUSIONAL]

3. That is one hell of a rigged craps game!

Is this a way to explain to to people without any knowledge of statistics:

If a double sided coin was tossed 100 time -- bush would have only won a fraction of one coin.

If there were 100 sheep -- only one would be 15% black.

Any other visual picture to illustrate 15% of 1?

[prof_science]

5. If you can believe the exit pollsters, then...

...in only 1 out of more than 600 elections would * have won both FL and OH.

[ochazuke]

7. Where can we find all the pre-hacked exit polls?

[high density]

11. I'm curious as well

CNN's exit poll as of right now shows Bush 51.38/Kerry 47.62 in Florida, if my math is correct. I seem to recall that the exit polls published by Salon at around 2pm on election day showed about the exact opposite results, so I guess this means that there was about a 6 to 8 point swing in the exit polls by the end of the day. (Is this normal or is CNN cooking the books to make the exit poll match the "official" totals?)

[TruthIsAll]

14. No it is NOT normal; Yes, they WERE cooking the books.

TIA

[LiberteToujours]

8. What makes you think the polls were accurate?

[LisaL]

9. What makes you think electronic voting machines were accurate?

:eyes:

[TruthIsAll]

10. They always have been used to check the voting.

Stands to reason, if you ask someone who they voted for, then you know that
1) They are a very likely voter (they just voted)
2) They are not undecided (they just voted).
3) They are proud of their vote (they just voted).

That's why.

Somehow, I think you knew this already.

[LiberteToujours]

12. Sorry, I thought you were talking about telephone polls

Which are everyone's favourite bashing subject on DU, which is why I thought it strange you were using them in your analysis.

But you're right, exit polls have historically been much more reliable.

[ochazuke]

19. Nothing more reliable than the exit polls

For decades and decades they reliably predicted the final vote count...

until 2000 in Florida. There were thousands of people coming out of the polls that day who thought they voted for Gore, but their vote was counted for Buchanan because of the butterfly ballot, or their punch-card machine was so full of chads that they couldn't poke it through. Or they wrote in Gore in addition to punching his chad.

[ochazuke]

16. have you heard the looney theories about the exit polls?

I've heard the following explanations to reconcile the afternoon exit polls showing a big Kerry lead v. the final vote count showing a * win.

1. Bush voters didn't want to talk to the poll takers. Kerry voters did.

2. Kerry voters were anxious to register their vote because they hate Bush so much. Bush voters took their time and showed up in large numbers later in the day.

3. Men tend to vote later in the day, women in the morning. And of course men are for Bush, women are for Kerry.

These explanations are so flimsy that it almost in itself proves that the election was a fraud.

[AnIndependentTexan]

17. Explanation of rigging an election

http://www.truthout.org/docs_03/voting.shtml#part1

The story you are about to read is in this writer's view the biggest political scandal in American history, if not global history. And it is being broken today here in New Zealand.


http://www.commongroundcommonsense.org/index.php?showtopic=53

As John Hopkins University and Bev Harris have pointed out repeatedly in their first field studies; I have seen the duplicitous nature of just how easy it is to rig elections using the G.E.M.S system.

GEMS central tabulator is nothing more than an ATM machine on your home computer, with all the escape routes and backdoors. There are two columns for votes, one for the plus side and one for the negative. You can freely change the opposite column at any time you like, using MS Access. Doing this means you can add or subtract however many votes you want, within the GEMS system's rules.

What this means is that the opposing rule of positive, must turn out negative. If you take away 5000 votes from one side, reversing his percentage that exact same number will be applied to the other guy's total. It can not work any other way.

In order for the fraud to be workable, it must always come within the same vote total score GEMS tabulates. So yes, it does count every vote there is always. They never lied here. But, in order to keep the total accurate, they can freely remove or add the same number of votes, making it seem like nothing happened, to the other side. Thus the other candidate always wins. Ta daa.

[ShaneGR]

20. I, unfortunately, reject the exit polling as a reason

Why? Exit polling in 2000 was also wrong in both Ohio (Where Bush was projected to win by 10%) and Florida (Where Gore was predicted to win by 3-5%). The problem with exit polling, specifically in swing states where votes can go either way, is it depends on a number of completely unreliable data. Specifically, only people WILLING to be interviewed are exitpolled, exit polling is dependent on time of day, and exit polling tends to follow the pattern of the previous election (Which we all know doesn't currently hold true).


Look, the bottom line is exit polling is completely unscientific and doesn't prove anything, and never will. We need concrete evidence. Hopefully someone can provide it in the next few weeks.

[President Kerry]

21. EXCEPT.. why is there a difference between

paper-ballot-state discrepancies (which is almost none) and computer-voting-state discrepancies (both OH and FL)..

http://www.democraticunderground.com

it's right on the homepage.

[ShaneGR]

22. Not saying it isn't odd....

What I am saying is that it provides zero proof.

[TruthIsAll]

23. My old buddy SGR: Did you ever hear the term CIRCUMSTANTIAL EVIDENCE?

How've you been?

Its great to hear from old friends like you and ProgressiveJazz.

Just like the good old days during the 2002 elections.
You remember, the one where Cleland lost in GA? Hmmmm...

[ShaneGR]

24. Hey Truth, I admit it's good work on your part....

And I also admit things do look a little fishy. But, the problem I'm really having right now is that besides some exit polling data that's genuinely unscientific, we really don't have much except statistical irregularities. We need real proof, like fingers in the cookie jar proof.

My problem is thus: If we don't get proof, and the numbers sink in and Bush still has won, when do we start owning up to our own strategic misfaults instead of blaming everything on fraud or "idiot biblethumpers."

I don't know, I want to see this thing play out so we can catch any real fraud, but I really do see some serious serious problems in our overall strategy and vision.

[TruthIsAll]

29. A "little" fishy? I agree, we need something more.

But the probabilities provide an overwhelming prima facie case for fraud. Let's start from there.

If we don't win this, Bush is legitimized and we are all screwed - permanently. BushCo will NEVER give up the power they have TWICE obtained fraudulently.

[Paradise]

25. Haven't exit polls worked successfully for many years prior to baby *? nt

[TruthIsAll]

26. The probability I kicked this is 100%

tia

[Paradise]

27. Well, if you're basing your probability on an exit poll,

you are "wrong"! ;)

[TruthIsAll]

28. I BASE IT ON TWO EXIT POLLS. OK, PROVE IT! PUT UP OR SHUT UP!

REFUTE THE MATHEMATICS. GO AHEAD. YOU GET ONE SHOT.

I GIVE YOU 5 MINUTES.

[BUSHOUT]

33. How can it be refuted, you never showed your work.

I could pull a number out of a hat and claim that's now many pairs of socks I have.

Then I could ask you to refute my math.

My age = >35
# of feet = 2
therefore pairs of socks = 12

REFUTE IT!!!

[TruthIsAll]

34. What did I not show? Stop the crap. Be specific.

I show EVERYTHING:

The exit polls.

The probability function.

Everything.

You are sadly out of touch.

Everything is spelled out for you.

Why is it that ONLY you are having a problem understanding this?

[the_outsider]

37. Is there any way of getting the exit poll internals?

Which counties were sampled in which proportion, sample size etc.?

[TruthIsAll]

39. Don't know. Don't care. The only thing to know is: it's accurate. Always.

tia