Free -- TruthIsAll -- Elections Software– Do Your Own Analysis, Fire Bush!

[autorank]

Free TruthIsAll Elections Software – Do Your Own Analysis

The question of Who Really Won in 2004 is coming back strong.

The analysis does not need to be restricted to experts only. This software lets you
use the basic data from the Exit Polls, the Pre-Election Polls and the Demographic Data from the Exit polls.

Change assumptions, etc. and see if it’s possible to create a set that allow Bush to win. It’s nearly impossible.

Just go here for the down load.

http://us.share.geocities.com/electionmodel/InteractiveElectionSimulation.xls


Bush is about to hit a consensus 35% approval rating and
he's sure to drop more. Use this software to show people, there was no real way he won in the first place.

[nashville_brook]

1. wow -- so glad to see TIA alive and kicking

kick and rec'd

[autorank]

2. So glad to see you!!!

Bush is even less legitimate than he was a month ago...falling like a rock.

[yodermon]

8. I'm out of the loop

has TIA been tombstoned? someone PM me.
:(

[knowbody0]

3. kick recommended and appreciated

thank TIA for us, Autorank
and thank YOU

[autorank]

4. I will and you're welcome!

:hi:

[DemReadingDU]

5. thank you for sharing!

and thanks to TIA also

[autorank]

6. Thanks, nice seeing you. I'd like to do more reading! Xmas wish list.

[autorank]

7. SHARE this link with your "math" / Excel friends.

Get them hooked. It's time that the Excel community see what TIA has done with their product...made it dance, talk, reason, and argue...all in the name of democracy!

[ArkDem]

9. TruthIsAll won't convince anyone with a 'real' math background.

[In Truth We Trust]

11. Is this sarcasm? If not wtf?

[ArkDem]

14. His data are worthless.

[In Truth We Trust]

16. would you please substantiate your claim.

[ArkDem]

17. Look at his page, it speaks for itself.

[In Truth We Trust]

20. That is not substantiation at all. When you defame someone as you have

TruthIsAll by claiming he is not a "real" mathemitician you should substantiate it. I'm looking for concrete evidence to your claim if you don't mind please.

[OnTheOtherHand]

30. well, let's not get into the semantics of defamation

but I agree, it would be nice for ArkDem to back up the criticisms. (And there is no point in arguing about who has a real math background.)

If you check out the _other_ free software thread, you will find a couple of us trying to explain why we're not convinced by TIA's arguments and spreadsheet. There's also some of it on the "Waste of Time?" thread.

There are quite a few people around ERD who find TIA's arguments unconvincing, although we disagree in friendly fashion about a lot of other things. If there is anything in particular that you find convincing, I will be happy to tell you why I (probably) don't. Basically, TIA expresses high degrees of statistical certainty about things that no one can be that certain of. If people want to be certain of the same things, that's fine, but IMHO they shouldn't depend on his having proved any of them. I apologize in advance if that sounds catty; I'm just trying to be concise and accurate.

[Land Shark]

35. the standard of scientific truth is orders of magnitude higher than

the standards of proof in court necessary to award presidencies, execute murderers, or transfer millions of dollars in judgments. No court process would be likely to be successful if an honest adherence to standards of scientific proof were the order of the day.

Even if people like OTOH were correct, i fundamentally question whether they can take their "scientific" standards and apply them without modification, caveat or explanation to the areas of law and politics.

Just because this or that "coulda been", or there is some uncertainty as to an underlying assumption does not mean the court can not render judgment that becomes legal truth sufficient to (in our country) kill people, for example.

TIA should not be held to higher standards than courts, even if we assume that his critics math/science was 100% correct. And that's debated as well.

[OnTheOtherHand]

36. (sputter)

I have never held TIA or anyone else on the board to a standard of legal or scientific proof. This is a straw man, and it startles me.

TIA's hallmark has been to state as simple fact things that are at best controversial, sometimes Just Plain Wrong by ordinary standards, and to repeat them so loudly and aggressively that many people come to believe them.

I am perfectly willing to entertain arguments for fraud that fall short of proof. But some of them we would be better off without.

[Febble]

38. more sputter

I think the problem here is that different standards apply to different questions, even within the context of a prosecution.

To make a case for prosecution (I would imagine, but I'm no lawyer) you can pile stuff up - exit polls, long lines in Ohio, EIRS incidents favouring Bush, Diebold, Dieb-throat, Clint Curtis, tabulators, backdoors in software - and it doesn't matter if some of it doesn't prove to be solid, nor does it matter if something is contradictory. If (to reuse a metaphor) you find a body, and beside the body is a chainsaw, knife, a gun, and a bottle of untraceable poison, they all make the case for a prosecution for murder, even if only one of the weapons turns out to be actually used.

However, having made the case that prosecution is warranted, if you want to convict, you need a plausible story with a coherent narrative. And if your coherent narrative is based on statistics, statistical proofs give you your facts. Statistical significance allows us to say, more or less - this is so. Just as a detective can say "this knife was found at the scene". Both may have a 19 out of 20 chance of being true (if 1 in 20 cops are bent).

In statistics a 19 in 20 probability of something occurring by non-chance factors (i.e. a 1 in 20 of occurring by chance) is often acceptable (though there are strict rules about when more stringent "alpha" levels are required). But a one in two probability would certainly not be. It would be like coming to court and saying not that there is a 51% chance that murder was done (which I presume would be enough to convict), but that there was a 51% chance that murder weapons were found at the crime scene - which I presume would have the defence rolling in the aisles with mirth.

I think it's probably easier to bridge the gap between legal and statistical standards of proof by using Bayesian statistics, which work kinda like this. You start with a prior assumption that the probability of throwing heads is 50:50. Then someone tells you that that 5% of the coins in circulation have two tails. This is called a posterior. And it means that your chance of throwing heads is now 50% times .95%, i.e. only 47.5%. Nothing has changed except your knowledge of the facts. But it's knowledge that counts.

So: if we start with the estimate that there is a one in a trillion chance that the exit polls were wrong by chance (that's my current best estimate), this tells us that non-chance factors caused the discrepancy. For certain. We then have two possible non-chance factors: fraud and biased polls (and believe, me, they happen too).

We can then take a stab at which is the most likely: I'll play fair and say to start off with that fraud and erroneous polls are equally likely - there were means, motive and opportunity for fraud, but on the other hand, there is plenty of precedent for polls to be wrong. So we now have 50% chance of a massively stolen election - so far. That's our Bayesian prior.

So we then make a prediction - where is fraud most likely? - Ohio - let's say that the probability of fraud in Ohio was also 50%, given what we know of Rove and the importance of Ohio. OK. and the exit polls were a long way out in Ohio. That's our posterior. So our Bayesian probability goes up - maybe it's 25% chance that the election wasn't stolen .

OK - so we make another prediction: if fraud was responsible for the exit poll discrepancy in Ohio, we'd expect Bush's swing to be greater where the redshift is greater. So we correlate the two and get a correlation coefficient with a p value. If it's positive at a probability of p<.05 we can say that this can't have happened by chance, and unless we postulate some other cause for the correlation (and you can never absolutely prove causality with a correlation) our Bayesian probability of fraud in Ohio becomes close to 99%. It also raises the probability of fraud in our prior: we can now justify saying that the likelihood that the cause of the exit poll discrepancy nationally is fraud is substantially higher than the chance of it being due to polling bias.

However, if the relationship between the swing and redshift in Ohio is "insignificantly" different from zero (which the ESI finding suggests), we don't go back to 75% - we don't even go back to 50%. We now have evidence that our prior (50% poll, 50% fraud) was too generous to fraud - so the chances of the polling error being the cause of the discrepancy goes up.

However, that's only the logic for a massively stolen election. We still have our prior for Ohio - we think it was 50% that Ohio would be stolen. But it doesn't look as though the fraud was related to the exit poll discrepancy - so what kinds could there have been that wouldn't affect the exit polls: provisional vote scams; machine rationing; differential spoilage; maybe tampering with the votes here and there where the exit polls weren't looking. Well, there is good evidence for all these. How likely is this stuff to be deliberate? Let's say 50% again. So we are now up to 75% chance of deliberate fraud in Ohio.

Using this kind of logic, I'd happy with 51%. Actually I'd be happy with less. But using this logic we are nowhere near 50%, given the ESI finding, although further statistical "facts" might move the Bayesian probability yet again, in either direction.

The moral of the story is that you can't just add bits of statistical evidence together. You actually have to multiply them, and their values chance with more knowledge, and you don't always get the answer you expect. But some of the answers might be quite interesting all the same. And a lot more interesting than a claim that's going to be laughed out of court.

[Time for change]

39. I'm confused as to what you're saying here Landshark

Although I have disagreed with OTOH on a number of issues (most especially on the fact that I believe that Kerry won the elction, and he doesn't), I feel that it's not accurate to say that he has held TIA to a standard of "scientific proof". On the contrary, their basic disagreement is of a different sort. I also believe that it is important to understand the nature of this disagreement, because if we are going to argue the case for a Kerry victory to the public (which I believe is very important), it is important for us to have a good understanding of these issues. So, I will try to clarify our differences.

First, with regard to the relationship between the exit polls and who won the election:

TIA believes that the exit polls proove that Kerry won the election -- both the popular vote and the electoral vote. He believes that because statistics proove that the exit poll discrepancy from the "official vote" could not have happened by chance, and he is not willing to consider the possibility that the exit polls could have been biased.

OTOH believes that Kerry may have won the popular vote, but probably not. He believes that there is a greater likelihood that Kerry won the electoral vote (i.e., that he won Ohio) than the popular vote, and I'm not sure exactly where he stands on that. The reason that he believes that Kerry probably didn't win the popular vote is that he believes that the exit polls were biased.

I believe that Kerry won the popular vote, and I believe that the exit polls are one of numerous issues that significantly strengthen the argument that Kerry won the popular vote, but there are also a number of other issues that contribute to that argument. I also believe that it is even more likely that Kerry won Ohio and therefore the electoral vote. But I definitely do not believe that the exit polls proove that Kerry won the popular vote. To have a better understanding of where I stand on these issues and why, see this thread of mine, where I discuss 10 reasons to believe that Kerry won: http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=132&topic_id=2052179. Note that in my discussion of the exit polls I do not claim that they proove that Kerry won the popular vote, only that they add to that argument. But that is only one of 10 arguments that I make. Also see this in-depth discussion of the exit polls that I posted in May: http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=371726.

Anyhow, I suppose this was kind of a long prelude to the main point that I wanted to make, but here it is: The main argument that OTOH has with TIA is that OTOH believes that any argument about the relationship between the exit polls and who won the popular vote must at least consider the possibility of exit poll bias. He is quite correct in that belief, and I can tell you as an epidemiologist that I would be considered incompetent if I ever conducted a scientific study and dismissed the potential role of bias without even considering it. And if a court would simply dismiss the potential role of bias without even considering it, then I would have to say that the court was plain wrong. That is why whenever I write about this issue I make certain to discuss the potential role of bias before explaining why I believe that fraud is a more likely explanation for the exit poll discrepancy from the official vote (i.e., red shift) than exit poll bias. In fact, I believe that it is highly likely that there was some bias in the exit polls. I just believe that there was more fraud than there was exit poll bias. But even that I can't state with certainty.

And in summary, I think that OTOH is quite right to talk about the potential role of exit poll bias in this Forum because it helps us to gain a better perspective on reality by considering his arguments. As you can see from the above links, I have talked about the potential role of exit poll bias on a number of occasions myself, but I don't get flamed for doing that like OTOH does, because I say at the same time that I believe that Kerry won the election. So I must say that OTOH does us a service by bringing this issue into play, even though I disagree with his final conclusion. And, it is also important for me to say that in my pursuit of other mechanisms of fraud, OTOH has been of much help to me off of DU.

[anaxarchos]

43. I'm just getting used to violently agreeing...

...with you on a number of points (role of conspiracy theory, etc), TfC, but what you say here is simply not true. The main argument TIA (and I) have with OTOH is that he has become the most dogged proponent of the "Bush Won" camp. He is the first one to answer my posts and he played the same role with TIA.

Now, this is not a crime. It is legitimate. But to put him in the camp of simple or scientific "doubt"? ... gimme a break.

I admit that there is a certain duality to his posts - that's why I concocted the Dr. Other/Mr. Hand thing. But that all gets old really fast.

The issue is not about OTOH's doubts or his person, it is about his position (politically speaking).

He comes from a long line of posters of whom Mistwell comes to mind.

And, he's not the best by a long shot. At least Mistwell tried to argue the case. I have never been able to get past "well it seems to me..." with your guy.

[Febble]

44. This is a decidedly strange post

Is this a "game" where we start by deciding "who won" then proceed to decide who has the best argument?

Or is it a forum where we try to find out what happened?

Or campaign where we press for election reform?

The last two make sense to me. The first makes no sense at all, yet it seems to me to be the way you are trying to frame the debate.

Speaking for myself, I think that probably "Bush won" in the sense that I think probably more legally cast votes were cast for Bush than for Kerry. I even think that the balance of probabilities is that this also happened in Ohio. I can cite reason for thinking this, but I won't in this post, as I you know I have done it elsewhere.

And I suppose that puts me in the "Bush {probably} won {the popular vote}" camp as well as the "Bush {possibly} won the {electoral vote}" camp. But there is assuredly nothing political in those positions. Nor do they prevent me from:

1. Wishing fervently that Kerry had won
2. Thinking that Bush's victory was achieved by certainly unethical and probably illegal means
3. Trying to find out what those means were and where they were used.
4. Pushing for transparent, secure election systems
5. Pushing for an end to voter suppression.

So that's my own "political position".

Now tell me: how does failing to conclude from a long hard look at the evidence that "Bush won" say anything about ANYONE'S political position? All it says, to me at least, is that OTOH is a rigorous mathematician and scientist. I hope I am too.

Because I think it will take intellectual rigour to find out what happened and to succeed with any prosecutions that result.

[anaxarchos]

45. This is a political board and it is full of debate on political issues...

..look around you. Sides everywhere. OTOH is a rigorous partisan. He is primarily interested in promoting his position. Whatever else he may be is not clear to me.

You too, I think...

Does debate happen in such a place?... Sure.

Is it primarily designed to form or promote one's opinion? Depends on the person and the subject...

Mostly, the only "open-minded" debate about fraud in this forum is about how it may have happened.

There are a few who want to say it didn't happen. The onus is on them to make their case until the tilt of the forum changes. Then, the onus will change as well.

I see no indiciation of "open-mindedness", nor do I demand it.

But let's be straight with each other...

[foo_bar]

46. AnaxLight'n'Dark is decrying "duality"?

Isn't there some DU rule against irony?

Sides everywhere.

Flatland, romance of many dimensions!

He comes from a long line of posters of whom Mistwell comes to mind.

As opposed to the RaulVB-davidgmills-TIA lineage, all diamonds in the rough.

Whatever else he may be is not clear to me.

When your last refuge is "freepers!" (or "Jews!" in the case of davidgmills), it means you lost the meta-game of being taken seriously. You can then reformulate your beliefs, or you can defend them to death in increasingly smaller venues.

The onus is on them to make their case until the tilt of the forum changes.

If you mean the "Election Reform" forum, I'd say the fulcrum has shifted since you first made that argument.

[anaxarchos]

51. f_b, ya gotta watch the coherence thing....

You are becoming more indecipherable every day (at least to me).

If you are thinking that I am calling OTOH a freeper, you are entirely wrong. I don't even think you are a "freeper".

As for the tilt changing... might be... I kinda expected it from day one. Take no comfort in it though. The first guy to get arrested will be you.

And as far as Light and Dark goes, you claim to admire that.

[foo_bar]

52. it seems like you got the gist of it

The first guy to get arrested will be you.

Not sure about this part, though. Is it an allusion to Niemoller's "first they came for the Communists"?

And as far as Light and Dark goes, you claim to admire that.

"as Light and Dark goes,"
groaned anaxarchos.

[anaxarchos]

53. Nah... On both counts.... n/t

[foo_bar]

54. can you elaborate on that?

As Eliza would say:
http://www-ai.ijs.si/eliza-cgi-bin/eliza_script

[anaxarchos]

59. Ha... You've got some noive...

I'm the one who lets it all hang out there. You're the one who lurks in the shadows and actually works at producing the most obscure references possible. Remember?

Sure, I can elaborate, and it's very simple. It's the same reason I asked you why we were on "different sides". I think you won't stop. I think your "intellectual indulgences" will get you shot as soon as peace arrives in the land... figuratively speaking of course. It's kinda one of those "takes one to know one" references. WTF, I could be wrong...

I'll be happy to tell you about my "two voices" too if you really wanna know. That is equally mundane. It's just "training" and it dates me. When I was growing up, there was a deep seated and widely shared criticism of the smug, sterile talk of the larger society. It was generational but it went double for the "learned trades". We got used to dropping into "street voice" as a form of both self and group deprication (and to add "perspective"). At first, the "street voice" was a caricature but life gave it reality. Now, I wouldn't change it if I could. If you talked to me in person, the two voices actually sound different. Lucky for me they say exactly the same thing...

See what I mean? No mystery... That's your turf.

[foo_bar]

65. fugetaboutit!

A guy calling himself anaxarchos frets about "obscure references"?

I can't picture this figurative arrest and execution, but I don't really need to. The important thing is you blew off steam.

Dunno about the voices, I was thinking of the vast gulf between "deadly serious" and "Shall We Dance". My street voice drops all the R's, which loses something on duh intennet.

[anaxarchos]

66. Ya got me all wrong, gwasshoppa....

I almost never need to "blow off steam"...

Same with deadly serious... Was, wasn't it?

As far as "R"s go, that still puts you 25 letters ahead of most Republicans and those who work for MSNBC...

And my pal, Anaxarchos, obscure? You must be joking...

[foo_bar]

67. it's twue!

"anax" is rather obscure, and internet hubris doesn't translate into street smarts.

If you aren't here to blow off steam, why the death squad rhetoric? Isn't it a little melodramatic, even for a deadly serious internet chat?

[anaxarchos]

68. There you go skipping the meds again...

"Death squad rhetoric"...? and I'm "melodramatic"?

What ever you are hearing, I'm not sayin' it.

Man, ya gotta focus...

Tata...

[foo_bar]

69. now you're bringing back memories

JIM LEHRER: Was "There you go again" a line that just came to you spontaneously, or was it something that you had worked on?

PRESIDENT RONALD REAGAN: No, it just seemed to be the thing to say in what he was saying up there, because it was to me it felt kind of repetitious, something we had heard before.

http://www.pbs.org/newshour/debatingourdestiny/interviews/reagan.html

The first guy to get arrested will be you.
I think your "intellectual indulgences" will get you shot as soon as peace arrives in the land...
What ever you are hearing, I'm not sayin' it.

Yet OTOH is dualistic.

[yowzayowzayowza]

72. PMFJI, but...

Suggestion.

[Febble]

48. Yes, let's be straight

You're smart enough.

I've no objection to politics, nor to partisan politics. What I do object to is when people infer political partisanship from the scientific conclusions arrived at by other posters.

Speaking for myself, I'm certainly not promoting a position, whatever you may think. OK I'll rephrase that: I AM promoting the position that the evidence of the exit polls is far from clearcut evidence for fraud and may even be evidence against fraud.

However, I am only "promoting" that "position" because it is my honest reading of the facts as I currently see them. And, as I state in my sig, I do not claim any special authority for my views, apart from the statistical logic of my arguments. I'd love to be wrong. I simply point out what I see as the relevant statistical evidence, and what I think it can be said to tell us and what it cannot be said to tell us. It happens that I simply disagree with most of TIA's inferences. Not, I have to say, on the basis of his arithmetic, but on the basis of the assumptions he makes. All statistical inferences depend heavily on assumptions, and I think his are mostly faulty.

And the only "political" "position" I am promoting is the position I have stated elsewhere, which is that I think your democracy has been violated by the lack of transparency and accountability in your electoral system and by the unconscionable disenfranchisement of many of your citizens - the majority of whom would probably (statistically speaking) have voted for Kerry.

So why am I "promoting" the first position if I also want to "promote" the second? (You didn't ask me that, but you should).

Two reasons:

One is that I don't think it helps a case to present bad arguments, and I think that some of the statistical arguments, especially the exit poll arguments are bad arguments. I don't believe in presenting your opponents with ready made straw men.

Second is that I do think the exit polls tell us some important things about what unethical (and even fraudulent) practices may have occurred - by telling us what probably did not occur. If we take your own point (which was my starting point also) that Bush's victory simply does not make sense (because it makes no sense to vote for Bush) then if the exit polls are telling us, as I believe they are, that vote-switching probably did not occur, then it directs us to look for things like vote-suppression (a term I use advisedly, as it includes things like provisional vote allocation/rejection; spoilage, and voter suppression). In other words I actually think it is VALUABLE to look at the evidence dispassionately. And damaging to look at it passionately. I'm happy to have political incentives direct my science, but I'm not happy to have it affect the conclusions I draw from the evidence. And I'm even less happy to have other people make erroneous inferences about my political motivations from those conclusions.

AND even less happy still to see conclusions drawn about the political motivations of a friend who has, to my certain knowledge, spent a vast number of hours trying to find out why the hell Bush is president.

[anaxarchos]

50. Lemme get this straight...

You argue against fraud... But you do this in order to be dispassionate in the pursuit of science because you really WANT to prove fraud... But you pursue this on a political board... incessantly... But you don't want anybody to accuse you of arguing against fraud because that is deeply offensive...

What can I say?

I'm not that smart.

[Febble]

55. Listen

I argue that the exit poll evidence not only is poor evidence of massive fraud but even argues against it.

I think fraud or unethical, possibly illegal voter suppression may have cost Kerry Ohio, which means the presidency.

I think we are more likely to find evidence of the second if we look at what the exit polls are telling us about the first.

And I don't give a damn if anyone accuses me of "arguing against fraud" as long as they listen to the argument. I'd love to be persuaded the other way. I DO give a damn if people assume that I am "arguing against fraud" because I have political reasons for legitimizing Bush. Because nothing could be further from the truth.

Yes, I want to do dispassionate science, because I think it's by far the best way of uncovering the truth. And I think the truth stinks. I just don't think the truth stinks in the way that those believe that the exit poll discrepancy was caused by massive fraud think it stinks.

And I gave you two reasons arguing this "incessantly": 1. the exit polls are a straw man. 2. The exit polls CAN put constraints on where fraud might have occurred, and in what way it might have been done as long as we look at them properly.

And I do all this on a political board because it is about the functioning of the democracy of the most powerful nation on earth and the legitimacy of its leader.

[anaxarchos]

61. And that's what I said... What's your beef? n/t

[Febble]

62. Well, I certainly haven't got one

if you haven't. That was rather my point.

But if you don't have a beef, what's with the "'Bush Won' Camp" stuff?

As far as I can see there isn't one. A Bush Won camp, I mean. There's just a few of us less then convinced by the exit poll argument that he didn't. But AFAICT we are all united (pretty well) that it was a stinking election, and that we want to make sure that future elections are fair, clean, and seen to be fair and clean.

And some of us fear that presenting the exit poll story as though it indicated a 16,000,000 to 1 probability of fraud is a way of getting election reform laughed out of court rather than taken seriously.

I'll put myself happily in the "let make sure our claims are not easily falsifiable" camp. As well as in the "what legitimacy has a democracy if you can't be sure who won camp". And of course the "a democracy that systematically and disproportionately disenfranchises its ethnic minorities is no democracy" camp.

[Time for change]

47. Ok then, let's talk about OTOH's positions

Obviously I can't presume to talk about all his posts, because I haven't seen them all, and you may be referring to some stuff I haven't seen. So, I'll just stick to what I know.

You're probably wondering why I, who have posted numerous threads on DU that provide evidence for and support the view that Kerry won the 2004 election, would defend someone who is one of the chief arguers against that view. That's a legitimate question, and here are my legitimate reasons:

1. NOBODY has helped me more than OTOH in my attempts to accumulate non-exit poll related evidence that Kerry won the election. We have exchanged hundreds of e-mails off of DU on this and similar subjects, and it is safe to say that some of my best posts on this subject would not have occurred without his help.

2. I don't judge a person's motivations solely on the basis of their conclusions on technical/scientific issues. Yes, I have disagreed and do disagree with OTOH about the likelihood that Kerry won the election, and we have argued about this a great deal. And certainly it is true that far more Kerry voters believe that Kerry won the election than do Bush voters. But I certainly don't see OTOH's beliefs in this regard as an indication that he is less of a Democrat than the rest of us. What is important in this regard is not who you believe won the election, but how hard you are willing to work to find the truth.

3. You may disagree, but OTOH presents very good arguments. As such, I think that it's very important to have him participate in our discussions. I have learned a lot from his arguments, and they have helped me in pursuing election issues. So how do I explain our disagreement on the main issue? I think that he is overly scientific/cautious about this. Many scientists are. But also, see reason # 4.

4. I understand OTOH's "devil's advocate" kind of stance towards things because I have that kind of mind set myself. In fact, I used to drive my father crazy with my constant disagreements with him. That kind of mind set is one that has the view "If an argument has a weakness and I discuss that weakness, even though I may not be against the conclusion, then we're all likely to learn something in the process". So, OTOH attacks what he sees as weak arguments whenever he sees them. I don't do that myself these days, because I have other priorities, but that doesn't mean that I don't see the usefulness of what he does.

5. It IS true that one of OTOH's pet peeves is over-reliance on the value of the exit polls, including the refusal on the part of many DUers to even consider the possibility of bias. OTOH is absolutely correct that bias needs to be at least considered in any argument about the exit polls, because not to do so dis-credits us. One of my best posts was a long discussion of why I felt that the exit poll discrepancy with the official vote indicated fraud rather than exit poll bias: http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=371726&mesg_id=371726. Who was the first person to support this thread? And why was he so supportive of a OP that argued for a Kerry victory (and I didn't even know him then)? Because it at least discussed the need to consider the role of bias in an assessment of the meaning of the exit polls.

6. And just to drive home the point that OTOH is willing to support what he sees as a good argument for a Kerry victory, who was one of two people to come to my support against at attack on my latest thread to make a case for a Kerry victory?: http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=104&topic_id=4998616&mesg_id=4998616. (See post # 54.)


So, in summary, I think that we Democrats need to realize that we're on the same team, and not get hostile with those who disagree with us, not over moral or political issues, but over technical/scientific issues. Good arguments can help us to find the truth, even if their conclusions are wrong. We ought to use those arguments as they were meant to be used, to assist us in our search for the truth, rather than waste our energy by getting angry over them.

[anaxarchos]

49. TfC, you miss my point...

You can think whatever you like about whatever you like...

What you can not do is set up a false dichotomy. Your dichotomy is "certainty" versus "skepticism". You also imply a modifier such as "stubborn certainty" or "fanatical certainty" versus "healthy skepticism" or "scientific skepticism".

Ain't true...

You have two sides of a political debate. They disagree. That's all.

The two sides are "fraud" / "no fraud"... Everything else is smoke.

Wanna be scientific?
Then step outside yourself and your opinions - do an experiment.

Dig up the last 10 threads that TIA did...
See who shows up almost immediately.
What do they say?
What do you conclude from this?

or

What's the subject and intent of this thread?
And what exactly are we talking about?

Listen, I don't care either way... I have already said that I think most political positions are legitimate.

As far as stifling dissent, gimme a break again....

I hadn't realized that the Democratic Party was purging people who DIDN'T believe there was fraud in 2004. These are truly rare opinions you want to defend. Almost "endangered"...

[Time for change]

56. I think that you miss my point anax

I don't look at our efforts here as a two sided political debate. If it is, it really doesn't matter much who "wins" that debate, because if everybody on ERD is convinced that Kerry won the election how is that going to help us?

What IS important IMO is three things, all of which I detailed in another post, so I'll cut and paste it here:

1. We can fight for election reform, using as our primary argument the proven fact that our current election system is easily manipulated and provides no basis for confidence that our elections are fair. Therefore, we can even make the case that our Democracy is in grave danger, if not gone, until we fix this problem.

2. We can expose all the fraud that we are aware of and try to educate the public about all the good reasons to believe that the 2004 election was probably stolen.

3. We can continue to look for actual proof that the election was stolen. I believe that it exists out there. Some have pointed out that the most likely scenario for finding the proof will be to continue to identify statistical and other evidence for a stolen election, which in turn will encourage whistle blowers to come forward with evidence that will eventually add up to actual proof.


All three of these things are very important. The part that I am most interested in, and that OTOH is most interested in, because that's what we are best at doing, is # 3. Doing this does not require that we feel certain that Kerry won the election. And, arguing against weak arguments that Kerry won the election may help us reach that goal (obviously OTOH feels that this is a more important activity than I do, but I have no argument with him on that score). And I can assure you that OTOH also supports #s 1 and 2. And he also believes that there was a good deal of fraud in 2004. He just hasn't been able to find what he considers solid evidence to quantify how much fraud there was, though he continues to work on that.

As far as the subject of this thread is concerned, here's the sequence of events that led to my involvement: Landshark accused OTOH of holding TIA to an unreasonable standard of scientific proof. I felt that Landshark didn't understand that OTOH's main argument with TIA is that a thorough assessment of exit poll bias needs to be an important part of all exit poll related arguments for a Kerry victory. I feel that it's important that Landshark understand that, because I believe that if someone tries to argue the exit polls without addressing that issue they will be discredited in court. Obviously Landshark knows a lot more about the law than I do, but I know more about science than he does, so I thought that this might be useful information for him to consider.

I was faced with the same problem myself when in the first week of January a participated in a small group that went to Capital Hill to lobby Senators to stand up against the election results. At the time, the best evidence I had was the exit poll evidence, which I had accumulated from what Jonathon Simon had put out on the Internet. I went around spewing out the million to one/ billion to one sort of statistics, and I had a chart with all the state results, including the 17 states that were outside the margin of error, all in Bush's direction. But I was very careful not to imply that exit poll bias was an impossibility. I didn't emphasize it, but when Senatorial staff brought the subject up, rather than telling them that that was a ridiculous thing to talk about, I honestly admitted that it was a possibility, and then I proceeded to tell them the reasons to believe that the exit poll discrepancy with the official count was more likely the result of fraud than of exit poll bias. I believe that that gave me more credibility than if I had simply dismissed the issue.

And I certainly never said that the Democratic Party (or DU) was purging people who didn't believe there was fraud in 2004.

[anaxarchos]

60. I am not missing your point, TfC,

I am explicitly rejecting it...

And, you are ignoring mine so I am going to politely bow out of this discussion. I have no particular need to escalate this.

For the record:

1) I'm kinder to your guy than you are. I think he goes from thread to thread looking for "bad apples" in order to discredit the barrel (I wasn't kidding about Mistwell). I believe he does this because he genuinely doesn't believe in the barrel ("fraud"). You believe he merely wants to substitute for the slogan "Fraud!", the far more "accurate", "I think there might have been fraud but everything that indicates such is either wrong or could be explained by other factors so you are free to believe fraud existed, despite the fact that I personally don't...". If I was convinced of that, I would really be pissed off.

2) Your guy is not right. He gets chewed up in debate because he starts the fight and then has no case... not because he is the unsuspecting victim of "group think". You are free to believe otherwise.

3) None of this has ANYTHING to do with "science". You should accept your guy's testimony to that effect. I once counted 126 references to "false memory" or its variants on this board without a single basic citation. Science without references? Science without citation? Science without empirical evidence? Science through "expert assertion" alone? I am tempted to ask WTF you social sciences types have done with "science" (no, I don't mean it). This ain't "science" (on this point alone let me give "expert testimony"). If it ain't "politics" either as I claimed, we've all got a problem.

4) I don't need to give your guy a hard time... He gives me more than enough chances on his own. This is about your "ringing defense" which, I believe, "frames" the debate very badly. This is not the first time we have disagreed on this very issue.

I respect your efforts, nevertheless. Good luck with them and it goes without saying that I will help out if you have a need.

[OnTheOtherHand]

71. well, as long as we are speaking for the record

My political stance on the board is that activists will look like idiots if they insist on untenable arguments, so I vehemently oppose arguments I so regard. (No, that doesn't mean that everything I post is vehement opposition -- but then, anyone who reads my posts already knew that.) This doesn't strike me as difficult to understand, so I don't know why you keep getting it wrong.

I presented direct evidence of survey respondents changing their reports about their 2000 vote. You briefly acknowledged it -- which was refreshing, because it might have been the first admission that the 'false memory' argument had any merit. You then went on to argue that because the data were from the NES, they could not be mechanistically applied to the electorate. That might have been a devastating rebuttal except that (1) I had already conceded it; (2) you presented no argument that the actual level of false reporting would be expected to be substantially lower than in the NES; and (3) it's sort of wishful to trash the NES while accepting the exit polls as gospel. All in all, you offered a polemical rebuttal rather than a serious response, and I revised my expectations accordingly. You may still sincerely believe that false vote reporting could not account for the exit poll results, but you have not shown why not.

[Febble]

58. I haven't checked

the last 10 TIA threads. I certainly didn't turn up on all of them, as that would have been a full-time job.

But many of them were specifically titled to challenge "Naysayers"; "Febble"; "OTOH"; "quants" - so not surprising if OTOH and I and other naysaying quants showed up.

I am never sure, anax, from your elliptical style, what you are proposing/refuting/asserting/discrediting.

If the "two sides" to this debate are "fraud" and "no fraud" then I'll come down on fraud. The chances that there was no fraud in the 2004 debate seems to me to be rather small. If the "two sides" are electoral vote winning sized fraud versus no fraud, I'll sit on the fence. If the "two sides" are popular vote winning fraud versus no fraud, I'll go for "no fraud".

But all these places seem daft places to put the fence. The important questions are:

Are elections stealable? - YES
Was the 2004 election fair? = NO
Was there deliberate fraud? = Probably.

And following from those there are various things we can do to fix it. I'm mostly concerned with the last one. And as TfC says, if statistics are going to help with finding out if, where and how fraud occurred, then we need to listen to the statistics, even when they are telling us that some forms and magnitudes of fraud probably did not occur (which is what my reading of the evidence tells me).

So enough with false dichotomies. We don't need any dichotomies. We just need to know what went wrong, what could go wrong, and how to fix both.


(edited for typo)

[In Truth We Trust]

10. Thanks Autorank! TruthIsAll is my hero and a true American patriot!

I would love nothing more than to see his data presented to the masses for their awakening!!!

Paper ballots NOW!
Hand counts NOW!!!
Impeachment NOW!!

[autorank]

24. You're welcome. It's like the Gutenberg Bible...give to everybody

Let them make up their own minds.

No more damn intermediaries.

We are the mathematicians we've been waiting for.

[Peace Patriot]

12. Recommended! And I wish I could recommend it a thousand times more!

Wake up, People! Diebold and ES&S election theft machines into 'Boston Harbor' NOW!

[autorank]

23. Well you could try but would'nt that be...

a real case of "vote fraud":evilgrin:

[Kurovski]

13. Kicked and recommended.

Best wishes to TIA!

[redacted]

15. Shrub is only about 6 rating points above Nixon just before impeachment.

[autorank]

18. Oh, you are so on target.

And it looks like he'll catch him on the way down...and then it will be obvious to everybody in the 70% negative/disapprove group that HE STOLE THE DAMN ELECTION, TWICE!

Here's one way he gets to 30%;

[redacted]

29. Thank you autorank! Appreciate the link and have one for you

in exchange. Took me a long time on Google to find this. Lots of plots relevant to what's happening right now. Check this out:

http://66.102.7.104/search?q=cache:JiDy17__H5MJ:www.nyu.edu/its/socsci/Docs/scandals.html+nixon%27s+approval+ratings&hl=en&start=11

Presidential Scandals and Job Approval
Impact Analysis with SAS

by
Robert A. Yaffee

Statistics and Social Science Group
Academic Computing Facility
e-mail: [e-mail]

[anaxarchos]

32. Thank you for this...

It's an interesting angle, isn't it?

[texpatriot2004]

19. Thanks TIA what a good idea. n/t

[autorank]

25. I'm autorank, the guy who did the elections software is TIA;)

Glad you out there, undeterred by the forces of nature!

[melissinha]

21. Question

I read it in passing, I haven't kept up with the Election forum, but you guys consider the Brazilian Electronic Voting model? From what I understand an independent agency in charge of voting and source code is checked.

On that thinking, you guys know that voting ismandatory in Brazil, interesting huh? I think once you get to like 70 it isn't mandator

[autorank]

22. Good point. The Brazilians reasoned through the process. Paper is the

only solution, however, because it takes any chance of fraud out of the election if you have citizen observers at every precinct. That's where the count starts and where it should end. Just post the agreed upon number by counters and observers and put it on the internet and in the paper. People can do their own calculations to match the tabulated vote of all precincts/counties. Cool.

India has a very good electronic system, very low tech, hard to fool with. Their parties really hate each other, make us look like wimps. They don't have challenges

Since Brazil can do it and India can do it, the fact that we can't means the people in charge here DO NOT want to. Sad isn't it.

Here's what I think would be a perfect solution (modestly put, haha).

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x394982

:hi:

[LibertyorDeath]

26. Big kick for TIA

:patriot:

[mom cat]

27. You are the best...You have clung to the truth and are now being

vindicated :yourock:

[Darkhawk32]

28. Where IS TIA? n/t

[kster]

31. kick.nt

[kiwi_expat]

33. I just tried TIA's software. I was trying to explore a hypothesis.

I was trying to ADD enough Bush voters-who-didn't-vote-in-2000 (but did vote previously) to see how many I would need in order to change the early-exit-poll data to a Bush victory of 3 million.

The software was highly interactive and a definite improvement over using just a calculator. It did nicely re-calculate the various votes (in millions), which was very helpful.

But I still had to use my calculator to re-calculate the percents for variables that I was not modifying (to add up to 100%). It would be lovely if TIA were to have a version that would do that work for me. :-)

[anaxarchos]

37. You are looking at more than a simple "scenario"...

You are recreating nothing less than the Republican 2004 Presidential Strategy. The problem the Republicans were addressing was the 2000 near dead-heat and the likelihood of recreating it in 2004. Given the Gore plurality in 2000, the likelihood of the Nader vote going to the Democratic nominee and the geography, originally they posed an ambitious plan for building a "New Republican Majority". The theory was that by using the patriotic wind of 9/11 at their backs and by focusing the "Power of The Presidency", Republicans could chip away at the conservative edges of traditionally Democratic constituencies. Thus the concentration on Hispanics, the invitation of conservative Black ministers to the White House (even as Bush refused to speak before the NAACP), etc.

As the election approached, it became clear that the impact of Iraq and the faltering economy (particularly in potential battlegrounds like Ohio and Pennsylvania) had undone that "strategy". The situation left few good GOP alternatives.

A battle of attrition for electoral turnout is mostly counterintuitive. It appears to be largely subjective but, in truth, it is typically fought "by the numbers" (i.e. money, phone calls, response rates, etc.). In 2004, the Democrats had nearly as much money to spend and were highly motivated. They had already won the battle for registration and historically the Republicans have only been able to better mobilize a smaller constituency (i.e. extract a higher percentage turnout from a smaller number of total partisans that could theoretically be mobilized). There are a number of genuinely knowledgeable veterans of these types of fights on this board if you need expert opinion.

In the 2004 elections, there is no reason to believe that the Republicans won the trench battle (and many reasons to believe that at best they got a draw). There is also no reason to believe that Rove put his faith in it. It happened but it was largely background noise to the Republican "strategy".

The other possible alternative was the tried and true one of winning the "undecideds", independents, and new voters. This was openly advocated by several Republican "strategists" and was proposed through the mechanism of swinging Bush's position "to the center". The interesting thing was that this was openly opposed by Rove. The basis of his opposition was that the electorate was already very polarized and that the hit rate among the "undecideds" would be too low.

Instead, Rove proposed a gamble which has since been widely regarded as a work of "genius" but which, at the time, was seen by many Republican veterans as all but conceding the election. Rove proposed a "turn to the right" instead, and, more specifically, a turn to the evangelical right. Bush undertook a series of initiatives toward fundamentalists (whom he had largely ignored for three years) which was crowned by support for a constitutional amendment against gay marriage (which, tellingly, hasn't been heard from since). The gamble was that this would mobilize the fundamentals to such a degree that it would overcome any erosion that it might cause in the tiny "middle".

The object of this strategy was 3 to 4 million conservative fundamentalists who were thought to have sat out the 2000 election. This constituency was the ONLY rapidly mobilizable and sizable group within the Republican constituencies which COULD be brought to the 2004 elections, thus justifying the "gamble" according to Rove. By this mechanism (perhaps as many as 3 million votes), the Republicans could just offset the Democratic plurality of 2000, the Nader vote, and a modest Democratic swing of the tiny "middle". Because this strategy was thought to be disproportionately effective in certain battleground states (Florida, Missouri, and a few others), it was thought that it could gain near parity or even a tiny plurality in the popular vote and that it could be focused geographically to win a narrow victory in the electoral vote.

All of the above probably happened... But where the rub comes in is with two factors. First, instead of 110 to 112 million votes as was expected in a "rerun of 2000", the turnout produced nearly 123 million votes, effectively swamping the Rovian strategy with a much bigger "middle". Second, Bush achieved, not narrow parity, but a 3 million vote majority in the popular vote. Some Republicans were "stunned" and this accounts for the claims of a "landslide", etcetera, immediately following the elections. Of course the election did not break any ground historically in presidential races (it was in fact a very close race), but what was really being referred to was a 3 million vote majority where no such vote could exist...

This "what if" is probably the closest to reality that could be proposed and the irony is that it is supported by the 12:22 AM exit polls. In fact, it is not so much that the exit polls suggest "fraud" so much as they support a logical understanding of an election which can not achieve that outcome WITHOUT fraud. The unadjusted exit polls suggest a 39/41 split which neatly follows the expected results and, if you followed the Prisuta reference in the "Game" thread, it also accounts for this kind of response error as it is generally described in the literature (i.e. a 39/39 actual split with nearly 3 million fundies reporting themselves as Bush voters in 2000 despite the fact that they didn't vote PLUS a proportionately greater swing in nonvoters for Kerry). Of course, Kerry wins handily in the 12:22 exit polls...

Please report the conclusions of your exercise....

[autorank]

40. Bush winner contingent: answer this or forever hold your piece!

Which you may be doing anyway.

This is just brilliant. In addition to the number, and I'm talking about TIA, althecat, and eomer, and everybody else, this says it all.

This is the real question that has to be answered.

But where the rub comes in is with two factors. First, instead of 110 to 112 million votes as was expected in a "rerun of 2000", the turnout produced nearly 123 million votes, effectively swamping the Rovian strategy with a much bigger "middle". Second, Bush achieved, not narrow parity, but a 3 million vote majority in the popular vote. Some Republicans were "stunned" and this accounts for the claims of a "landslide", etcetera, immediately following the elections. Of course the election did not break any ground historically in presidential races (it was in fact a very close race), but what was really being referred to was a 3 million vote majority where no such vote could exist...

Thanks Anax. I needed that!

[OnTheOtherHand]

41. no offense, but what is the brilliant part?

I don't know who was expecting 110 to 112 million votes in a "'rerun of 2000.'" Interest in the 2004 campaign was much higher, and AFAIK everyone (including the pollsters who were tweaking their Likely Voter models) knew it. I respect anax's faith that a Big Middle turned out to swamp Bush, but I don't really understand it.

I can't quite figure out what the second argument here is. It seems to be an unsubstantiated assertion that some Republicans knew they Could Not win by 3 million votes, and so does anax. Umm, okay.

Believe what you want, but I see no signs that political observers woke up this morning and thought to themselves, "Migosh, turnout in 2004 was really high, I bet Bush stole the election after all!"

[mgr]

64. Illuminating strawman argument without a citation

I believe the more realistic Rovian stategy, would be: If one assumes there is no advantage to being the incumbent in the next election, then the only strategy left to us is to either increase our base, or increase our turnout. Given that our voters turnout out in very high numbers, our only reasonable strategy is to increase our base.

We know from Ohio that a republican GOTV effort was apparently underway since 2000. The corporate money available to them throughout the previous four years, exceeded the overall impact in 2004 than the monies available to the democrats during the election year. The impact of the democratic GOTV will be felt in 2006 and 2008 as Bush's foibles likely weaken conservative turnout.

Mike

[mgr]

63. It would be really nice to provide some links to support these contentions

I think you ask for such upthread.

http://polysigh.blogspot.com/2005/10/larry-bartels-on-whats-matter-with.html
http://www.emergingdemocraticmajorityweblog.com/donkeyrising/

Seems to undermine the conservative chipping away motif. The pattern in Cuyahoga where african americans voted for Kerry and for measure 1, as noted by Mebane also does this.

There are a lot of unsupported or unsubstantiated contentions here, particularly the one regarding who won the registration battles--the evidence appears that all the democrats did in Ohio was to regain the number of registered voters likely to vote democratic between the 2000 and 2002 purges, and did not increase their margin--their Ohio success was with a heavier turnout compared to 2000, not the larger base. The NY Times article on which it appears the contention that Dems out registered Reps was based appears to only focus on Pennsylvania and Ohio GOTV. Obtaining all the data is a bit daunting, but if the empirical evidence you have to support it is the exit poll results, then you are making a circular argument.

Mike

[tritsofme]

34. Who?

I'm 99.99982% certain that name doesn't ring a bell...

[autorank]

42. Many NEW features on INTERACTIVE MODEL

http://us.share.geocities.com/electionmodel/InteractiveElectionSimulation.xls

The model includes final pre-election and national poll data, interactive. The simulations are based on undecided voters and confirm the exit polls. A nice fit. (see new doc below)

This confirms that the professional pollsters (all documented) were on target. This would seem to argue against poll bias.

The undecideds go to challengers, almost always.

Here's the doc for the interactive model
--------------------------------------------------------------------

INTERACTIVE ELECTION MODEL
created by TruthIsAll

This model includes the following sheets:

Main Data input and summary analysis

StatePreExit Projections and analysis of pre-election and exit state polls
NationalPre Projections and analysis of 18 pre-election national polls

National Exit Analysis of National Exit Poll demographics at the following time lines:
a) Preliminary (13047 respondents) updated Nov. 3 at 12:22 am.
b) Final Exit Poll (13660 respondents) updated Nov. 3 at 1:25 pm.

Voted2000 Sensitivity Analysis of the National Exit Poll "Voted in 2000" demographic.

STATE POLLS
The analysis includes:
1) Kerry popular vote percentage/win probability based on pre-election state polls.
2) Kerry electoral win probability based on 200 Monte Carlo simulated election trials
3) Exit Poll percentages and deviations from the final recorded vote.

Enter calculation MODE = 1 to run the simulation/projection using final PRE-ELECTION polls.
The pre-election state polls are from Zogby, ARG, Gallup, etc.
Kerry's projected vote is calculated as his poll percentage plus an input undecided voter allocation.
Undecided voters traditionally break for the challenger by 60-80%. Adjust this margin up or down
to see the effects on popular and electoral votes.

Enter MODE = 2 to run the simulation based on EXIT polls. Compare to pre-election results.
Review the expected electoral vote and win probabilities.

Play "what if" by changing just two inputs: undecided voter allocation and cluster effect.
Find undecided voter allocation necessary for Kerry to win 50% of the popular vote
and 270 Electoral votes.

Enter the cluster effect as a percentage increment to the margin of error (MoE).
The MoE defines the range in which the final vote will fall with a 95% probability (confidence level).
For example, if a poll with a 3.0% MoE yields 51% for a candidate, then the probability is 95% that his/her vote will
fall in the range 48%-54%.

The cluster effect is a controversial factor added to the theoretical (calculated) MoE.
It is applied to account for the physical "clustering" of individuals with similar characteristics in precincts.
The number of states deviating beyond the exit poll MoE will decrease as the cluster effect increases.

Review the following simulation output:
Electoral and popular vote split and win probabilities.
Deviation probabilities for pre-election polls.
Deviation probabilities for exit polls.

Note that the number of states deviating beyond the exit poll MoE will
decrease as the cluster effect inflates the MoE.

NATIONAL POLLS
The average of 18 Pre-election polls predicted a virtual tie - which meant that
Kerry was poised to win, since historically the challenger wins a majority of the undecided vote.

The Preliminary National Exit Poll ( 12:22am, 13047 respondents) followed the 4pm ( 8349 ) and 7:33pm
( 11027 ) time lines. Kerry was leading at each of the three time lines.

The Final National Exit poll (13660 respondents) was posted at 1:25pm.
In the Final Exit Poll the weights and percentages were adjusted to match the recorded vote.

Ask "what if" by changing exit poll demographic weights and vote percentages.
You can also change the exit poll "cluster" effect.
Note how the popular vote split and corresponding deviation probabilities change.

Exit poll percentages do not all sum to 100% horizontally. This will not have a material
effect on the probabilities.
The demographics are calculated independently.

Key demographics to test:
Gender - Kerry lost 3% between the preliminary and final.
How Voted in 2000 (the 43% Bush/ 37% Gore weights are impossible).
Party ID - Preliminary was 38% Dem/35% Rep/37% Ind ; the final was changed to 37%/37%/36%

DEFINITIONS
Monte Carlo - a simulation technique applied to a mathematical model of a system to determine probable outcomes.
This simulation consists of 200 trial "elections" to determine the expected Electoral Vote and win probability.
The state win probability is based on the final exit poll split. A typical state poll consists of 600 sample-size with 4% MoE.

The Electoral Vote is calculated for Kerry and Bush in each of the 200 election trials,.
The average electoral vote is the arithmetic mean of the 200 trials.
The median EV (the middle value) is usually within a few votes of the average.

Margin of error - is based on poll sample size and given by the formula:
MoE = 1.96* Sqrt at the 95% confidence level, where P and 1-P is the likely characteristic split.

Normal distribution - calculates the probability of a win based on the poll spread and MoE.
Binomial distribution - calculates the probability of N states exceeding the MoE.

NORMAL PROBABILITY FUNCTION
For example, assume the following poll:
Size 3000 2-party vote
Kerry 49% 52.13%
Bush 45% 47.87%

MoE 1.79%
Cluster 20%
Adj. MoE 2.15%
StdDev =
MoE /1.96 1.09%

Probability of a Kerry win:
=NORMDIST(Kerry,Bush,StdDev,TRUE)
=NORMDIST(0.5213,0.4787, 2.15,TRUE)
Probability = 99.99%

BINOMIAL PROBABILITY FUNCTION
Calculate the probability that the MoE would be exceeded in AT LEAST N states.
This is 1 - the probability that N-1 is the MAXIMUM number of states in which the MoE was NOT exceeded.

= 1 - Probability that the MoE would NOT be exceeded in more than N-1 states.
= 1 - BINOMDIST(N-1,50,P,TRUE)

Let N= 16 The number of states which exceeded the state MoE for Bush.
P = 0.025 The probability that the vote discrepancy would exceed the MoE in a given state for Bush.

The Probability that the MoE would be exceeded in 2004 for Bush in 16 states is:

Prob = 1- BINOMDIST(15, 50, .025, TRUE)
Prob = 5.24025E-14 or 1 in 19.083 trillion

[jarnocan]

57. that TIA's web site and links

are great. http://www.progressiveindependent.com Check out his multimedia and other links. Awesome. Thanks again!
In TIA's honor, Andy's and for all of US- don't just do your usual stuff on NOV 2nd!!! because our www.worldcantwait.org !!!! DANG!
I just watched this anti- DU video on there. Abolish DU! (not us:) http://www.bushflash.com/pl_lo.html We are living in a scary world and BU**SH**Inc. plays a big part of much of the world's nightmares.

[autorank]

70. Those are great sites jarnocan! Thanks!!!