OPTIMIZER: Assume the PLAUSIBLE scenario that the exit polls were RIGHT

[TruthIsAll]

Let's assume the PLAUSIBLE scenario that the exit polls were
right.

Target Kerry to win 51.80%
Assume the given weighted average 53% response rate.

But assume there is NO response bias: 
Weighted average Alpha =1.0 

Restrict partisanship precinct response to 40-60%.

What is Kerry's:
Final Total Exit Poll percentage? 
Percentage in the five partisanship groupings?

We also want the optimizer to tell us:
WPE and Alpha for each of the five groupings, subject to the
weighted average constraints (53% AND 1.0).

Key results: 
1-Bush response was relatively heavy (K/B=.93) in the vote
rich non-partisan middle, but not in the partisan precincts.

2-Kerry won the exit poll with 51.44%, under his 51.80%
assumed vote.

Hmmm....



EXIT POLL RESPONSE OPTIMIZATION MODEL								
6/8/05 1:51 PM								
								
Objective: Determine values of constrained variables required
to derive a target								
Kerry/Bush percentage split using aggregate exit poll response
data.								
								
Precinct Variables Range Constraints:								
1-Response: within (Min, Max) and equal to weighted avg
rate.								
2-Kerry win percentage:within (Min, Max).								
3-Alpha (K/B): within (Min, Max) and equal to weighted
average.								
4-WPE within E-M actuals or (Min, Max). 								
								
TARGET INPUT								
Kerry 2-party vote		51.80%						
Bush 2-party vote		48.20%						
								
Wtd Avg Response		53.0%						
Wtd Avg Alpha (K/B)		1.00						
								
RESPONSE INPUT CONSTRAINTS								
								
1250	Strong Bush			Strong Kerry				
Prcts	40	415	540	165	90			
								
KERRY WIN%								
Min	0%	20%	40%	60%	80%			
Max	20%	40%	60%	80%	100%			
								
RESPONSE								
Min	40%	40%	40%	40%	40%			
Max	60%	60%	60%	60%	60%			
								
ALPHA								
Min	0.10	0.10	0.10	0.10	0.10			
Max	1000.00	1000.00	1000.00	1000.00	1000.00			
								
WPE								
E-M	-10.0%	-6.1%	-8.5%	-5.9%	0.3%			
Min	-100.0%	-100.0%	-100.0%	-100.0%	-100.0%			
Max	100.0%	100.0%	100.0%	100.0%	100.0%			
								
OPTIMIZER OUTPUT SUMMARY								
								
	PERCENTAGE SPLIT			VOTE SPLIT (mm)				
PCT	POLL	VOTE	Diff	POLL	VOTE	Diff		
Kerry	51.44%	51.80%	0.36%	62.24	62.68	0.43		
Bush	48.56%	48.20%	-0.36%	58.76	58.32	-0.43		
								
Bush needed	47.99%	of refusers to match his vote						
								
PRECINCTS						Wtd Avg	
Number	40	415	540	165	90			
Weights	3.2%	33.2%	43.2%	13.2%	7.2%			
								
Resp.	40.0%	51.6%	51.8%	60.0%	60.0%		53.0%	
Dev	-13.0%	-1.4%	-1.2%	7.0%	7.0%			
								
ALPHA								
K/B	1.25	1.02	0.93	1.05	1.09		1.00	
Dev	25.2%	2.1%	-6.5%	5.2%	8.9%		0%	
								
2-PARTY VOTE 								
Kerry	18.6%	38.6%	57.3%	59.8%	79.5%		51.80%	
Bush	81.4%	61.4%	42.7%	40.2%	20.5%		48.20%	
								
Votes	3.87	40.17	52.27	15.97	8.71		121.00	
Kerry 	0.72	15.52	29.97	9.54	6.93		62.68	
Bush	3.15	24.65	22.31	6.43	1.78		58.32	
Diff	-2.43	-9.13	7.66	3.11	5.14		4.36	

EXIT POLL								
Kerry	22.5%	39.2%	54.5%	61.5%	83.5%		51.44%	
Bush	77.5%	60.8%	45.5%	38.5%	16.5%		48.56%	
								
Votes	3.87	40.17	52.27	15.97	8.71		121.00	
Kerry 	0.87	15.75	28.51	9.83	7.28		62.24	
Bush	3.00	24.42	23.76	6.14	1.43		58.76	
Diff	-2.13	-8.66	4.75	3.69	5.84		3.49	

WPE	-7.8%	-1.2%	5.6%	-3.6%	-8.0%		0.72%	
E-M	-10.0%	-6.1%	-8.5%	-5.9%	0.3%		-6.77%	
Diff	2.2%	4.9%	14.1%	2.3%	-8.3%		7.5%	
								
								
								
OPTIMIZATION MODEL								
								
Categ.	HighB	Bush	Even	Kerry	HighK		Total/Avg	
Prcts	40	415	540	165	90		1250	
								
ALPHA								
K/B	1.25	1.02	0.93	1.05	1.09		1.00	
AvgDev	25%	2%	-7%	5%	9%		0%	
								
RESPONDERS								
Total	16	214	280	99	54		663	
Pct	40.00%	51.56%	51.76%	60.00%	60.00%		53.0%	
								
Kerry	4	84	152	61	45		346	
Pct	23%	39%	55%	62%	84%		51.44%	
Bush	12	130	127	38	9		316	
Pct	77%	61%	45%	38%	16%		48.56%	
								
REFUSERS								
Total	24	201	260	66	36		587	
Pct	60.00%	48.44%	48.24%	40.00%	40.00%		47.0%	
								
Kerry	4	76	157	38	26		301	
Pct	16%	38%	60%	57%	73%		52.0%	
Bush	20	125	103	28	10		286	
Pct	84%	62%	40%	43%	27%		48.0%	
								
VOTE								
Kerry	7	160	310	99	72		647	
Pct	18.6%	38.6%	57.3%	59.8%	79.5%		51.80%	
Bush	33	255	230	66	18		602	
Pct	81.4%	61.4%	42.7%	40.2%	20.5%		48.20%	
								
WPE								
Kv-Bv	-62.8%	-22.7%	14.7%	19.5%	59.0%		3.60%	
Kp-Bp	-55.0%	-21.6%	9.1%	23.1%	67.1%		2.88%	
								
WPE	-7.8%	-1.2%	5.6%	-3.6%	-8.0%		0.72%	
E-M WPE	-10.0%	-6.1%	-8.5%	-5.9%	0.3%		-6.77%	
								

[MrModerate]

1. Ya know . . . I'm sure this is wonderful, damning stuff . . .

But I can't make heads or tails of it. And neither can anyone else except statisticians and the like. Which doesn't characterize either 1) most Americans or 2) the people who need be influenced by strong proof that the election was stolen (again).

I think a little summary of the implications in narrative form wouldn't be out of place. In fact, without such a narrative summary, I'm afraid your efforts are a tad on the pointless side.

Just a suggestion, now.

[TruthIsAll]

2. Yes, but: the analysis is not geared to the general populace.

They don't know squat about response rates, alphas, precinct partisanship.

The debate is down to this:
Was it fraud?
or
was it rBr bias, as Mitofksy has theorized?

Hopefully, this model will help show that rBr is implausible, if not infeasible.

[WI Independent]

3. The problem is...

That ALL of this is dependent on the exit polls being right. For the exit polls to be right the pollsters had to know what they were doing in setting them up.

In proving the rBr theory implausible, you bring into question the competence of those that developed that theory. The problem is these are the SAME people that set up the polling methodology and weighting to begin with, therefore it casts doubt on their ability to do that properly either.

But they had to do that properly for any of your analysis to be good... round and round we go.

[TruthIsAll]

5. Come on, they KNEW what they were doing in setting them up.

Polling ability is not something that you LOSE.
It is NOT rocket science.
The methodology IS well-known.
Historical demographics ARE available.
Human nature does NOT change.

For you to assume the pollsters did not know what they were doing when the designed the poll is a pure canard.

The question is not the polling result; the question is why they made up the rBr fiction as an explanation for the discrepancy.

They WANT you to believe that rBr was the cause.
Do you believe it? If you do, then you must also NOT believe that the election was stolen.

[WI Independent]

7. Don't you see the conflict...

in trying to discredit someone using data that is as only as valuable as the same person's credibility? Seems like a type of circular logic... they must be wrong because they were right.

No, I don't really believe the rBr theory... but I also don't believe their original polling methodology has any more credibility than the rBr.

Was the election stolen? I don't know and I don't think analysing polls from now to eternity will answer that question definitively. Right now I believe what Kerry and others have said... there was fraud (or irregularities or whatever other nice word they choose to use), but it was not enough to change the outcome of the election.

Do YOU really think so many of the Democratic leaders are just plain ignorant of what really happened? Don't you have more faith than that in them? (serious questions, not trying to question your "Democraticness")

[LightningFlash]

8. Actually you are incorrect.

The Cencus ID weightings confirm that the election was stolen. There is compelling, long lasting evidence of the crime in effect. And if it isn't corrected, the same thing will transpire in 2006.

http://www.huffingtonpost.com/theblog/archive/jim-lampley/revisiting-the-biggest-st_1619.html
http://www.tucsonweekly.com/gbase/Currents/Content?oid=oid:69132
http://www.dispatch.com/news-story.php?story=dispatch/2005/02/12/20050212-B1-00.html
http://www.thousandreasons.org/get_article.php?article_id=13
http://www.buzzflash.com/interviews/2002/11/04_Palast.html
http://www.cronus.com/electionfraud/
http://www.freepress.org/columns/display/3/2005/1116
http://freepress.org/departments/display/19/2005/1284

As for why the DLC won't cover it? Well the DLC is corrupt, that's the real answer.

[Internut]

9. You have yet to show, even once

how the "cencus" (you were corrected a few dozen times, will you learn how to spell this word please) weightings in any way contradict the final exit polls. You're still welcome to offer that information if you have it.

[LightningFlash]

10. You've already been discredited.

There's no reason to go back over the census material.

As shown in the final I.D. weightings, 43% of the exit poll respondents were Bush voters from the year 2000. That was statistically impossible because at least 1.5 million bush voters died, and the gender weighting sample reversed by reducing the percentage. The final weighting is bogus, and you are going by nothing but rhetoric disproved by TruthIsAll and everyone else.

[Internut]

11. As someone here has shown already

the "who did you vote for before" questions in polls are notoriously unreliable.

If you go to http://webapp.icpsr.umich.edu/GSS/ click on "Analyze", pick the "Frequencies or crosstabulation (with charts)" option and give the variable PRES00, this is what you get:

Voted for:

GORE 44.7%
BUSH 50.6%
NADER 3.3%
Other .9%
Didnt .6%


That clearly demonstrates that people misreport their past voting behavior when asked in polls.

In addition to that, the census information does not contain anything about whom anyone voted for, whether in 2000 or in 2004. So what you wrote has nothing to do with "census material".

[LightningFlash]

12. Wrong, simply read the earlier thread.

It was completely layed out in bold detail. Just because some "right wing" talking points didn't awcknowledge it, won't change the fact it was already completely refuted. Please check "weighting by gender" catergory and assorted appendix.

[Internut]

13. Go ahead and give your argument - I will ask again

"how do the census survey figures contradict the final exit poll?".

No references to "previous posts". No "right wing talking points". Put up. Show the numbers and explain how they contradict.

[LightningFlash]

15. Read your own previous thread. I'm not even linking to his.

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=375544&mesg_id=375544

[Name removed]

17. Deleted message

Message removed by moderator. Click here to review the message board rules.

[TruthIsAll]

16. I am not questioning their competence.

You say:
In proving the rBr theory implausible, you bring into question the competence of those that developed that theory. The problem is these are the SAME people that set up the polling methodology and weighting to begin with, therefore it casts doubt on their ability to do that properly either.

I say:
That is totally false logic which reeks of naivete.
I do not question their competence.
I question their integrity.

Even Mitofsky says the methodology was correct.
But he cannot just come out and say the polls indicate Kerry won.
He is caught in a contradiction.
So he takes the safe way out.

"There must have been polling bias"
"Bush voters were shy".

What do you expect him to say?
He works for the networks which continue to protect Bush.
Do you expect him to say anything else?

He knows damn well that Kerry won this election.

[WI Independent]

19. Or, they just ran a sloppy exit poll...

and is using rBr to cover that. He couldn't very well say he screwed the pooch either. Seems an equally plausible explanation to me.

Incompetent OR dishonest... why should I believe him? And if I choose to believe him, which contradictory analysis should I believe?

What of the competence of Kerry and most other Democratic politician? Hardly any politician has even hinted publicly that this election was stolen. Do you think they are that ignorant of what happened or do you have another explanation for their silence?

[TruthIsAll]

20. Sloppy exit poll? No, sloppy spin on your part. n/t

/

[WI Independent]

21. You still haven't addressed my question...

"Hardly any politician has even hinted publicly that this election was stolen. Do you think they are that ignorant of what happened or do you have another explanation for their silence?"

This is a serious question. Numbers CAN lie. The silence of virtually all Demoratic politicians means most of what you're concluding doesn't pass the "sanity check" for me.

[TruthIsAll]

23. I have answered your question. And you are in denial.

You assume:
If there was fraud the Dems would have screamed bloody murder.

That is a well-known RW talking point.
Or you are very naive.
Or both.

Well, they had very little evidence early on.
And most Dems are a bunch of wimps.

The media would not touch it.
So they only a few had the gonads to stand up.

Barbara Boxer.
John Conyers.

The other Democrats, including Kerry, are loathe to yell "fraud".
But they know it was stolen.
They are not stupid.
The process takes time.

The evidence, in case you don't know it, is overwhelming.
Eight-six of 88 touchscreens turned Kerry voted to Bush.
The odds: 1 in 79,000,000,000,000,000,000,000

And over 98% of 50,000 election incidents favored Bush.

And on and on..

[WI Independent]

29. So I must be RW or naive if...

I don't except your perception of reality? I assure you I am neither.

No, I'm not a Democrat... that should be obvious by my screen name (kinda screwed that up though... should be little "i"). I generally vote for individuals, not parties. I'm pretty much middle of the road but can lean right or left depending on the issue.

As to naive... am I naive to not blindly accept the validity of polling and WEIGHTING methodolgy which the pollsters won't release?

I've never claimed there was not fraud... I'm just not convinced it rose to a level that changed the election or that it was a carefully orchestrated plan at the national level. I won't deny it COULD have happened... just not convinced it did.

And if Kerry really believed the election was stolen, he would find a way to use it... legally if there was proof... politically if there was no proof that met the legal standard.

[TruthIsAll]

30. More RW talking points. You admit there was fraud, but not enough...

to turn the election.

"I've never claimed there was not fraud... I'm just not convinced it rose to a level that changed the election or that it was a carefully orchestrated plan at the national level. I won't deny it COULD have happened... just not convinced it did".

Ok, a few questions for you.

Why was there ANY fraud, if not to turn the election?
Why bother stealing ANY votes, if not to turn the election?
Why not just it alone, if Bush was going to win any way?
Why steal votes when you don't need them?
Why bother strealing even one lousy vote?

Why cheat at poker if not to win?
Why cheat on your taxes if not to win?
Why go for a knockout in the final round if you have already won the fight on points?

Do you see how that silly talking point of yours falls apart when you ask yourself a few simple questions?

[WI Independent]

31. Just because you cheat at poker...

doesn't mean you wouldn't have won the hand anyway.

"Why was there ANY fraud, if not to turn the election?
Why bother stealing ANY votes, if not to turn the election?
Why cheat at poker if not to win?"
Of course it was the intent, but was it neccessary?

"Why not just it alone, if Bush was going to win any way?
Why steal votes when you don't need them?
Why bother strealing even one lousy vote?
Why go for a knockout in the final round if you have already won the fight on points?"
Poor arguments... all would have required the cheaters knew the outcome without their intervention. I guess they weren't comfortable with "too close to call".

"Why cheat on your taxes if not to win?"
Not sure how this one relates...

Do you believe that Kerry had NO fradulent votes? All your questions would apply equally to people cheating FOR Kerry as well.

[LightningFlash]

32. It means there's an impossible weighting at the end.

Gender reversal, showing women declined to vote for Kerry when the opposite was true according to http://exitpollz.net at the end of the night.

Population reversal where 43% turned out for Bush when the fact is at the minimum 1.5 million Bush voters had died. Can't aruge with numbers, AND still look straight faced.

https://voteprotect.org/index.php?display=EIRMapNation&cat=ALL&start_time=00%3A00&start_date=11%2F2%2F2004&end_time=23%3A00&end_date=5%2F5%2F2005&search=switched&go=Apply+filter

All incidents by catergory with "switched" votes favoring Bush, and on top of that full-scale election audits back up the exit polls. Fraud was all over Florida, and Kerry physically won Florida.

http://www.bbvdocs.org/FL/PalmBeach/Nov2004/pollsite01008-machine7301.PDF
http://www.bbvdocs.org/FL/PalmBeach/Nov2004/pollsite01006-machine6545.PDF
http://www.bbvdocs.org/FL/PalmBeach/Nov2004/pollsite01004-machine21285-oct.PDF
http://www.bbvdocs.org/FL/PalmBeach/Nov2004/pollsite01004-machine21284.PDF

Yes the infamous troll Bev Harris/bailey77 has proved that the exit polls were right in Florida. You were simply wrong, Florida had fraud which switched the election. Democrats in name only were involved.

[TruthIsAll]

34. Oh, so now you say "They all do it". Another RW talking point.

Your argument is specious, in tatters.

The massive documented evidence is that fraud (vote-switching, etc) was a Kerry vote to Bush phenomenon, not vice-versa.

The exit poll deviations to the vote favored Bush in 42 of 50 states.
All 22 states in the eastern time zone deviated to Bush.

16 states deviated BEYOND the margin of error to Bush.
None to Kerry.

Bush votes were inflated by 4 milion when you analyze the maximum number of Bush and Gore 2000 voters still alive who could have voted in 2004 and apply the equivalent National Exit Poll voting percentages to THESE TRUE weightings.

Of 88 documented touchscreen incident, 86 switched Kerry votes to Bush. The odds: 1/ 79,000,000,000,000,000,000,000

Do you care to cite your evidence of Kerry fraud?

[WI Independent]

35. It would explain why many Democrats aren't anxious...

to launch a full-scale investigation. Some lip service to it... but for the most part, no action. Do you have a more plausible explanation?

Search for Milwaukee voter fraud for an example that likely went Kerry's way. This appears to be voter fraud (carried out by individual voters) as opposed to election fraud (carried out by workers at some level), but the end result is the same (although likely not as successful).

All your evidence to support the exit polls are based on the exit polls themselves... My main disagreement with you is whether or not the exit polls prove anything at this point. IF the polling and WEIGHTING methodology is ever released and can be verified... it's a whole new ballgame.

I believe there has been some degree of fraud and honest mistakes in EVERY Presidential election. At this point I don't accept exit polls as proof that it rose to an election swinging level in 2004.

[TruthIsAll]

36. You are really reaching. All the evidence based on exit polls themselves?

You say:
"All your evidence to support the exit polls are based on the exit polls themselves".

You totally ignore the tons of documented evidence in the EIRS database.

You totally ignore the BBV machines made BY Republicans FOR Republicans with NO paper trail.

You totally ignore Blackwell.

You have shown your true colors.
You are in total denial of the facts.

There is no point in any further discussion.

[autorank]

41. Now why did you go do that "doesn't pass the "sanity check: for me."

Wow, "doesn't pass the 'sanity' check"...therefore, we're crazy. That's a common theme of transient visitors to this forum. Why would you say such a thing? Could it be slander?

[WI Independent]

46. Sorry, just noticed this...

when the thread was revived. No, I didn't mean TIA or other believers are insane. It's just an expression, not a accusation. I use it in reference to my own ideas that seem to make sense but fall apart when looked at from another angle.

[MostlyLurks]

4. Who is it geared to?

I agree with MrModerate and have, in the past, said something similar.

I wholly respect your tenacity and intellect, as you're way over my head with respect to statistical analysis. Put to good, industrious use I'm sure the data could prove foundational in a larger movement.

But if you're not directing its use to "normal" people (i.e. those like me and MrModerate) who is using it or what are you doing with it? That's not meant to sound confrontational - I'm genuinely curious as to how the findings are being dissemintated to those who *can* understand, and more importantly, use it.

Not to put words in MrModerate's mouth, but I think both of us have the same problem and related query: We'd like to understand this, even at a basic level, but there's no context in which the data is described, or summarized for us statistically-retarded citizens. Will it be at some point? Could it be, or is it just too complicated and intertwined with "specialized knowledge"?

Cordially,
MostlyLurks

[TruthIsAll]

6. Read USCV, or Lampley, or the threads here at DU.

You just have to get up to speed on the topic.
Have you had any statistics or probability courses?
If not, pick up "Statistics for Dummies" (no offense).

How about basic algebra?
That's a minimum.

[TruthIsAll]

44. Those are good points. See post # 18 for an explanation

I did not mean to blow you off. If I appeared to, it was unintentional.

[Peace Patriot]

14. There is no reason to disbelieve the exit polls, given so much OTHER...

...evidence of fraud (for instance 86 out of 88 reported incidents of touchscreens changing Kerry votes to Bush votes--how COULD that happen without fraudulent programming? --it couldn't!)(--and there is so much more!), and given exit polls' constant improvement over the years, and good record in recent years, in the US and elsewhere. They are standard election practice specifically used to detect fraud in other democratic countries. And, you think corporations don't want accurate polls on consumer habits? (They do!). Exit polls are like all polls, only much better--they poll actual voters as they leave the polling place.

What there IS reason to disbelieve is Edison-Mitofsky's lame and non-fact based "theory"--invented after the fact--that the reason Kerry won the exit polls is the Bush Republicans are shy creatures and didn't respond to being asked who they voted for.

Such hogwash! People who voted for Kerry in Republican districts--like those folks in Waynesville who were recently thrown out of their church for refusing to renounce their Kerry votes--are much more likely to have shunned a pollster in their neighborhood polling place than loudmouth, fascist Bushites who don't think anyone else is entitled to an opinion, let alone entitled to vote.

THAT's what we don't trust--after the fact excuses, with no basis in reality or the data.

E/M were the jerks who permitted the news monopolies to LIE to the American people about who won the exit polls. They put FALSE DATA on everybody's TV screens on election night!

And now they're lying to cover their asses. It's very simple: They did their exit polls in good faith, with time-tested methods. It showed a fraudulent election--or was/is strong evidence pointing in that direction. Somebody put the heat on (late night call to the networks by Karl Rove?). Now they're trying to cover it up--fuzz it over, muddle the issue.

And it stinks to high heaven.

As for TIA--their plea is legit, TIA. I, too, want to better understand WHAT you are saying, and as much of your method as I am able to. Don't slough it off. We are (or at least I am) committed activists who are trying to get the truth out, and we need help in translating your insights into the data, into terms that others can understand.

It's a bit snotty to say: go read a book! You don't have to write a treatise. Just a bit of context, and some terms, and the import of your findings.

For instance, what an optimizer is, why you are doing a "plausible" scenario that the exit polls are right, the significance of the "key results," and...

I had trouble with this: "under his 51.80% assumed vote." (Where did the assumed vote come from?)

You are doing these optimizer studies in a certain context. Just briefly give the context. (And explain your "Hmmm...". Does the result surprise you? Conflict with other data? Confirm something?)

It's up to you, of course. I'm just urging you to explain more--for my own sake, and that of others. I've been riveted by your posts, and have taken a lot of time and trouble to understand them, and convey your findings to others. You are good at explaining--when you take the trouble to do so (your posts were excellent in the Febble controversy, for instance--and have been in other cases). (Your "To believe Bush won, you have to believe..." series is superb!) We activists are constantly engaged with others who know less than we do--and it's vital, it's the heart of our democracy, really, that we communicate what is going on here, with the exit poll analyses, with the other 2004 election evidence, with the news monopolies, and with election reform. So, if you can, help us out.

(Or, if you're just thinking out loud here--just working something out--if it has no great significance to YOU, as yet--then just say so.)

[TruthIsAll]

18. Peace, you are the one of the most gifted writers on DU.

You must understand.

I am not trying to convince, nor educate the masses. I am just trying to provoke thought and discussion among those who are technically at a certain level. I am only adding my analysis to the debate between USCV, DUers and the naysayers.

As you know, I do many types of analysis. I have done my best to explain but cannot be expected to educate all lurkers. If they have a specific question, I will try to answer it, but I cannot try to explain the details. The naysayers and trolls would like to bog me down in minutia - and wear me out. That is their MO.

As for those who are sincere, I am just being very practical in telling them to get up to speed on the subject. I hope that is not taken as arrogance. But I do not have time to educate.

There is no question that some of this stuff is complicated; I had to struggle with the USCV paper the first time I read it, as well as with Febble's work. By the way, she is much better than I will ever be in her explanatory skills. I guess I just don't have as much patience or writing ability. That also applies to Ron Baiman - he is an excellent writer. I do my best, but I know my limits.

Let's face it. The crux of the matter is that the exit polls say that Kerry won. I have shown this in a myriad of ways by slicing and dicing the numbers. And we have shown that the probabilties that he did are very high

Dealing with rBr is more involved. Let's see if I can explain the Exit Poll Response Optimizer in a nutshell.

It takes the following exit poll data (the "constraints"):
1) percentage of responders to the poll (53%)

2) ratio of Kerry to Bush responders (56/50 or 1.12)

3) the five group weighting of partisanship precincts (High Bush, moderate Bush, Nonpartisan, moderate Kerry, High Kerry).

4) The Bush win percentage (51.23% of the 2-party vote).

5) The WPE (precinct "error" rate by partisanship). In other words the difference between the poll and the vote within each partisanship grouping (the total error is around 6.7%).

6) We restrict the variables to a min-max range in each precinct.
For example, the response rate may be limited to the 40-60% range in one or more precincts. Or it could be 30-70%, or 50-55%. The purpose of using ranhge limits is to provide feasibiltty to the results (we sure don't expect that any of the aggregate groupings would have a 5% response rate; most likely it will be between 48-58%)

Using this data, the model proceeds to calculate the required Kerry and Bush partisanship response rates for each of the 5 groups in order to satisfy the total weighted average input constraints.

In other words, it works backwards from the vote to calculate the the exit poll results, using the input error rates, applied to each of the precinct groupings. It determines what the responses have to be by precinct category to derive the Bush/ Kerry final vote (the "target").

We can also target Kerry as the winner, and assume that there is no reluctant response bias (K/B =1); that is, Kerry and Bush voters responded equally to the pollster. That is what I believe to be the more likely case; in fact, if there was anti-exit poll bias, it was more likely to have been reluctant republicans who voted for Kerry.

I hope that clears it up somewhat.

Try to read the output of the model. If there is anything you want clarified, let me know.

[Peace Patriot]

24. I can't tell you how much a statement like this helps me!

"In other words, it works backwards from the vote to calculate the exit poll results..."--TIA

I am almost 100% human language oriented (with a dash of talent in geometry). I am one of these people whose brain freezes when I am faced with a page of numbers and formulas. If I concentrate, and am able to get a good mental diagram of what the numbers/formulas may mean--for instance, "working backwards" (very helpful phrase), then I CAN comprehend the numerical--often quite well. But my initial response is to run for the hills!

I understand about lurkers and naysayers. I've followed your threads for a long time and I think I have some idea what you are up against. I am actually amazed at your patience, at times. But do keep in mind that any given "lurker" may be a potential freedom fighter (or a current one) who is trying to understand what's wrong with his/her country and looking for ammunition to fight back with. If they are in a nascent stage, your abruptness might turn off the revolutionary lightbulb that is trying to light up their minds. If they are already active, it might delay or stop their inquiry into the exit polls, thus depriving them of vital information. (Naysayers and mere troublemakers are another story--and I know it's hard to tell the difference sometimes.)

Also, I can grasp the analogy to my own work--creative writing (fiction, poetry) and public advocacy, often in written form, on sometimes quite complex issues. I tend to be comprehensive about facts and details, and I tend to see all their nuances, and I sometimes bristle at having to simplify (and over-simplify!) a complex matter, or explain it to people who are insensitive to facts, details and nuance, have other things on their minds, or are just intellectually lazy, or worse, have ulterior motives for not paying attention (such as corrupt government officials).

What I do with this need to simplify, to get over being irritated by it--and it is truly a need in some cases--is to regard it as an intellectual discipline (maybe akin to the way mathematicians reduce the elements of a problem to symbols in a formula). I think simplifying is also related to compassion--or, rather, to a desire to be compassionate. Intellectual types can go off the deep end, and plan bombing raids over the whole of the Vietnamese countryside (or today, the villages of Iraq), leaving their own humanity--and humanity in general--completely out of the picture. One of the smartest men in our country in the 1960s--Robert McNamara--did just that. So, anyway, simplification is a way of reigning in pure intellect, so that we (I) do not commit the sin of coldness and obliviousness. It is also a way of respecting other modes of being (other than the highly intellectual) and other ways of learning and wisdom.

That has been a big life lesson for me--learning that dancers and musicians and organic farmers and "Mother Earth" and "Father Earth" types have OTHER ways of knowing--and may indeed have tremendous intelligence, in its broadest meaning--that they cannot translate into words, and sometimes their mode of being, or mode of learning, makes it difficult for them to understand me!

Example: I find it very hard to respect any writer who makes frequent grammatical or spelling errors out of ignorance. Yet I've had to learn the meaning of the following: that William Shakespeare spelled his own name six different ways, because he didn't give a damn about spelling--it was IRRELEVANT to his genius! And the fact that there WERE no rigid rules of English in those days positively contributed to his genius.

So, what the hell am I doing being so "school marmish" about language? I've learned the value of language discipline--and it does have value--but I've also learned that there is value in tossing the rules right out the window, and allowing the FREE SPIRIT of language with all its joy and emotion and fundamental connectedness to the body and to the earth, to come forth. Balance between these two modes--intellect and body--is in fact the key to language genius, and perhaps to all happiness in this life--and we Westerners have a great deal of difficulty maintaining that balance.

And so, to take a long trail here back to my original request--that you explain these models more--it is simply a request, and it is absolutely not a demand, nor is it a criticism (except for the word "snotty"--which I think was fair re: "go read a book").

Please know that when you do explain, you explain well, and it is very helpful.

[Peace Patriot]

25. P.S., one thing that I think newbies need to know is this most basic of...

...assumptions of your work: it's not just that Kerry won the exit polls (that is acknowledged by the pollsters, and has been confirmed by USCV and others), but that the exit polls confirm a Kerry win in all other ways, when you look at the exit poll and official tally data from different angles. For instance, the east-west skew that you discovered. There is simply no innocent explanation for why the eastern states are more at variance with the exit polls than the rest of the country--and the most plausible explanation is that the Bush Cartel stole most of the votes there first, to get an early lock on the election in the earliest closing polls.

Another example is the sudden midnight switch from Kerry to Bush--the result flipping over, based on a mere 660 final exit poll respondents. As you have pointed out, to get that flip (from Kerry to Bush), they had to infuse impossible numbers of Bush voters into their exit poll/official tally mix.

Another basic item: That the pollsters PERMITTED the TV networks to MIX the two numbers (exit polls vs. official tally), thus denying the American people strong evidence of election fraud--Kerry's clear win of the exit polls (a great journalistic, as well as intellectual, crime); and that the pollsters have denied basic data to qualified researchers and to a senior Congressman, John Conyers, who is investigating the election--and have issued statements to try to explain Kerry's win of the exit polls (the "reluctant Bush responder" theory) that are unsupported by any data.

What your current model does is to try to create a realistic scenario, based on the known data and on reasonable inferences, to find out what the voter response rate was (to the exit pollsters) in this plausible scenario.

-------

QUESTION:

At the beginning, you say...

Key results:
1-Bush response was relatively heavy (K/B=.93) in the vote
rich non-partisan middle, but not in the partisan precincts.

This model seems to show that, yes, Bush voters responded (to exit pollsters) in relatively equal numbers to Kerry voters in the middle range of precincts (those that had about even numbers of Republicans and Democrats, and those with only a slight edge either way), but that when it comes to highly partisan precincts (presumably heavy Dem or Rep registration), voters become very reluctant to respond in the most Republican areas (60% refusal) and much more likely to respond in the Democratic areas (only 40% refusal).

So--if I'm reading the model correctly--I would re-state your key result this way: Bush response was relatively heavy (K/B=.93) in the vote rich non-partisan (equally divided) middle, as was Kerry voter response, but in the highly partisan precincts, voter response to the pollsters dips significantly (60%) in heavily Republican precincts and increases significantly in heavily Democratic areas (non-responders are only 40%).

In other words, the above response rates are what are shown in a realistic model of a Kerry win. (--i.e, no outsized or unexpected "reluctant Bush responder" phenomenon).

And if this is what really happened (more or less), it confirms intuitive and anecdotal information--that, a) some Republicans in heavy Republican areas don't trust outsiders (pollsters) but are much freer of that prejudice in the vote-rich middle precincts (mixing people more evenly produces more open-mindedness and less fear?); b) some voters in highly Republican precincts may have been afraid to state their vote publicly (to a pollster, at their neighborhood polling place)--and (my inference) those wouldn't likely be Bush voters, but rather Kerry voters surrounded by Bush voters (as with the folks recently thrown out of their church in Waynesville for refusing to renounce their votes for Kerry); and c) voters in mixed Dem/Repub areas, and high Dem areas, tend to be more open about how they voted (whoever they voted for), and represent the bulk or mainstream of the country (most of the precincts).

Am I reading these results of the Optimizer correctly?

I feel a little confusion and uncertainty with regard to the categories of "Bush voter" and "voter in a heavy Bush or Republican precinct." DOES this model say that Bush as well as Kerry voters in heavy Dem precincts were more open about saying who they voted for, while both Bush and Kerry voters in heavy Rep precincts were much less open--with the vote rich middle areas being a wash, as to response?

[Peace Patriot]

26. Nope, I think I just completely misstated your findings. Your goal was:

"What is Kerry's:
Final Total Exit Poll percentage?
Percentage in the five partisanship groupings?"

The first, given the constrants, Kerry 51.44 % in the exit poll (compared to 51.80% of assumed vote) (assumed from what?)

The second, I'll have to go back and read the numbers again.

I think I mistook your constraints for your goals. (Right?)

(They are nevertheless realistic constraints.)

[TruthIsAll]

28. The National Exit Poll and my electionmodel projection.

The National Exit Poll had Kerry at 51-48, which is 51.5% of the 2-party vote.

My projection model had it 51.8-48.2%.
http://www.geocities.com/electionmodel/

So that was the Goal.

[TruthIsAll]

27. Beautifully written. You are perfectly correct.

Your writing humbles me. Your interpetation of the numbers is right on.

I'm at work right now, so I will be brief.

IN THIS SCENARIO:
The percentage of responders increases from 40-60% as we go from Bush to Kerry precincts.

Kerry got 51.44% of total responders.
Kerry got 55% in the non-partisan middle, Bush 45%.

Kerry got 52% of total refusers, Bush the other 48%.
Refusers ranged from 60% to 40% (high Bush to High Kerry)
Kerry's percent of refusers ranged from 16% (high Bush) to 73% (high Kerry)

Bush got 61-77% of responders in Bush precincts.
Kerry 62-84% in Kerry strongholds.

[TruthIsAll]

22. 11 ALPHA SCENARIOS: BASE CASE (NO BIAS) ALPHA = 1.0

			ANALYSIS OF EXIT POLL RESPONSE				
			BY PARTISANSHIP FOR VARIOUS ALPHA (K/B)
	             AND CONSTANT RESPONSE RATE (R=53%)				
			
ASSUME KERRY WON: 51.8% - 48.2%				
							
			HIGH	SOLID	NON-	HIGH	SOLID
	AVERAGE		BUSH	BUSH	PART. KERRY	KERRY
K/B	0.95		1.25	0.97	0.87	1.03	1.08
Resp.	53.00%		40.00%	51.56%	51.76%	60.00%	60.00%
Kvote	51.80%		18.61%	39.10%	56.94%	59.84%	79.56%
Kpoll	49.52%		22.47%	38.21%	51.14%	60.93%	83.09%
WPE	4.55%		-7.71%	1.78%	11.59%	-2.19%	-7.07%
							
K/B	0.96		1.25	0.98	0.88	1.04	1.08
Resp.	53.00%		40.00%	51.56%	51.76%	60.00%	60.00%
Kvote	51.80%		18.61%	39.00%	57.02%	59.82%	79.55%
Kpoll	49.92%		22.48%	38.41%	51.84%	61.06%	83.18%
WPE	3.77%		-7.73%	1.17%	10.36%	-2.48%	-7.27%

K/B	0.97		1.25	0.99	0.89	1.04	1.08
Resp.	53.00%		40.00%	51.56%	51.76%	60.00%	60.00%
Kvote	51.80%		18.61%	38.90%	57.10%	59.80%	79.54%
Kpoll	50.30%		22.49%	38.61%	52.53%	61.18%	83.27%
WPE	2.99%		-7.76%	0.58%	9.15%	-2.76%	-7.46%
							
K/B	0.98		1.25	1.00	0.91	1.04	1.08
Resp.	53.00%		40.00%	51.56%	51.76%	60.00%	60.00%
Kvote	51.80%		18.60%	38.81%	57.18%	59.78%	79.54%
Kpoll	50.69%		22.50%	38.81%	53.21%	61.30%	83.36%
WPE	2.23%		-7.79%	-0.01%	7.95%	-3.04%	-7.66%
							
K/B	0.99		1.25	1.01	0.92	1.05	1.09
Resp.	53.00%		40.00%	51.56%	51.76%	60.00%	60.00%
Kvote	51.80%		18.60%	38.72%	57.26%	59.77%	79.53%
Kpoll	51.07%		22.51%	39.02%	53.88%	61.43%	83.45%
WPE	1.47%		-7.82%	-0.59%	6.75%	-3.32%	-7.85%
							

BASE CASE:
ALPHA = 1.0 

K/B	1.00		1.25	1.02	0.93	1.05	1.09
Resp.	53.00%		40.00%	51.56%	51.76%	60.00%	60.00%
Kvote	51.80%		18.59%	38.64%	57.33%	59.75%	79.52%
Kpoll	51.44%		22.52%	39.22%	54.54%	61.55%	83.54%
WPE	0.72%		-7.84%	-1.16%	5.57%	-3.60%	-8.04%
							




K/B	1.01		1.25	1.03	0.95	1.06	1.09
Resp.	53.00%		40.00%	51.56%	51.76%	60.00%	60.00%
Kvote	51.80%		18.59%	38.55%	57.40%	59.74%	79.52%
Kpoll	51.81%		22.53%	39.42%	55.20%	61.67%	83.63%
WPE	-0.03%		-7.87%	-1.73%	4.40%	-3.87%	-8.24%
							
K/B	1.02		1.25	1.04	0.96	1.06	1.09
Resp.	53.00%		40.00%	51.56%	51.76%	60.00%	60.00%
Kvote	51.80%		18.59%	38.48%	57.46%	59.72%	79.51%
Kpoll	52.18%		22.54%	39.62%	55.84%	61.80%	83.72%
WPE	-0.77%		-7.90%	-2.28%	3.24%	-4.15%	-8.43%
							
K/B	1.03		1.25	1.05	0.98	1.06	1.10
Resp.	53.00%		40.00%	51.56%	51.76%	60.00%	60.00%
Kvote	51.80%		18.58%	38.40%	57.52%	59.71%	79.50%
Kpoll	52.55%		22.55%	39.82%	56.48%	61.92%	83.81%
WPE	-1.50%		-7.93%	-2.83%	2.09%	-4.42%	-8.62%
							
K/B	1.04		1.26	1.06	0.99	1.07	1.10
Resp.	53.00%		40.00%	51.56%	51.76%	60.00%	60.00%
Kvote	51.80%		18.58%	38.33%	57.58%	59.70%	79.50%
Kpoll	52.91%		22.56%	40.02%	57.11%	62.04%	83.90%
WPE	-2.22%		-7.95%	-3.38%	0.95%	-4.70%	-8.81%
							
K/B	1.05		1.26	1.07	1.00	1.07	1.10
Resp.	53.00%		40.00%	51.56%	51.76%	60.00%	60.00%
Kvote	51.80%		18.58%	38.26%	57.64%	59.68%	79.49%
Kpoll	53.27%		22.57%	40.22%	57.73%	62.17%	83.99%
WPE	-2.94%		-7.98%	-3.91%	-0.18%	-4.97%	-9.00%

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


		WPE SENSITIVITY TO ALPHA				

K/B	AVG	HB	B	BK	K	HK
 0.95 	4.55%	-7.71%	1.78%	11.59%	-2.19%	-7.07%
 0.96 	3.77%	-7.73%	1.17%	10.36%	-2.48%	-7.27%
 0.97 	2.99%	-7.76%	0.58%	9.15%	-2.76%	-7.46%
 0.98 	2.23%	-7.79%	-0.01%	7.95%	-3.04%	-7.66%
 0.99 	1.47%	-7.82%	-0.59%	6.75%	-3.32%	-7.85%

 1.00 	0.72%	-7.84%	-1.16%	5.57%	-3.60%	-8.04%

 1.01 	-0.03%	-7.87%	-1.73%	4.40%	-3.87%	-8.24%
 1.02 	-0.77%	-7.90%	-2.28%	3.24%	-4.15%	-8.43%
 1.03 	-1.50%	-7.93%	-2.83%	2.09%	-4.42%	-8.62%
 1.04 	-2.22%	-7.95%	-3.38%	0.95%	-4.70%	-8.81%
 1.05 	-2.94%	-7.98%	-3.91%	-0.18%	-4.97%	-9.00%

[kster]

33. Kick.nt

[TruthIsAll]

37. Here is the latest Updated Model for slight fix to refuser rates

EXIT POLL RESPONSE OPTIMIZATION MODEL							
6/10/05 12:13 PM							
							
OBJECTIVE:							
Determine values of constrained variables required to derive a
target Kerry/Bush percentage split using aggregate exit poll
response data.							
							
Precinct Variable Input Range (Min, Max) Constraints: 							
1-Response rates: will equate  to input weighted
average							
2-Kerry 2-party win percentages							
3-Alpha (K/B): will equate to input weighted average							
4-WPE: must input  (optional: set Min= Max= E-M)							
							
OPTIMIZER TARGET VOTE (2-party)							
Kerry		51.80%					
Bush		48.20%					
							
WEIGHTED AVERAGE RESPONSE							
Overall Response Rate		53.0%					
Alpha (K / B)		1.00					
							
RESPONSE INPUT CONSTRAINTS							
							
1250	Strong Bush		Strong Kerry				
Prcts	40	415	540	165	90		
							
KERRY WIN%							
Min	0%	20%	40%	60%	80%		
Max	20%	40%	60%	80%	100%		
							
EXIT POLL RESPONSE							
Min	30%	30%	30%	30%	30%		
Max	70%	70%	70%	70%	70%		
							
ALPHA (K/B)							
Min	0.10	0.10	0.10	0.10	0.10		
Max	10.00	10.00	10.00	10.00	10.00		
							
WPE							
E-M	-10.0%	-6.1%	-8.5%	-5.9%	0.3%		
Min	-10.0%	-10.0%	-10.0%	-10.0%	-10.0%		
Max	10.0%	10.0%	10.0%	10.0%	10.0%		
							
OPTIMIZER OUTPUT SUMMARY							
							
			EXIT POLL / VOTE DEVIATION 				
	Percentage			Amounts (mm)		
	Poll	Vote	Diff		Poll	Vote	Diff 
Kerry	51.59%	51.80%	0.21%		62.42	62.68	0.26
Bush	48.41%	48.20%	-0.21%		58.58	58.32	-0.26
							
PRECINCTS	
Categ.	HighB	Bush	Even	Kerry	HighK		Total/Avg
Number	40	415	540	165	90		
Weight	3.2%	33.2%	43.2%	13.2%	7.2%		
							
EXIT POLL							
Kerry	22.0%	40.0%	53.1%	64.0%	86.3%		51.59%
Bush	78.0%	60.0%	46.9%	36.0%	13.7%		48.41%
							
RESPONSE	
       48.0%	49.9%	53.6%	57.4%	57.8%		53.00%
WtdDev	-5.0%	-3.1%	0.6%	4.4%	4.8%		0%

ALPHA							
K/B	1.20	1.04	0.94	1.07	0.97		1.00
Dev	20.1%	3.7%	-5.8%	6.5%	-3.2%		0%

2-PARTY VOTE 							
Kerry	20.0%	39.3%	54.7%	62.0%	87.7%		51.80%
Bush	80.0%	60.7%	45.3%	38.0%	12.3%		48.20%
VOTES (mm)							
Kerry 	0.77	15.78	28.59	9.90	7.64		62.68
Bush	3.10	24.40	23.68	6.07	1.07		58.32
TOTAL	3.87	40.17	52.27	15.97	8.71		121.00
Diff	-2.32	-8.62	4.90	3.83	6.57		4.36


EXIT POLL (mm)							
Kerry 	0.85	16.07	27.76	10.23	7.52		62.42
Bush	3.02	24.10	24.51	5.75	1.19		58.58
TOTAL	3.87	40.17	52.27	15.97	8.71		121.00
Diff	-2.17	-8.04	3.24	4.48	6.32		3.84
							
WPE	-4.0%	-1.5%	3.2%	-4.1%	2.8%		0.42%
E-M	-10.0%	-6.1%	-8.5%	-5.9%	0.3%		-6.77%
Diff	6.0%	4.6%	11.7%	1.8%	2.5%		7.2%
							
							
							
							
							
							
OPTIMIZATION MODEL							
							
Categ.	HighB	Bush	Even	Kerry	HighK		Total/Avg
Prcts	40	415	540	165	90		1250
Kerry	20.0%	39.3%	54.7%	62.0%	87.7%		51.80%
							
Kerry	8	163	295	102	79		647
Pct	20.0%	39.3%	54.7%	62.0%	87.7%		51.80%
Bush	32	252	245	63	11		602
Pct	80.0%	60.7%	45.3%	38.0%	12.3%		48.20%
							
ALPHA							
K/B	1.20	1.04	0.94	1.07	0.97		1.00
AvgDev	20%	4%	-6%	7%	-3%		0%
							
RESPONDERS							
Total	19	207	289	95	52		663
Pct	48.0%	49.9%	53.6%	57.4%	57.8%		53.00%
							
Kerry	4	83	154	61	45		346
Pct	22.0%	40.0%	53.1%	64.0%	86.3%		51.59%
Bush	15	124	136	34	7		316
Pct	78.0%	60.0%	46.9%	36.0%	13.7%		48.41%
							
REFUSERS							
Total	21	208	251	70	38		587
Pct	52.0%	50.1%	46.4%	42.6%	42.2%		47.00%
							
Kerry	4	80	142	42	34		301
Pct	18.1%	38.5%	56.5%	59.3%	89.6%		52.07%
Bush	17	128	109	29	4		286
Pct	81.9%	61.5%	43.5%	40.7%	10.4%		47.93%
							
							
WPE							
Kv-Bv	-60.0%	-21.5%	9.4%	24.0%	75.4%		3.60%
Kp-Bp	-56.0%	-20.0%	6.2%	28.0%	72.6%		3.18%
							
WPE	-4.0%	-1.5%	3.2%	-4.1%	2.8%		0.42%
E-M WPE	-10.0%	-6.1%	-8.5%	-5.9%	0.3%		-6.77%
							
							

[TruthIsAll]

38. KEY RESULT: AVERAGE WPE = 0 .42% (NO FRAUD)

OPTIMIZER TARGET VOTE (2-party)							
Kerry		51.80%					
Bush		48.20%					
							
WEIGHTED AVERAGE RESPONSE							
Overall Response Rate		53.0%					
Alpha (K / B)		1.00	

[helderheid]

39. kickety kick

[autorank]

40. Spank it, then Kick it

:spank::kick:

Contact the DNC and Give 'em Hell About NOT Acting on Election Fraud

NEW LEADERS FOR A NEW DEMOCRATIC PARTY

[kster]

42. Thanks TIA AND KICK.NT

[tommcintyre]

43. kick

[kster]

45. kick.nt