TIA: Who still believes the 2006 vote counts were accurate?

[mom cat]

TIA: Who still believes the 2006 vote counts were accurate?

Then you are invited to a free vacation in Sarasota to check
out the voting machines.

This is a summary comparison of the reported National vote vs.
120 pre-election Generic Poll model projection, preliminary
(7:07pm) and Final National Exit Polls.

There is a major discrepancy in the reported CBS News national
vote total from in the EDA summary table and the state totals
in the detailed table. The National vote totals
(57.7%D-41.8%R) are also posted on Wikipedia, but the data
source is not indicated.

What is the explanation for these obvious discrepancies? We
know that there is incontrovertible evidence of reported
incidents where votes were miscounted or lost (FL-13, IL-6,
FL-24, OH-2, etc.). And voter disenfranchisement, which
amounted to millions of votes, is a major factor apart from
the miscounted votes. 

The bottom line is that the Democrats won by a much bigger
margin than indicated the reported vote count indicates. If
the Wikipedia reorted vote is correct, then the Democrats
received 57.7% + X%, where X =3-5%.

But it's impossible to know the actual vote count - there is
no paper trail. It's a sad commentary on the lack of
transparency in our elections. That's why it ranks far from
the top of world-wide democratic voting systems. Just ask
Jimmy Carter. Of course, the voting machine vendors think
everything is just fine and say the machines performed
brilliantly.

Is the corporate media (NYT, MSNBC, etc) finally waking up to
the extent of the problem? And what about the Democrats? Will
they just glow in the results of an election in which 15-20
House seats were probably stolen and two Senate seats were
almost stolen? 
____________________________________________________ 

This is a summary table of polls and reported votes. Details
and links are provided below. 

Summary .... Votes Dem .. Rep. Other Margin Dem . Rep . Other
Average .... 74.95 40.33 33.22 1.40 . 7.11. 53.8% 44.3%. 1.9%
Pre-election 

Generic-10.. 76.58 42.96 32.09 1.53. 10.87. 56.1% 41.9%. 2.0%

National Exit Poll 
CNN-7:07pm . 76.58 41.05 34.40 1.13.  6.65 . 53.6% 44.9%. 1.5%
CNN-Final... 76.58 40.24 34.43 1.91 . 5.81 . 52.6% 45.0%. 2.5%
NYT (adj)... 76.58 40.65 34.40 1.53 . 6.25 . 53.1% 44.9%. 2.0%
(I adjusted the NYT National Exit Poll results slightly to
include Other).

Reported National Vote
Wikipedia... 68.06 39.27 28.46 0.32. 10.80 . 57.7% 41.8%. 0.5%
CBS(11/09).. 76.58 40.32 34.57 1.69 . 5.76 . 52.7% 45.1%. 2.2%
CBS State... 73.69 37.80 34.20 1.69 . 3.60 . 51.3% 46.4%. 2.3%

Average 
Generic-10.. 76.58 42.96 32.09 1.53. 10.87 . 56.1% 41.9%. 2.0%
NatExitPoll. 76.58 40.65 34.41 1.52 . 6.24 . 53.1% 44.9%. 2.0%
ReportedVote 72.78 39.13 32.41 1.24 . 6.72 . 53.9% 44.5%. 1.7%

____________________________________________________ 

Final 10 Generic Polls
---------------------
Poll.... Date . Dem .. Rep 
Harris.. 1023 .. 47 .. 33 
AP...... 1030 .. 56 .. 37 
CBS..... 1101 .. 52 .. 33 
Nwk..... 1103 .. 54 .. 38 
Time.... 1103 .. 55 .. 40 

Pew..... 1104 .. 47 .. 43 
ABC..... 1104 .. 51 .. 45 
USA..... 1106 .. 51 .. 44 
CNN..... 1106 .. 58 .. 38 
FOX..... 1106 .. 49 .. 36 

Average........ 52.0%. 38.7% 
Adj UVA......... 4.1% . 3.2% 
Projected ..... 56.1%. 41.9% 

Reported....... 51.3%. 46.4% 
Deviation...... -4.8%. +4.9% 

____________________________________________________ 

CNN National Exit Poll (7:07pm)
Sample-size: 13,208
MoE: 0.87% 

GENDER 
...........Mix Dem . Rep . Other
Male...... 48% 51% . 47% . 2%
Female.... 52% 56% . 43% . 1%
Total.... 100% 53.6% 44.9% 1.5%

CNN NATIONAL EXIT POLL (1:00pm) 
Sample-size: 13,251
http://www.cnn.com/ELECTION/2006/pages/results/states/US/H/00/epolls.0.html

GENDER
...........Mix Dem . Rep . Other
Male...... 49% 50% . 47% . 3%
Female ... 51% 55% . 43% . 2%
Total.... 100% 52.6% 45.0% 2.4%

____________________________

NYT National Exit Poll
http://graphics8.nytimes.com/images/2006/11/08/us/politics/1109-nat-webPOLL.gif

GENDER 
...........Mix Dem . Rep . Other
Male...... 48% 51% . 47% . 2%
Female ... 52% 55% . 43% . 2%
Total.... 100% 53.1% 44.9% 2%

____________________________

Wikipedia
---------
Dem 39.267mm 57.7%
Rep 28.464mm 41.8%
http://en.wikipedia.org/wiki/United_States_House_election,_2006
____________________________

CBS 11/09
---------
(reported by the Election Defense Alliance-EDA)

http://electiondefensealliance.org/landslide_denied_exit_polls_vs_vote_count_2006
http://electiondefensealliance.org/files/A1.gif
http://electiondefensealliance.org/files/A2.gif
http://electiondefensealliance.org/files/A3.gif
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=364x2775205

The CBS summary vote totals differ from the state totals by
2.902mm votes.
Of this difference, 2.527mm are Democratic votes and 0.371mm
are Republican. 

It's a Supreme Mystery:
Baby, baby, where did the 2.9mm votes go?

Reported Vote (millions)
CBS........... Total Dem ....Rep ..... Other
Summary...... 76.582 40.323 . 34.565 . 1.694
Detail....... 73.680 37.796 . 34.194 . 1.690
Deviation..... 2.902. 2.527 .. 0.371 . 0.004

____________________________

The projected 10-Generic poll 14.2% Democratic margin deviated
by -9.3% from the CBS Reported margin, a 6.681mm vote
difference. The corresponding Democratic vote share deviated
by -4.8%.

U.S. House 
(in thousands)
............... Dem Repub Other Dem Repub Other Marg
VoteMargin
Generic.......41334 30872 1474 56.1% 41.9% 2.0% 14.2% 10463
Reported..... 37796 34194 1690 51.3% 46.4% 2.3%. 4.9% 3602
Deviation ....-3538. 3322 216. -4.8%. 4.5% 0.3% -9.3% -6861

____________________________

MATCHING THE NATIONAL EXIT POLL TO THE FINAL GENERIC-10:

In the following tables, National Exit Poll weights and vote
shares were adjusted in order to MATCH the Generic 10-poll
projection. Since the Final NEP is ALWAYS matched to the
REPORTED vote and virtually everyone agrees that the reported
vote does NOT reflect the TRUE vote, we are surely justified
in matching it to the final 10-Generic poll projection.

Final 10-Generic poll projection
.....................Dem . Rep. Other 
Generic-10......... 56.1% 41.9% 2.0% 

National Exit Poll
(adjusted to match the GP)

VOTED 2004 
........Mix Dem . Rep Other
Kerry.. 50% 92% . 6%. . 2% 
Bush... 45% 15% . 83% . 2% 
Other... 1% 76% . 22% . 2% 
NoVote...4% 65% . 32% . 3%
Total. 100% 56.1% 41.9% 2.0% 

PARTY ID 
........Mix Dem . Rep .. Other
Dem.... 40% 93% . 6% ... 1% 
Rep ... 33% 12% . 86% .. 2% 
Ind ... 27% 55% . 41% .. 4% 
Total. 100% 56.0% 41.9%. 2.1% 

GENDER 
........Mix Dem . Rep .. Other
Male... 47% 54% . 44%... 2% 
Female. 53% 58% . 40%... 2% 
Total .100% 56.1% 41.9%. 2.0%

[The Count]

1. Some still believe kerry lost in 2004, a few , Gore lost in 2000...

I am convinced most don't want to think of this uncomfortable topic anymore....But I thank you!

[mom cat]

11. Perhaps 650,000 Iraqis would be alive today if we did think about

it. So much damage has been done by the election thefts. So much of it can never be undone.

[goclark]

50. Kicked and Bookmarked

[mom cat]

15. ***** TIA: DIRECT LINKS TO DATA SOURCES *****

TIA: DIRECT LINKS TO DATA SOURCES


FINAL CNN NATIONAL EXIT POLL (1:00pm)
http://www.cnn.com/ELECTION/2006/pages/results/states/US/H/00/epolls.0.html

NYT National Exit Poll


Wikipedia
http://en.wikipedia.org/wiki/United_States_House_election,_2006

CBS
(reported by the Election Defense Alliance-EDA)

http://electiondefensealliance.org/landslide_denied_exit_polls_vs_vote_count_2006







http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=364x2775205

[EST]

2. Never in doubt.

Give 'em hell!

[Roland99]

3. "it's impossible to know the actual vote count - there is no paper trail."

:(

[mom cat]

12. Precicely.

[slackmaster]

17. Depends on where you are

Here in California there are always paper trails.

[Roland99]

22. Not on a national level. There can be no complete paper trail.

[slackmaster]

51. I can't think of any office that we actually elect nationally

President comes close, but there is still that Electoral College layer in there. We have 50-some separate elections to choose electors. But that is beside the point - There were no national issues in the 2006 election.

My state and many others are not messed up. The ones that are (Ohio and Florida come to mind) need to get their shit together. Federal law needn't dictate the details of how they do it, but it could require a standard of auditability that those states and no doubt others have not achieved.

I don't support requiring all of the states to conduct their elections in any specific manner, but would support a federal law requiring a paper audit trail. DRE machines with no paper backup are unacceptable.

[Roland99]

70. Hand-counted. Paper. Ballots.

The only acceptable way.

[nashville_brook]

4. it was intimated in another thread that the Margin Of Theft was based on pre-October

polling numbers (then fed into electronic voting machines via last minute software updates) and therefore fell short of what was needed to overcome popular will.

this is an interesting twist; that the fraudsters are limited by time in their ability to access the software.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x460564

Rural Counties like Hill, with little technical expertise, are easy targets for this type of manipulation, but it does not stop there!!! ALL of the voting machines were deployed throughout the country pre-programmed with a formula that Karl Rove was probably talking about when he said he had "...done the math...". The logistics of pre-programming these machines meant that the formula had been done based on the "math" in early October at the latest. By the time early voting and E-Day rolled around, the mood in the country had changed considerably, but it was too late to re-program unnoticed. All across the country the voters beat the percentage or got close. Tons of money (that we can never match) was poured into races that just didn't make sense, but for only the reason to give the impression that IT made the difference. For example, one of the Republican Commissioner candidates here spent over $8000, with "...little of his own money...", and the balance coming from "donations".

for me, this explains a lot AND clearly suggests that something precious was stolen even if Congress as a whole was won -- we've lost our ability to KNOW OUR WILL. i've said this a dozen different ways since 2004, but the most insidious theft is the theft of our IDENTITY.

if Congress was really won with significantly higher numbers than before (if 2004 belonged to Kerry and 2000 belonged to Gore) WE'VE LOST OUR SOUL. we have no way of knowing ourselves anymore -- we can only be who the people who own the machines say we are.

and fuck if that doesn't piss me off.

[EFerrari]

9. "the theft of our identity"

That's a very powerful way to put it.

Agreed.

[mom cat]

13. It pisses me off royally too! A larger margin in the House would make the difference in hundreds of

bills and perhaps in a vote to impeach Bush/Cheney. Too much is at stake to dismiss the seats that were stolen.

[tom_paine]

20. That gets to the heart of our national voting untrustworthiness

And I agree that 2006 election, while suggestive that perhaps the problem is not so bad as worst case, does not mitigate the problem that our electoral system needs to be ovrehauled and receive a strong dose of checks and balances to insure turstworthiness and prevent election fraud.

[Peace Patriot]

28. One of my theories about the 2004 election was that the electronic voting

systems had to be pre-programmed to certain percentages, that Rove & Co. were caught off guard by the size of Kerry's win, and that's why Ohio happened. They had the Republican election theft machine all primed to go in Ohio, if needed, as Plan B. I had wondered to myself why, if they have TRADE SECRET, PROPRIETARY programming code in all the new and extremely insider hackable voting machines and central tabulators--controlled by Wally O'Dell and Howard Ahmanson (80% of the nations' vote)--did they risk the blatant, egregious, broad-daylight Voting Rights Act crimes in Ohio? Why not just tweak the code a little more? And I think the answer IS access. It's not so easy to change the programming code on election day.

It was just a guess. They couldn't count on the Democratic Party leadership just ignoring the broad daylight robbery of black, student and other Democratic votes in Ohio. So it was a risky business. And, even with John Ashcroft as A.G., providing immunity from any prosecution, still, it looks really bad, and, even if they could count on the war profiteering corporate news monopolies to place an "Iron Curtain" over the whole thing, Ohioans might get uppity about it, and kick up a fuss (and some tried damned hard, bless them forever!!!). So why do it?

Answer: They had to. I toyed with the idea that they did it out of pure malice. They had the power to stomp all over black and poor voters, so they did. Smash 'em down. Crush all the little people. Beat 'em up so bad that they can never rise up to vote Democratic again. And I don't doubt that Bushites get their thrills stomping on the helpless for no reason at all. But I'm pretty convinced now that it was necessity. TIA's latest, here, helps confirm it. Likely, they underestimated the American people's determination to throw them out. The machines (and central tabulators) have to be pre-programmed to steal, switch or eat votes, in the factory or opportunistically during "servicing" (or maybe they do it when they conduct their secret industry "testing" of the machines--a nice irony), but there are too many safeguards and eyes watching on election day to alter the code to meet changing circumstances. So, with Kerry surging to victory--even with the secret formulae in place--Plan B had to be activated. (I've sometimes wondered, too, how it could be predicted, ahead of time, that it would "all come down to Ohio"? Everybody kept saying this, and I kept wondering--why? The answer may be that Rove/Diebold-ES&S directed it there, into Blackwell's corrupt jurisdiction--a state that Rove could count on to steal, suppress, switch, disappear, or throw away as many votes as needed--and Rove planted the "meme" with the usual suspects.)

I have my thoughts about Plan C, which I won't go into here. (Okay, it has to do with Cheney running off to Hawaii two days before the election--odd choice, Hawaii, not a lot of Republican votes there--and Cheney's plane being out over the Pacific on election day; all those phony "terrorist" alerts seeded into the newsstream leading up to election day, including a really odd one involving a school in Los Angeles; and, if you wanted to shut down the voting on the west coast, or seriously disrupt it, how you would do it, combining these elements?)

But Plan A and Plan B combined well enough to flip a 3% Kerry win (reflected in the exit polls) into a 2.5% Bush win (as "tabulated" by Diebold/ES&S's secret formulae). Kerry probably won by more like a 5% margin, if all those who wanted to vote had been able to. The 3% margin is the poll of those who made it to the voting booth and actually voted. (Greg Palast estimated that a million black voters had been purged from the voting rolls before the election ever started.)

Another thing that was happening that year was CA Sec of State Kevin Shelley's lawsuit against Diebold, which included a demand to see Diebold's source code. This was in late spring 2004, six months before the election. What strikes me about the whole arc of the Shelley story now is the Dark Lords' desperation to get him out of office before he could go any further. He was clearly onto Diebold. He had decertified their touchscreens. (I think now he must have been suspicious about the Recall, too.) And he certainly would have pursued it--he is a bulldog. So they had to cripple him--ruin his career, his credibility, drive him out in disgrace. And that's what they did. That's how important it was to them to NOT have a smart Sec of State opening up their machines.

ES&S is playing games about the code, too--in FL-13--saying they may or may not cooperate with examination of their code (--in the most blatantly fraudulent election of them all!) (--even the NYT thinks it was bad!). (18,000 "undervotes", the bulk of them Democratic votes, who voted for all the Democrats, but couldn't be bothered to vote in the hottest Congressional race in years.)

I have insisted on these statistical threads that CONTEXT is important. If known criminals have the means to commit a crime, undetectably--and have also gone out of their way to set up the conditions for the crime--in this case election theft--then you have to start with the presumption that election theft has occurred, and use what tools you can to figure it out--as TIA has so magnificently done. It violates logic and common sense to presume that vote tabulation with secret formulae, owned and controlled by people with very close ties to the Republican Party and far rightwing causes, is correct, by default. That is "faith-based" voting. It's nuts. The opposite presumption should be made: Means--yes. Motive--yes. Opportunity--yes. Known criminal history--yes. All in spades. Easy, undetectable means. Massive coverup of other crimes, and the opportunity for the more massive looting of our treasury, provide plenty of motive. Opportunity--every opportunity, from the factory to selective "machine breakdowns"--all controlled by rightwing corporations. Known criminal history--100,000 murders, at least.

Another test is the political landscape around the election theft, and resulting from the theft. In this case, Bush's 30% approval rating (and rarely over 40% for a two year period--since his "re-election" in fact), and all the other many indicators of citizen disgust and rebellion (including the huge increase in requests for Absentee Ballots for these midterms--indicating vast distrust of the machines and their "results"). SEVENTY-PERCENT of the people wanting the Iraq War ended. EIGHTY-FOUR PERCENT wanting no part of a widened Mideast war.

And the result? "Impeachment is off the table." Bush, Cheney, Abazaid and others STILL saying "stay the course." Congress critters discussing various plans to "withdraw" U.S. troops a year from now, or two years from now (to nearby emirates). And a group of Democrats in Congress, like the ones who voted for torture and suspension of habeas corpus last month--Bushite Democrats--poised to block moves to end the war, to reduce funding for the war, to impeach the perpetrators of the war, to enforce subpoenas against the perpetrators, and to achieve any real reform. Even with the Democratic wins, the new Congress is still not very representative of the American people. And you have to ask why. And, if you have a nearly completely non-transparent vote counting system run by rightwing corporations--in both the PRIMARIES and the general election--you're really damned stupid not to start there. (I really think the primaries need close scrutiny, too. Who was selected out?) I heard the Iraq vets didn't do well--but I don't have stats on it. If I was a warmonger, those are the people I would LEAST want in Congress, and, if I had the secret power to select some of the Democrats out, via "trade secret," proprietary programming code--in craftily designing a Congress that looked like a win for the people, but wasn't really--that's where I would start. Analysis is needed, not just of Dem vs. Repub, but which Dems lost? What was their policy on the war? Were the strong antiwar candidates the ones weeded out--or particularly progressive or populist candidates, who might tend to scrutinize war spending and support other serious reform?

Another reason that the new Congress is still not very representative of the American people is that only one third of the Senate was up for re-election. And in the Senate, we have a big blockade against the wishes of the American people--a very tight Dem majority with Joe Lieberman as the pivot. The people changed the Senate as much as they could--even with only one third of it open to change, and rigged voting machines. But it's still full of Bush "pod people," and War Democrats. That is where most progressive proposals will die--in the Senate--including, possibly, election reform with any teeth. (Diane "You too can learn to love the Corporate Rulers" Feinstein heads the Senate elections committee.)

One final thought on additional analysis that is needed: Lieberman's election. I understand that CT still has the old reliable, and virtually unriggable lever voting machines, but I picked up that CT introduced an electronic central tabulator between the primary and the election. I'm still wondering about that. The pre-election polls showed Lieberman ahead--and it may turn out that war profiteers simply bought the election (poured money into it), that Lieberman had too many chips he could cash in (while Lamont had none), and that the Democratic Party failed to support the choice of the rank and file. It all seems very corrupt. Lamont did a magnificent thing in running, and at last eloquently expressing the American people's overwhelming opposition to this war, in a political forum. He is the "Eugene McCarthy" of this era--the harbinger, the political catalyst. If he did lose it, in truth, I'm sure it was the forces that were arrayed against him, and not anything he did wrong. But it doesn't make much sense. I repeat, SEVENTY PERCENT of the American people want the war ended. EIGHTY-FOUR PERCENT want no part in a widened Mideast war. So what is it about CT that caused it to buck this tremendous trend? Or did it? It needs an explanation.

[nashville_brook]

31. this post totally kicks ass -- "it's the CONTEXT, stupid"

thank you for taking the time to write all this out. please consider making it a thread. it needs to be read b/c our position on election fraud isn't just about statistics...

for the regular joe, it's about the context: the means, motive, opportunity and history of theft.

[druidity33]

46. please make a journal...

i often find your posts insightful. I've found myself saying on DU before, "someone brought up that ____ ____" only to be told downthread it was Peace Pat or H2O man who saw through the gauze...

thanks for your work.

[In Truth We Trust]

5. K&R Hand Counted Paper Ballots NOW!

[mom cat]

6. Clickable versions of the links in the OP and supporting charts:

http://www.cnn.com/ELECTION/2006/pages/results/states/US/H/00/epolls.0.html

GENDER
...........Mix Dem Rep Other
Male. 49% 50% 47% 3%
Female 51% 55% 43% 2%
Total. 100% 52.6% 45.0% 2.4%

____________________________

NYT National Exit Poll


GENDER
...........Mix Dem Rep Other
Male. 48% 51% 47% 2%
Female 52% 55% 43% 2%
Total. 100% 53.1% 44.9% 2%

____________________________

Wikipedia
---------
Dem 39.267mm 57.7%
Rep 28.464mm 41.8%
http://en.wikipedia.org/wiki/United_States_House_election,_2006
____________________________

CBS 11/09
---------
(reported by the Election Defense Alliance-EDA)

http://electiondefensealliance.org/landslide_denied_exit_polls_vs_vote_count_2006



http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=364x2775205

[Judi Lynn]

10. First time seeing the NYT National Exit Poll you posted. Thank you. n/t

[mom cat]

16. I ad not seen it either.

[fooj]

7. KandR

:patriot:

[Febble]

8. Not me. So? n/t

[FogerRox]

14. KnR & Bookmarked

[No DUplicitous DUpe]

18. To say nothing of the thousands who would have voted, but for the..

long lines, caused by voting machines, that cannot be scaled up to demand.

Tale of tally: Voters gave up
20,000 fewer people voted in Denver than officials had expected.
http://www.denverpost.com/news/ci_4702680

SNIP:
Denver finally finished counting its votes Tuesday, confirming that about 20,000 fewer people voted in Denver than officials had expected.

The final count came exactly two weeks after an Election Day so troubled that officials are considering an overhaul.

An analysis of past voting trends indicates a sizable number of voters, confronted with lines that took up to three hours to get through, likely gave up and went home.

"We're not going to pretend that people didn't leave without voting, because they did," said Alton Dillard II, the Denver Election Commission's communications director.

There's no way to know exactly how many voters were affected, but when compared with past gubernatorial races, it's clear the Denver vote came up short.
http://www.denverpost.com/news/ci_4702680

[garthranzz]

19. Not me. Nor in 04 or 02 or 00 - but, hey, we only know the truth here

And this is one reason why retaking Congress just isn't enough.

[mom cat]

21. You are so right. Now we have to stay moblized.

[texpatriot2004]

23. K & R for Transparent Democracy nm

[Bernardo de La Paz]

24. Fundamental Premise is Bogus. A generic poll WILL NEVER MATCH actual candidates.

There surely was fraud and skulduggery in some places. Find it. But the generic poll is not the way.

[OnTheOtherHand]

25. I think this is sort of a content-free thread n/t

[mom cat]

47. TIA responds: So, this is a "content-free" thread?

TIA responds: So, this is a "content-free" thread?

I was just over at Pollster.com perusing a series of threads regarding comparing Generics and individual congressional polls. The analysis and prose were very impressive, but something very fundamental was missing. I kept looking for any mention of the word: F-R-A-U-D. Why do you and your fellow analysts at pollster.com NEVER once even consider F-R-A-U-D as an explanation for the vote/poll discrepancies?

Not to mention the failure to consider the discussing a few of the recorded instances of machine/software problems ("hacking" and "glitches") which muck-up the vote counts. Such as the missing 18,000 votes in FL-13 heavily Democratic precincts.

"Hear no evil, see no evil" seems to be a common theme, not only at pollster.com, but also of Charlie Cook, Chris Bowers, Ruy Texeira and many other polling "experts". In fact, some have even DISMISSED the notion of fraud. Hell, just take a look at all the DU posts in this forum for the specific horror stories.

And you call THIS a content-free thread?

SO, OTOH, PLEASE ANSWER THIS ONE QUESTION DIRECTLY:

HOW COME THE FRAUD FACTOR IN ELECTION 2006 WAS NEVER DISCUSSED?
IS IT FOR THE SAME REASON IT WAS NEVER DISCUSSED IN 2004? OR 2002? OR 2000?

[OnTheOtherHand]

49. TIA, you are full of shit on this

You know damn well that the fraud factor has been discussed in 2006, and in 2004. (EDIT TO ADD: and also in 2002 and 2000, although you may not know that.)

[mom cat]

68. TIA: Touchy, touchy...

TIA: Touchy, touchy...

Have I struck a nerve? Are you feeling some heat?

How else to describe your choice of words?

You know, I never said you were full of shit. But I do admit it's implied each and every time I destroy your convoluted, misleading arguments in full view. Like I have in this thread.

Instead of your that one-line response, you could have pointed out exactly where the subject of F-R-A-U-D was discussed in the specific pollster.com thread I referred to. In fact, it would be nice if you showed us a few examples where the subject was brought up by someone other than myself. Did YOU ever bring it up? You may want to check out mydd.com. I never saw F-R-A-U-D mentioned there, either. Check out Charlie Cook's website while your at it.

The bottom line is that professional pollsters and analysts don't want to discuss F-R-A-U-D at all. After all, they are creatures of the media - the black lagoon which you should have referred to. They are very concerned about keeping their status and income intact, so they avoid the issue. I have not made one dime directly or indirectly from any of my postings. Can you make that statement?

It seems obvious that at this point, with all the instances of "computer glitches" in 2006 and prior elections, that any discussion of polling discrepancies from the reported vote count would discuss the F-R-A-U-D factor.

Now show us where it was discussed here:
http://www.pollster.com/blogs/

[OnTheOtherHand]

71. TIA, you just disappoint me all the time

Once I thought you were at least trying to engage the substance of criticisms, but now I see no sign of it. I regret it.

I'm still waiting for your explanation of why it "doesn't make a difference WHAT Gore (or Bush) voters SAID in 2004 about who they voted for in 2000," and yet you keep brandishing a table about what voters said in 2004 about who they voted for in 2000. It's not that you can convince me that these stances are consistent, but I can't help being curious whether you have actually convinced yourself.

I have no idea whether fraud was discussed in that particular pollster.com thread, and I couldn't care less. What you wrote was that I "NEVER once even consider F-R-A-U-D as an explanation for the vote/poll discrepancies," and that is wrong. In fact, it's ridiculous, since I've had substantive posts on several threads about the topic, including yours.

What actually bothers you isn't that I don't consider fraud, it's that I have the temerity to disagree with you. Maybe you can't even believe that anyone actually disagrees with you, and so you have to tell yourself that we just don't consider your arguments, no matter how long we spend responding to them.

However, if it is important to you to see me discussing fraud as an explanation for vote/poll discrepancies on pollster.com, try here.

"The bottom line is that professional pollsters and analysts don't want to discuss F-R-A-U-D at all.... They are very concerned about keeping their status and income intact, so they avoid the issue."

See, you could have been discussing it, but here you are evading the substance and smearing entire professions instead. Your choice.

"I have not made one dime directly or indirectly from any of my postings. Can you make that statement?"

Absolutely. I don't make a dime directly or indirectly from posting on DU. But after you finish trying to smear me, maybe you would like to start on Skinner?

[Name removed]

75. Deleted message

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

[OnTheOtherHand]

78. I'm afraid of spending the rest of my life arguing with a rock

An abusive rock, to be sure.

"THE 43/37 WEIGHTS WERE NOT ONLY IMPOSSIBLE, THEY ARE PROOF THAT E-M HAD TO FUDGE THE IN ORDER TO MATCH THE FINAL NEP TO A CORRUPT VOTE COUNT."

Care to buy a noun?

Never mind, it won't help. The 43/37 figures appear in a table about whom people report having voted for in 2000. Once you say that it doesn't matter who they say they voted for in 2000, your argument is gone. End of story. To repeat myself, the fact that the table is 'impossible' because 43% of 2004 voters can't have voted for Bush in 2000 no more proves fraud than would a table -- weighted or unweighted -- that showed that 2% of 2004 voters admitted to having run red lights. You are confusing polls and real life. This is not an especially subtle distinction.

I've pointed to the fact that exit polls have overstated the previous winner's margin in every ICPSR-archived presidential exit poll going back to 1976, but you refer to a "FICTITIOUS RATIONALE." What is your problem?

In the "Game" thread, I used defection percentages actually observed in the 2000-2004 National Election Study panel survey. Amazingly, in the face of direct evidence of people changing their reports of who they voted for four years earlier -- and the effect it had on the apparent defection percentages -- you seem to think I dreamed the whole thing up. Did you ever look at that panel study?

I have no way of knowing whether I misinterpreted you before, but this new statement seems unambiguous: "the Mystery Pollster and others of his ilk have consistently avoided the F-R-A-U-D issue." That, of course, is again ridiculous. You can't have missed his four-part critique of the RFK Jr. article.

Trying to turn you against Skinner? Have you read his posts? He disagrees with you, plain and simple, as I do. It's nice that you apparently don't dare to say the things about him that you say about me, but I don't find that it reflects especially well on your character.

EDIT: omitted word

[Febble]

79. TIA doesn't appear

to understand what weights are.

He also appears to be posting abuse on a forum from which he was banned.

[BeFree]

27. Premise is allowable

Generic polls have shown to have tracked voters' intent many times without serious error. Why do you think they expend the time and effort to do the poll? If generic polls had been proven to be no good, the poll wouldn't happen.

Sorry, but you sound like others here who seem to say that polling people are a bunch of shits and we should pay no attention to their product.

Generic polls are a useful tool to discover voter intent, and what they show here is another bunch of stolen votes.

[Bernardo de La Paz]

37. No, don't confuse polling people with generic polls.

A generic poll is one that asks "do you favor the Democratics or the Republicans?" However, people in the voting booth vote for people, specifically, since they are given names (and affiliations) to vote for. That is why the incumbent wins so often, even when generic polls favor the other party.

Generic polls have been shown to NOT track voters' intent many times. There are many reasons why pollsters ask the generic question, and a common reason is to compare with prior elections, not for comparison within an election.

[BeFree]

43. So, you are the expert?

You seem to know all there is to know about these polls, but you are just being shy with the inside scoop, eh? Otherwise you'd post up something at least as half as comprehensive as TIA's body of work.

So here we sit, at your mercy, waiting for you to show us that the polls are shit and should just be trashed, of no use, just wasted money. But, I'd wager, all you do is come in trash the place and leave. See ya.

[Bernardo de La Paz]

53. Please drop the personal attacks. I'm not a know-it-all. Don't shoot the piano player.

Please be civil.

[OnTheOtherHand]

44. you're making stuff up again

"Generic polls have shown to have tracked voters' intent many times without serious error."

Prove it.

"...you sound like others here who seem to say that polling people are a bunch of shits and we should pay no attention to their product."

Uh, no. There are people here who "seem to say" that polling people are all right, but we shouldn't take the polls literally. And there are people here who "seem to say" that polling people are a bunch of shits, but their polls are pure gold. Go figure.

[BeFree]

52. Just because

Its over your head doesn't mean I am making it up.

Are you one of those who thinks that the polling people make up their numbers? That they have no basis for their computations? I would suggest you take it up with them, as it seems you have a serious problem with their work. That's all they need: OTOH looking over their shoulders. Go for it, son!

[Bernardo de La Paz]

55. Please stop the personal attacks on the user "OnTheOtherHand" and get back to debating.

[OnTheOtherHand]

57. if you aren't making it up, then present it

"Generic polls have shown to have tracked voters' intent many times without serious error."

Link, please?

I'm not randomly hurling polemic here. I think your fingers wrote a check that reality can't cash.

[BeFree]

58. Too many too list

But the general consensus is that polling has been quite accurate over the years and pollsters are continually refining their procedures to make their reports as accurate as possible.

Actually, I can't think of many polls that were inaccurate before e-voting came on the scene, can you?

[OnTheOtherHand]

59. in other words, you've got nothing n/t

[BeFree]

61. And you have, what?

I thought you studied polls? I thought you were an expert? I'm no expert, but I have read an awful lot and I stand by my memory of polling numbers being accurate, especially before e-voting came on the scene.

[OnTheOtherHand]

62. you are trying to change the subject, twice

First, you are trying to change the subject from generic House polls to polls in general.

Second, you are trying to change the subject from facts to, well, me.

"Generic polls have shown to have tracked voters' intent many times without serious error."

Prove it.

[BeFree]

63. How?

What would satisfy you? You seem unsatisfiable and you are nuts if you think I am gonna lay out proof for you too just say 'not good enough'. So, lay out what constitutes proof to you and I will deliver.

On edit: This little jewel about forecasting:
As explained previously, to arrive at her own prediction ***** performs simple computations on the forecasts of others, namely averaging. Our ***** first computes the average within the several methods by which forecasts are made, and then the average of these averages, i.e., the average across methods. This procedure has been shown to increase forecasting accuracy.

See, that's what TIA has done.

Now then, as for proof, while waiting on your parameters, I present the Generic polls used in this TIA study..... were the generic polls accurate, or were they not?

[OnTheOtherHand]

64. ah, heck, let me help

You could've linked to this:
http://www.pollster.com/guest_pollsters_corner/bafumi_erikson_wlezien_forecas.php

Only problem is, you need to look to see which generic polls they construe to have been accurate. Do some reading, check out the 2006 results again, and get back to me.

[BeFree]

65. Thank you

See, now was that so hard to do? Why stoop to personally attacking me when you had info at your fingertips? Must be that *sick fun* you go on about.

I'll look at your link later and get back to you... gotta run right now.

[OnTheOtherHand]

73. what, are you kidding?

You are relying on me to find supporting evidence for your claims, when I can't even be sure what your claims mean, and don't know whether you know either?

I said that you were making things up, and apparently you were. If that's "personally attacking," well, so be it.

[BeFree]

74. Hello, thanks for the hand

Cool. You ain't gonna like this one bit, Mr. Hand, but your link seems to support TIA's position; that generic polls have been good predictors of elections.

To begin with, we estimate a regression equation predicting the House vote in the 15 most recent midterm elections, 1946-2002, from the average generic poll result during the last 30 days of each campaign. The generic polls turn out to be very good predictors, as we have shown. Based on the current average of the generic polls (57.7% Democratic, 42.3% Republican) the forecast from this equation is a 55% to 45% Democratic advantage in the popular vote (1).

http://www.pollster.com/guest_pollsters_corner/bafumi_erikson_wlezien_forecas.php

Then at a link there we go to:

http://www.temple.edu/ipa/workingPapers/

And there a pdf "generic" link leads us to:


Forecasting House Seats from Generic Congressional Polls
(October 24, 2006)
Joseph Bafumi
Dartmouth College ([e-mail])
Robert S. Erikson
Columbia University ([e-mail])
Christopher Wlezien
Temple University ([e-mail])
According to the frequent polling on the generic ballot for Congress, the Democrats hold
a large advantage leading up to the vote on November 7. But does this Democratic edge mean
that the Democrats will win a majority of House seats? Doubts are often expressed about the
accuracy of the generic ballot polls. And even if the polls are correct in indicating a majority of
votes going to Democratic candidates, further doubts are expressed about whether the
Democrats’ vote margin will be sufficient to win the most seats.1
This paper is intended to provide some guidance for translating the results of generic
congressional polls into the election outcome.2 Via computer simulation based on statistical
analysis of historical data, we show how generic vote polls can be used to forecast the election
outcome.


And at another link we find this website:
http://politicalarithmetik.blogspot.com/2006/08/votes-seats-and-generic-ballot.html

Where it is stated:

The first bit of "good news" for Dems is that despite these problems, it is still the case that as the generic ballot goes up for Dems, their national vote share has also generally gone up. So if we throw out all the econometrics and just ask "Does higher generic ballot support usually mean higher vote share on election day?" then the answer is yes.

Still, the good news for Dems is that high generic ballots do tend to go with more seats. Bad news for Reps.

And this is the good news for Reps and the very bad for Dems. IF the votes-seats relationship in 2006 follows the pattern of the past six elections, then EVEN THE EXTRAORDINARY SUCCESS currently forecast by the generic ballot may not be enough to give Dems control of the House. In the figure above, the red line for the 1994-2004 votes-seats relationship remains below the magic 218 seats even when Dems win the 53% of the national vote which current generic ballot results would predict. Rather than the 251 seats they would be able to expect under the 1946-1992 relationship, they will expect only 215 seats, 3 short. The Dems would need 55% of the vote to reach a predicted number of seats of 218, and even the current excellent generic ballot results are not enough to sustain that number of national votes.

I find that extraordinary, don't you? Since 1994, when I first suspected the vote machines had been 'fixed' the generic polls took a dive off their historic 1946-1992 predictions. Makes one wonder, doesn't it?

This means that I would be very reluctant to assume that the historic relationship between votes and seats is still true. (And a statistical test confirms that the slopes differ to a statistically significant extent between 1946-1992 and 1994-2004.) And the implication of that is that even a very successful vote for Democrats need not translate into control of the House.

Comments:

My sense has been that the election-eve generic ballot works pretty well in non-presidential-years in identifying emerging trends, if not actual party control.

Then another comment:

I hate to say this but the first thing that leaps to mind is increasingly corrupt voting apparatus. Out of curiosity do you have any information on the prevalence of electronic voting versus hand counting over time?

All in all, I have to say thanks, Mr. Hand, before now I had just trusted TIA, when he said the generic polls were good, and now, having gone through the link you provided, my faith in TIA has been reaffirmed!

[OnTheOtherHand]

76. but you are cherry-picking

As you quoted yourself: "Based on the current average of the generic polls (57.7% Democratic, 42.3% Republican) the forecast from this equation is a 55% to 45% Democratic advantage in the popular vote."

So, the expected Democratic advantage in the popular vote (based on this equation) is smaller than the margin in the generic polls.

"The first bit of 'good news' for Dems is that despite these problems, it is still the case that as the generic ballot goes up for Dems, their national vote share has also generally gone up."

Did you notice what the problems were? (I never denied that "as the generic ballot goes up for Dems, their national vote share has also generally gone up." That doesn't mean that they match.) Here, let me help some:

In the figure, the gray diagonal line is where the points should be if the generic ballot exactly predicted the national two party vote. What we actually see is that in all but 3 of the last 30 elections, the polls fall well below this gray diagonal line-- they overstate the Democratic vote. (emphasis added)

This appears to be consistent with the other paper, which -- as you quoted but apparently didn't notice -- also indicates that the generic gap overstates the observed gap in the popular vote.

[BeFree]

81. Ya done good,

Now what? Are you saying the generic polls are not perfect? Who said the polls were perfect?

Can you argue this point? Generic polls have been good predictors of elections. That's what everyone else is saying, and who I am to argue with them?

Or this: Generic vote polls can be used to forecast the election
outcome. Seems they have more faith in their work than you do.

Or this: "Does higher generic ballot support usually mean higher vote share on election day?" then the answer is yes."

Ya done good helping us to get past the stuff that says the generic polls are no good at predictions, thanks, Mr. Hand.

[OnTheOtherHand]

83. Read. Think. Learn.

Both these sources indicate that the popular vote margin should be smaller than the generic poll margin. The popular vote margin was smaller than the generic poll margin. Not some weird new e-voting anomaly; totally run-of-the-mill.

It's hardly helpful to argue that the generic polls are actually pretty good at predictions if you ignore the predictions.

Here is part of what Bafumi et al. stated in their earlier paper:

From the literature (Erikson and Sigelman, 1995; Moore and Saad, 1997), it is known that the answer to the question "how accurate are the generic polls?" must be nuanced. We know that they perform poorly as point estimates. For instance, an 18-point Democrat lead from early in an election year most likely will translate into a far smaller vote lead on Election Day. However, regression equations accounting for the vote in terms of the generic vote do predict well, as they properly discount the exaggerated sizes of the generic poll leads. (emphasis added)

http://www.temple.edu/ipa/workingPapers/Documents/Bafumi,%20Erikson%20and%20Wlezien,%20Midterm%20Elections.pdf

This is almost the opposite of TIA's position, which just might be why you didn't quote it.

But the problem goes even further than that. Not all generic polls are conducted in all years, and in 2006 we can see that the late generic results are mutually inconsistent. Ah, but don't all such differences average out? well, sort of, on average. Those papers report that the equation based on generic polls in the last 30 days has a "Root MSE" of 1.90. The root MSE, as Google will tell you, is an estimate of the standard deviation of the error term in the model. The conventional "margin of error" will be about twice that, or 3.8 points. But notice that the measure is Democratic vote share. So if you are interested in the margin, the margin of error will be over 7 points.

That counts as pretty good prediction compared to picking a result out of a hat -- nice healthy R^2. But it won't support an inference of fraud in 2006.

Did it ever cross your mind, or TIA's, to wonder why in fact these scholars haven't written a follow-up paper claiming evidence of fraud? It could be that you are smarter than they are, or braver. Or maybe, just maybe, it could be because you've misunderstood what they were saying -- which, I have to admit, is pretty much what I expected, although I didn't expect you to go so far over the top.

[BeFree]

84. Actually what was said

was that as the Dem share increased more Dems seats were won in the house. And that that showed in the majority of elections excepting the elections from 1994-2004 when the trend of thirty+ years was skewed, but it still held that as Dem share of the generic poll increased so did the number of Dem house wins.

So what happened? No one even mentions fraud except for the one comment. And there is your division, and why the case has to be made for fraud in this election. I guess you just can't see that the fraud effected the vote. I can. Maybe I am smarter?

[OnTheOtherHand]

86. the number of Dem House wins DID increase

There could be a case for fraud in this election, but that doesn't seem to be the one.

But maybe you are smarter, so by all means, keep trying.

[mom cat]

69. TIA: ONCE AGAIN, OTOH, YOU COME UP SHORT

TIA: ONCE AGAIN, OTOH, YOU COME UP SHORT

Don't you spend time at Pollster.com?
Then how did you miss this?

http://64.233.187.104/search?q=cache:zjC6OVUBA2kJ:www.pollster.com/guest_pollsters_corner/bafumi_erikson_wlezien_forecas.php+2004+generic+polls&hl=en&gl=us&ct=clnk&cd=3

THE 3 PROFESSORS INDEPENDENTLY CONFIRMED MY GENERIC PROJECTIONS.
THEY USED A SLIGHTLY DIFFERENT APPROACH: A REGRESSION ANALYSIS OF THE 15 MOST RECENT MID-TERMS.

READ IT AND WEEP.

"To begin with, we estimate a regression equation predicting the House vote in the 15 most recent midterm elections, 1946-2002, from the average generic poll result during the last 30 days of each campaign. The generic polls turn out to be very good predictors, as we have shown".

October 27, 2006
Bafumi, Erikson & Wlezien: Forecasting House Seats from Generic Congressional Polls

(Editor's note: Today's Guest Pollster's Corner contribution comes from Professors Joseph Bafumi of Dartmouth College, Robert S. Erikson of Columbia University and Christopher Wlezien of Temple University. The post is based on a larger paper available for download here).

Although the Democrats hold a large advantage in generic ballot polls, there has been considerable uncertainty regarding whether the Democrats would win the most House Seats. Doubts are often expressed about the accuracy of the generic ballot polls. How district lines are drawn raises further doubts about whether the Democrats could win a sufficient majority of the vote to win the most seats. We estimate how the generic ballot "vote" translates into the actual national vote for Congress and ultimately into the partisan division of seats in the House of Representatives. Based on current generic ballot polls, we forecast an expected Democratic gain of 32 seats with Democratic control (a gain of 15 seats or more) a near certainty.

To begin with, we estimate a regression equation predicting the House vote in the 15 most recent midterm elections, 1946-2002, from the average generic poll result during the last 30 days of each campaign. The generic polls turn out to be very good predictors, as we have shown. Based on the current average of the generic polls (57.7% Democratic, 42.3% Republican) the forecast from this equation is a 55% to 45% Democratic advantage in the popular vote (1).

But would this mean that the Democrats also win the most seats? The Democrats winning 55% of the vote would represent a 6.4 percentage point swing from 2004, when they received 48.6%. If Democrats were to win exactly 6.4% more of the 2006 vote in every district than they won in 2004, they would win 228 seats. However, an average swing of 6.4% percentage points will be spread unevenly-sometimes more than 6.4% and sometimes less. Moreover, the prediction that the average vote swing will be 6.4% is itself subject to error.

more...

[OnTheOtherHand]

72. actually, I linked to it before you did -- and I read it, and thought about it

Based on the current average of the generic polls (57.7% Democratic, 42.3% Republican) the forecast from this equation is a 55% to 45% Democratic advantage in the popular vote (1).

In other words, based on a 15.4% gap in the generic polls, they forecast a 10% advantage in the popular vote. Is the clue phone ringing yet?

Does this mean that based on an 11-point gap in the generic polls, they would forecast an 11% advantage in the popular vote? Probably not. :eyes:

[mom cat]

82. TIA: WELL, I GUESS THAT PROVES THE GENERICS ARE USEFUL.

TIA: WELL, I GUESS THAT PROVES THE GENERICS ARE USEFUL.

ONCE AGAIN, YOU AND FEBBLE MISS THE MAIN POINT. YOU BOTH CONTINUE THE SPIN THAT USING PRE-ELECTION AND EXIT POLLS IS A USELESS ENDEAVOR IN FORECAS AND ANALYSINF=G THE RESULTS, RESPECTIVELY.

The THREE professors employed a multiple regression of historic mid-term polling data. Their analysis proved that Generic polls ARE useful in projecting the national vote. The fact that their projections differed from mine by 1-2% is NOT THE ISSUE HERE. THE POINT IS THAT OUR TWO DISTINCT MODELS BOTH PROJECTED THAT THE DEMOCRATS WOULD WIN BIG THIS TIME.

AND DON'T FORGET: MY MODEL ASSUMED ZERO FRAUD. THEIR MODEL USED HISTORICAL VOTE COUNTS WHICH UNDERSTATED THE TRUE DEMOCRATIC VOTE DUE TO BALLOT SPOILAGE AND LOST VOTES. THERE IS YOUR 2% DISCREPANCY.

YOU CAN VIEW THE GENERIC MODEL FORECAST HERE.
http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=460141&mesg_id=460141

In addition to the Generic model, I also developed a Monte Carlo Risk Analysis Model of 61 GOP-held seats. The model projected that the Democrats would win 40 seats (242D-193D) but that 15 were likely to be stolen. The key fraud assumptions were 3% ballot spoilage and 4% vote-switching. It turns out the model was very accurate. It's currently 230D-202R.

As far as I know, not ONE political pundit did a similar risk analysis.
OTOH and Febble, do you know of any who did?

YOU CAN VIEW THE 61 GOP-SEAT SIMULATION MODEL HERE.
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=364x2523087#2524208

[OnTheOtherHand]

85. the straw man cometh

(edit to clarify subject header)

For someone who likes to be called TruthIsAll, you sure have issues. Neither I nor Febble has ever claimed that pre-election or exit polls are "useless." You ought to defend (or abandon) your own lousy arguments, not invent lousy counterarguments to attribute to your critics.

Or do you really not understand the difference between saying that polls are "useless" and that they don't provide evidence of fraud in this instance? It's hard for me to imagine how you could not.

As far as I know, you haven't even calculated what Bafumi et al.'s projected national vote would have been, since the paper I cited (and, later, you cited) was prepared two weeks before the election. Nor did you look at their root mean square error. The generic average probably wouldn't change much, maybe +14 points instead of +15 points; so their projection might have been somewhere around Democrats +9%, with a margin of error of over 7 points on that margin. Big whoop. Even your most aggressive cherry-picking tactics aren't likely to salvage much of an argument there.

You also seem to have skipped their projections of Democratic seats won (228 assuming uniform swing, 235 using a more elaborate simulation method). Those hardly support your suggestion of 15 stolen seats, although given the error in the model, almost anything is possible.

Nor, apparently, have you looked at the comparisons between polls and vote counts in individual House races, conveniently summarized here -- almost no difference on average.

I'm sort of tired of pointing to all the things you ignore, since you typically continue to ignore them.

Tell you what, though: if Bafumi et al. come out with a paper that agrees with your interpretation of their work, that will certainly change my perspective on the whole thing.

[Febble]

89. Well it seems to me that

it is TIA who has missed the point. Neither I, nor OTOH have said that pre-election and exit polls are useless for forecasting results. We are both social scientists, and regard survey data as invaluable for predicting lots of things, including election results. However, as social scientists, we are aware that there won't be anything like a 1:1 relationship between polls and results. Other factors will affect that relationship.

One of those factors could be fraud. But there are many others - and the fact of their existence can be ascertained very easily from the between-poll variance in the 10 pre-election polls TIA cites.

So it is not at all surprising that a multiple regression model should predict the polling data historically. Indeed the exit poll data (unadjusted) were an excellent predictor of the official result, as TIA once showed in one of his own graphs. At precinct level, the correlation is about .9

So of course, polls are a good predictor of the count, whether or not there is fraud, and whether or not there is error in the poll (sampling or non-sampling).

What I HAVE said is that simply inferring fraud from a discrepancy between a poll and the official count is difficult because of the other factors that can (and indeed have been shown to) contribute to that discrepancy. With the generic polls, these factors have been studied, and the historical finding that the count tends to be lower than the generic poll. This does not make the generic poll non-predictive - it just means that in the regression equation that predicts the value of the counted two-party vote-share:

counted two-party vote-share= b*generic poll-share + error

b is less then 1.

[Febble]

77. Read it and what?

Here's the regression plot:



http://politicalarithmetik.blogspot.com/2006/08/votes-seats-and-generic-ballot.html

On the horizontal axis is the Generic Ballot Dem two party vote-share(the "predictor")

On the vertical axis is the National Dem two party vote-share.

There is a positive correlation between the two. Therefore the Generic Ballot is a "predictor" of the National vote.

To "predict" the National vote, you'd find the Generic Ballot vote-share on the horizontal axis, and read off the corresponding National vote-share value on the vertical axis.

However, you will notice that scale on the two axes is the same, and a diagonal line is drawn for reference. If the National Dem voteshare tended to be the same as the Generic Ballot Dem voteshare, the datapoints would congregate around that diagonal line. For Generic Ballot voteshare of 60%, you would read off a 60%.

However, you can see that the regression line is actually less steep than the diagonal. If you find a Generic Ballot voteshare on the horizontal axis, and read off the prediction of the National voteshare on the vertical axis, you find it tends to be lower. A Generic Ballot voteshare of 60% predicts a National voteshare of about 54%.

Yes, the Generic ballot would seem to be a good predictor of the National voteshare. But to get the prediction you have to knock a bit off.

[NobleCynic]

26. As this flawed analysis by TIA keeps getting kicked....

For the last time it is statistically inaccurate to compare a generic poll to the actual vote. The generic poll doesn't account for name recognition, among other things. That alone introduces enough potential error that one cannot consider any results found proof. Circumstantial evidence maybe. Proof? Most certainly not.

This is not to say TIA's conclusions regarding the election being affected by fraud are wrong. However the implications about the scale of the fraud he draws cannot be considered the gospel truth.

Bad statistics are bad statistics people. It doesn't matter if the conclusions drawn are what we want to hear. Honest debate cannot be made around dishonest numbers. Those who fight the status quo must hold their own to a higher standard.

[BeFree]

29. BS

Where does TIA say it is the gospel truth?

There are margins of error, by gawd, and estimates of the UVA, to name just two things. What TIA has done is used statistical analysis to uncover what we all know by now to be a bad case of fraud just under the surface.

Now, you can honestly show why you disagree with TIA but it seems you are just barking up a big ol'tree. Here, I'll give you something to bark at:

If polls were no good, why do they do polls?
Are you saying the polls have dishonest numbers?
Are you claiming that TIA's analysis is mathematically mistaken?

[foo_bar]

36. did anyone actually read the OP?

It's TIA's confession that his beliefs about exit polls and margins of error were a sham:

National Exit Poll
CNN-7:07pm . 76.58 41.05 34.40 1.13. 6.65 . 53.6% 44.9%. 1.5%
CNN-Final... 76.58 40.24 34.43 1.91 . 5.81 . 52.6% 45.0%. 2.5%
NYT (adj)... 76.58 40.65 34.40 1.53 . 6.25 . 53.1% 44.9%. 2.0%
(I adjusted the NYT National Exit Poll results slightly to include Other).

If we were to believe anything TIA wrote in the last two years, the exit polls are scientific "fact", the actual vote count is the "Diebold vote count", and the odds of an exit poll being off by 7-9 percentage points are billions and billions to one. But instead:

The bottom line is that the Democrats won by a much bigger margin than indicated the reported vote count indicates. If the Wikipedia reorted vote is correct, then the Democrats received 57.7% + X%, where X =3-5%.

So that's 60.7-62.7%, yet just days ago:

TRUTHISALL: 1 in 76 BILLION Odds of 5.1% Discrepancy in Dem House Vote vs. Generic Poll

The 116-poll House Generic Trend line projected a 56.4-41.6%
Democratic win, a 14.8% margin. The recorded vote was
51.3-46.4%, a 5.1% decline in Democratic vote share. The
Democratic margin declined from 10.6 million to 3.6 million
votes.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=364x2775205#2781808

So TIA's now saying the Generic Poll (which is a meaningless comparison for reasons yet to be explained away) actually went the other direction beyond its "margin of error" (as miscalculated by TIA to be 1.5% by waving house effects away with a magic wand), from 56.4 to >=60.7, and the exit polls are off by a minimum of 7 points in the opposite direction, in spite of the old TIA:

For example, if the Bush National exit poll deviation was 3.50% in going from from 47.0% in the preliminary poll to 50.5% in the recorded vote, then the odds are 1 in 288 billion for this result to have occured by chance alone.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x345177

So TIA doesn't have much literal belief in anything he writes.

[BeFree]

42. You don't quite get it, eh?

The odds are
If Wikipedia is right
If deviation was....

Besides analysis is just that: an analysis. Unlike others here who make absolutist claims, TIA rarely does, notice all the "if's and then's".

Look, like you, I don't like the idea that my vote and your vote have subjected to theft, and since you can't prove my vote or your vote was not stolen, then we have to pay close attention to how it might have been stolen.

We use the tools available to make that determination. TIA's mastery of the numbers is sure to alarm some people when he finds the numbers create odds that millions of votes were stolen, but that is no reason to attack TIA.

Either you think your vote, and mine, were counted as cast, or you don't.
Either you trust Diebold, or you don't.

[foo_bar]

56. did you understand my post, or TIA's?

Assuming a 1.5% MoE, the probability that the Democratic vote share would decline 5.1% from the Generic Poll (56.4%) to the actual vote (51.3%) is 1 in 76 billion. The probability is calculated using the Excel Normal Distribution function:

Prob = 1.310E-11 = NORMDIST(0.513,0.564,0.015/1.96,TRUE)
or 1 in 76,326,375,571

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=364x2775205

The bottom line is that the Democrats won by a much bigger margin than indicated the reported vote count indicates. If the Wikipedia reorted vote is correct, then the Democrats received 57.7% + X%, where X =3-5%.

Using all of TIA's faulty assumptions (1.5% MoE, discount of all non-sampling error and "house effects", all 116 polls having a 95% confidence interval, the different wording in the different polls, and a generic "Republican or Democratic" poll not being a predictor of named R vs. D contests), we would plug =1-NORMDIST(0.607; 0.564; (0.015/1.96); 1) into our freeware spreadsheet program (this is TIA's most conservative case in the aforelinked para, 60.7%). The odds of TIA being right, with strict deference to TIA's assumptions about his generic poll-of-polls, are .00000000962205948251693000 or 103,927,854 to 1. If we use 4% to represent TIA's "3 to 5%" fudge factor (=1-NORMDIST(0.617; 0.564; (0.015/1.96); 1)), we get .00000000000217470486063576 or 459.8 billion to one odds of TIA's assumptions today matching his assumptions last week.

It gets worse: TIA was arguing for a <1.0% MoE, "MoE = 0.97% =1.96*standard error = 1.96*SQRT((1-p)*p)/N) So I'm being conservative when I use a 1.5% MoE". Thus the mistaken spreadsheet function would be (=1-NORMDIST(0.607; 0.564; (0.0097/1.96); 1)), which has so many zeroes that OpenOffice can't find a significant digit (not quite infinity to one, but smaller than a 64-bit float can measure, so in the neighborhood of <= 1/18,446,744,073,709,551,616 depending on the mantissa and sign bit in question). So TIA created a rock even TIA can't push; either all of his assumptions from the last 2 years are bogus, and/or today's opus du jour doesn't employ the TIA Method.

Unlike others here who make absolutist claims, TIA rarely does, notice all the "if's and then's".

Understatement of the year.

Either you trust Diebold, or you don't.

Either you're with us or you're with the terrorists. In fact, it's logically consistent for both TIA and Diebold to be full of it.

[BeFree]

60. Nice work

That's something... running all those numbers, you should be proud of yourself.

You know the definition of success? Being right 51% of the time.

Seems someone like you could undertake a study as complete as TIA's and prove Diebold counted the votes correctly. If only it could be done, Diebold would pay you handsomely, eh?

All TIA has done is use statistics in an attempt to uncover the fraud we know exists. I don't understand the vitriol against him. If you don't like the fact that he is exposing fraud, ignore it.

[foo_bar]

88. that would be "no"

Seems someone like you could undertake a study as complete as TIA's and prove Diebold counted the votes correctly. If only it could be done, Diebold would pay you handsomely, eh?

I'm sure Diebold has enough in-house talent to generate another B.S. whitepaper (although there's a few fraudsters who'd make good PR flaks). I'm more of a whistleblower, having lost jobs over pesky ethical stances like not participating in insider trading, not lying to my boss's boss.

All TIA has done is use statistics in an attempt to uncover the fraud we know exists. I don't understand the vitriol against him.

It's true if you replace "use" with "abuse". Herein lies the rub:

If you don't like the fact that he is exposing fraud, ignore it.

He isn't exposing fraud, any more than he exposed the microbiologist conspiracy by using vague enough definitions to cherry-pick a desired outcome, one of an infinite number of conclusions to pick when you appoint yourself arbiter of bias and special feelings about particular data points. This variety of well-meaning junk science is precisely why we have peer review, and why none of TIA's tracts withstand it.

[mom cat]

66. TIA Responds

 TIA Responds

Kudos to you for doing all that work. Bet you thought you
finally got me.
Well, not quite.

I was being conservative when I included ABC, Pew, and
USA/Gallup, realizing that they were very suspicious late
outliers. 

The 7-poll 58.1% projection falls within the Wikipedia margin.

The 3-poll 52.3% projection lowered the 10-poll projection by
2.1%. 
Therefore, the 10-poll 56.0% projection was conservative.

Looking at the final 30 polls from Oct.2:
The 30-poll projection was 55.8%.
The 27-poll projection (excluding outliers) was 57.0%.

There was a 1.1% deviation between the 7-poll and 27-poll
projections.
Why? The Democratic trend was increasing. 

UVA: 60% of Undecided voters were allocated to the Democrats

Poll.... Date . Dem. Rep Oth. Und . UVA. . Dem
Harris.. 1023 . 47 . 33 . 2 . 18 .. 10.8 . 57.8
AP...... 1030 . 56 . 37 . 2 .. 5 ... 3.. . 59.0
CBS..... 1101 . 52 . 33 . 2 . 13 ... 7.8 . 59.8
Nwk..... 1103 . 54 . 38 . 2 .. 6 ... 3.6 . 57.6
TIME.... 1103 . 55 . 40 . 2 .. 3 ... 1.8 . 56.8
CNN..... 1106 . 58 . 38 . 2 .. 2 ... 1.2 . 59.2
FOX..... 1106 . 49 . 36 . 2 . 13 ... 7.8 . 56.8

Outliers: 
Pew..... 1104 . 47 . 43 . 2 .. 8 ... 4.8 . 51.8
ABC..... 1104 . 51 . 45 . 2 .. 2 ... 1.2 . 52.2
Gallup.. 1106 . 51 . 44 . 2 .. 3 ... 1.8 . 52.8

Averages
Polls... Dem . Rep. Other Undec UVA Projection
10..... 51.6% 39.1% 2.0%. 7.3% 4.4% 56.0% 
7...... 53.0% 36.4% 2.0%. 8.6% 5.1% 58.1% (excl. outliers)
3...... 49.7% 44.0% 2.0%. 4.3% 2.6% 52.3% (outliers)

Final 30 polls/projections from Oct.2:
30..... 51.0% 39.0% 2.0%. 8.0% 4.8% 55.8%
27..... 52.9% 38.3% 2.0%. 6.8% 4.1% 57.0% (excl. outliers)

Poll............. Date Dem . Rep
CNN LV........... 1002 53.0% 42.0%
AP-Ipsos RV...... 1004 51.0% 38.0%
Pew RV........... 1004 51.0% 41.0%
TIME LV.......... 1005 54.0% 39.0%
Newsweek RV...... 1006 51.0% 39.0%

ABC RV............1008 54.0% 41.0%
CNN LV ...........1008 58.0% 37.0%
Gallup LV.........1008 59.0% 36.0%
Harris LV.........1009 49.0% 36.0%
FOX LV............1011 50.0% 41.0%

CNN LV .......... 1015 56.0% 40.0%
NBC RV........... 1016 52.0% 37.0%
Newsweek LV... ...1021 55.0% 37.0%
Gallup LV........ 1023 54.0% 41.0%
ABC RV........... 1023 54.0% 41.0%

CNN LV .......... 1022 57.0% 40.0%
Hotline RV....... 1023 52.0% 34.0%
Zogby LV......... 1025 44.0% 33.0%
FOX LV........... 1025 49.0% 38.0%
Newsweek LV...... 1027 53.0% 39.0%

CNN LV .......... 1029 53.0% 42.0%
NBC LV........... 1030 52.0% 37.0%
CBS LV........... 1101 52.0% 33.0%
Newsweek LV...... 1103 54.0% 38.0%
TIME LV.......... 1103 55.0% 40.0%

Pew LV........... 1104 47.0% 43.0%
ABC LV........... 1104 51.0% 45.0%
Gallup LV........ 1106 51.0% 44.0%
CNN LV .......... 1106 58.0% 38.0%
FOX LV........... 1106 49.0% 36.0%

In closing, I want to emphasize that my probability
calculations are based on actual polling results, an assumed
UVA over a range of MoE assumptions. Do you have a problem
with that? And I was being conservative when I used a 1.5%
MoE. As for those "house effects", why don't you try
to quantify them?

As the number of polls increase, so-called "house
effects" are minimized. As long as the polls sample from
the same general population at approximately the same time,
averaging of the results makes mathematical and intuitive
sense. So does combining the sample-sizes. That's why analysts
combine the latest polls. 

A most important and famous result is The Central Limit
Theorem (CLT). Political analysts are aware of the Law of
Large Numbers, but are unaware of the ramifications of the
CLT. 

The CLT states that if the sum of the variables has a finite
variance, then it will be approximately normally distributed.
Since many real processes yield distributions with finite
variance, this explains the ubiquity of the normal
distribution.

You can test out the CLT yourself with this simulation:
http://en.wikipedia.org/wiki/Illustration_of_the_central_limit_theorem

[mom cat]

67. ****** Clickable Link *****

http://en.wikipedia.org/wiki/Illustration_of_the_central_limit_theorem

[foo_bar]

87. the solution is more cherry-picking?

I was being conservative when I included ABC, Pew, and USA/Gallup, realizing that they were very suspicious late outliers.

This is satire, right? Heck, just pick your five favorite polls from the bunch, today's "very suspicious" data was last week's "1 in 76 billion". You can reach an infinite number of conclusions by shrinking your sample-of-samples on bizarre subjective grounds ("realizing they were very suspicious"), after the fact, which you've already tried:

No, I leave them out because they are BIASED for Bush.

Why include them if they skew the averages against Kerry?

These organizations <FOX, CNN/Gallup, AP> are notoriously pro Bush. They constantly prop him up and thrash the Democrat, whether it was Gore in 2000 or Kerry today.

Taking them out of the averages can only improve the forecast.

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=104&topic_id=1998204#1998844

No, I leave them out because they are BIASED for Bush.
Why include them if they skew the averages against Kerry?

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=104&topic_id=1998204#1998844

That's why I stick with Zogby and ARG, etc. I've stopped blindly incoporating the latest poll that Votemaster puts up - unless t makes sense. Call me a cherry-picker. I throw away the bad ones. That's why my numbers seem so far out there for Kerry.

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=132&topic_id=1039319#1041539

After selectively accusing CNN, Gallup, AP, Pew, USAToday and ABC of bias, you proceeded to use their data when it fit your conclusion, and dropped some of it when you wanted a different conclusion. If that isn't the definition of cherry-picking, what is?

Test the hypothesis by collecting more data to see if the hypothesis continues to show the assumed pattern. If the data does not support the hypothesis, it must be changed, or rejected in favor of a better one. In collecting data, one must NOT ignore data that contradicts the hypothesis in favor of only supportive data. (That is called "cherry-picking" and is commonly used by pseudo-scientists attempting to scam people unfamiliar with the scientific method. <snipped crack at creationism> )

http://servercc.oakton.edu/~billtong/eas100/scientificmethod.htm

Which brings us to the real Central Limit Theorem:

The CLT states that if the sum of the variables has a finite variance, then it will be approximately normally distributed. Since many real processes yield distributions with finite variance, this explains the ubiquity of the normal distribution.

From your Wiki link:

They all express the fact that any sum of many independent identically distributed random variables will tend to be distributed according to a particular "attractor distribution". The most important and famous result is called The Central Limit Theorem which states that if the sum of the variables has a finite variance, then it will be approximately normally distributed.

http://en.wikipedia.org/wiki/Central_limit_theorem

Whether these approximations are sufficiently accurate depends on the purpose for which they are needed, and the rate of convergence to the normal distribution. It is typically the case that such approximations are less accurate in the tails of the distribution.

http://en.wikipedia.org/wiki/Normal_distribution#The_central_limit_theorem

The Central Limit Theorem tells us, quite generally, what happens when we have the sum of a large number of independent random variables each of which contributes a small amount to the total.

http://www.dartmouth.edu/~chance/teaching_aids/books_articles/probability_book/Chapter9.pdf

Some students having difficulty reconciling their own understanding of the central limit theorem with some of the textbooks statements. Some textbooks do not emphasize the independent, random samples of fixed-size n (say more than 30).

http://academic.cmru.ac.th/phraisin/teaching/4112105/home_ubalt_edu_ntsbarsh_Business-stat_opre504.pdf

A smattering of polls with unrelated wording and methodology aren't independent Bernoulli trials (read: "fixed-size") guided by the CLT, especially when n is a transient number depending on the day's cherry-picking algorithm, so in disregarding the "statistical fine print" you have your dilemma: billions to one probability of this OP being consistent with assumptions you presented last week.

Do you have a problem with that? And I was being conservative when I used a 1.5% MoE. As for those "house effects", why don't you try to quantify them?

I realize you were "being conservative", I got the string of zeroes by trying your less conservative assumptions. Quantifying the house effects of these 116 polls (or however many aren't "very suspicious") would be a major undertaking, but treating house effects in general as a conspiracy theory is easy to refute:

When combining polls from different survey organizations, house effects also are a problem. These effects represent the consequences of survey houses employing different methodologies, including survey design itself. Indeed, much of the observed difference across survey houses may reflect underlying differences in screening and weighting procedures. Results can differ across houses for other reasons, including data collection mode, interviewer training, procedures for coping with refusals, and the like (see Converse and Traugott, 1986; Lau, 1994; also see Crespi, 1988). Whatever the source, poll results can vary from day to day because polls reported on different days are conducted by different houses.

http://www.nuffield.ox.ac.uk/Politics/papers/2002/w27/wlezien.pdf

[Mr_Spock]

35. I agree

Don't you know you're supposed to just be awed by the sheer volume of data & type "K&R!"

[Febble]

39. You'd think that would be obvious

but it doesn't seem to be, to some people.

Anyway, nicely put.

[kster]

40. How ya doing Febble,still

hanging in there huh.........

[Febble]

41. Yup

still trying to get the good arguments heard above the bad ones.

There are plenty of good ones.

[Sancho]

45. Statistics never "prove" anything..

Statistics test hypotheses (at least the inferential stats discussed here) and provide an estimate that the null is true through probabilities.

Lawyers "prove" things.

You still have to decide if the odds are good enough for you, and take a chance that the odds were incorrect when reported with the statistics.

We had 100 years of doctors reporting that smoking caused cancer, but not enough "statistical proof" for the "scientific community". We're debating the "proof" of global warming now. For some, the waves can knock down their house and they wouldn't believe the water is rising.

The best "proof" is observation over and over so that any reasonable person doesn't care about hypothesis testing! How many thousands of reports of undervotes or ballot problems or crazy results does it take? At least the tide seems to be turning among the "average voters" that elections are hacked.

Meanwhile, polls (and particularly exit polls) COULD be useful for finding problems, confirming problems, and getting to the source of the problems, but I don't think pollsters (in some cases) are as open and honest to the issues as they might be...particularly exit pollsters. I realize this is debatable, but that's my view.

At least it's good that this post provides the descriptive evidence in the polls, without all the arguments about non-response bias and mathematical assumptions. This many polls is evidence of generalizibility and replicability regardless of the "level of significance" associated to a given result. If you believe that this evidence is biased, then ignore it.

[BeFree]

54. Exactly

If you think it is biased, just ignore it. Or come up with your own set of numbers that make your case.

The 'peanut gallery' 'experts' can only try to tear down this effort of TIA's, whilst producing nothing of any substance detailing alternate assumptions or reasons for the election outcomes.

Alas, we'll not see their product, never have. The only person who has taken a stab at explaining the election shenanigans is TIA. I guess the others believe Diebold's product?

[demo_not_full]

30. I said the same ...

In 92 I said that I bet Clinton won by a bigger margin. The pubs work hard to cheat people out of votes. It isn't hard to see what the people want in this country. Regardless of what the machines tell us. A voice speaks louder than paper and I wish more people would believe that!

[nashville_brook]

32. wow! welcome to DU!

rarely do i get the opportunity to say hello to someone with such a low post-count. welcome.

i remember being so innocent in 92 as to not question the veracity of the vote. that was a long time ago for me -- :evilgrin:

[bridgit]

33. k&r'd, mc...

;)

[Mr_Spock]

34. .

[kster]

38. Go Mom Cat, Thank you TIA

Keep the numbers coming, Stop the Rape of our Democracy!!!! PBHC........ K&R........

[Tsiyu]

48. Thanks for your work

and for keeping me in the know...


:pals:

[dchill]

80. Who still believes...

that Santa Claus is really the Tooth Fairy? In other words, lots of people. People who believe what they're told to believe.