Kerry: Exit polls (50.6%/311EV) MATCHED pre-election polls (50.52%/322EV)

[TruthIsAll]

THIS NEEDS REPEATING:
The EXIT polls AND the PRE-ELECTION polls agreed PERFECTLY.
John Kerry, you WON the election.
And the recounts will PROVE it.

Kerry%	Latest Pre_Election Polls		4pm EXIT Polls			
St	EV	Kerry%	Kerry	Bush	Diff	EV	Kerry	Bush	Diff	EV
AL	9	40.63	39	57	-18		41	59	-18	
AK	3	34.48	30	57	-27		40.5	59.5	-19	
AZ	10	47.37	45	50	-5		47	53	-6	
AR	6	50	48	48	0	6	46.6	53.4	-6.8	
CA	55	53.85	49	42	7	55	54	46	8	55
										
CO	9	49.47	47	48	-1		49.1	50.9	-1.8	
CT	7	55.32	52	42	10	7	58.5	41.5	17	7
DE	3	54.22	45	38	7	3	58.5	41.5	17	3
DC	3	87.64	78	11	67	3	91	9	82	3
FL	27	51.55	50	47	3	27	50	49	1	27
										
GA	15	44.68	42	52	-10		43	57	-14	
HI	4	50	45	45	0	4	53.3	46.7	6.6	4
ID	4	33.71	30	59	-29		33.5	66.5	-33	
IL	21	56.25	54	42	12	21	57	43	14	21
IN	11	40.21	39	58	-19		41	59	-18	
										
IA	7	53.19	50	44	6	7	50.7	49.4	1.3	7
KS	6	38.14	37	60	-23		35	65	-30	
KY	8	41.05	39	56	-17		41	59	-18	
LA	9	45.45	40	48	-8		44.5	55.5	-11	
ME	4	56.18	50	39	11	4	54.8	45.3	9.5	4
										
MD	10	55.67	54	43	11	10	57	43	14	10
MA	12	70.33	64	27	37	12	66	34	32	12
MI	17	53.61	52	45	7	17	52.5	47.5	5	17
MN	10	54.17	52	44	8	10	54.5	45.5	9	10
MS	6	45.16	42	51	-9		43.3	56.8	-13.5	
										
MO	11	47.31	44	49	-5		47.5	52.5	-5	
MT	3	38.71	36	57	-21		39.8	60.3	-20.5	
NE	5	34.41	32	61	-29		36.8	63.3	-26.5	
NV	5	50	49	49	0	5	49.4	50.7	-1.3	
NH	4	50	47	47	0	4	55.4	44.6	10.8	4
										
NJ	15	54.35	50	42	8	15	55	45	10	15
NM	5	50	49	49	0	5	51.3	48.7	2.6	5
NY	31	59.38	57	39	18	31	63	37	26	31
NC	15	48.14	46.7	50.3	-3.6		48	52	-4	
ND	3	38.89	35	55	-20		34	66	-32	
										
OH	20	51.55	50	47	3	20	52.1	47.9	4.2	20
OK	7	31.46	28	61	-33		35	65	-30	
OR	7	53.19	50	44	6	7	51.2	48.8	2.4	7
PA	21	52.63	50	45	5	21	54.4	45.7	8.7	21
RI	4	60.87	56	36	20	4	64	36	28	4
										
SC	8	43.3	42	55	-13		46	54	-8	
SD	3	44.68	42	52	-10		37.8	62.3	-24.5	
TN	11	48.73	47.8	50.3	-2.5		41.5	58.5	-17	
TX	34	38.54	37	59	-22		37	63	-26	
UT	5	25.81	24	69	-45		30.5	69.5	-39	
										
VT	3	56.99	53	40	13	3	65	35	30	3
VA	13	47.96	47	51	-4		47	51	-4	
WA	11	54.17	52	44	8	11	55	45.1	9.9	11
WV	5	48.52	45.8	48.6	-2.8		45.3	54.8	-9.5	
WI	10	53.68	51	44	7	10	52.5	47.5	5	10
WY	3	30.85	29	65	-36		29	65	-36	
										
Total	538	50.52	47.88	46.89	0.99	322	50.6	49.34	1.26	311
										
										
		
										

[DireStrike]

1. I don't think the recounts will prove anything.

They've been given too much time to cover their tracks.

[TruthIsAll]

23. ZOGBY = MITIFOSKY = STOLEN ELECTION n/t

.

[TruthIsAll]

32. A GRAPH OF KERRY'S PRE-ELECTION POLLS vs. EXIT POLLS vs. VOTE TALLIES

[TruthIsAll]

47. A NOTE ON MY USE OF THE SIMON/SCOOP EXIT POLL NUMBERS

My numbers came from the Simon SCOOP post. I mentioned this early on, but forgot to mention that I calculated 2-party percentages based on them for comparison purposes. My Election Model was 2-party based.

There was no bias in doing this. The percentages were slightly adjusted upward for BOTH Bush and Kerry, proportional to their actual percentages, to eliminate third parties Nader et al.

I want to emphasize: this was unintentional and I'm sorry for any confusion. I never claimed to have found these numbers anywhere but from Simon's exit poll analysis on SCOOP.

ALL analysis I have done based on these numbers is valid - there was no bias to Bush or Kerry. The deviations were NOT effected.

LET ME REPEAT: THE 2-PARTY ADJUSTMENT HAD NO VISIBLE EFFECT ON THE RELATIVE DEVIATIONS !

For example, look at Colorado:
The Simon exit poll numbers were
Bush 49.9
Kerry 48.1
DIFF = 1.8

My adjustment to convert to 2-party percentage:
KERRY adj. = Kerry /(Kerry+Bush)= =(49.9/(49.9+48.1) =49.08%
Bush Adj = = 1 - Kerry = 50.92%
DIFF = 1.84 (that's a difference in deviation of .04%)

That's not noticeable unless you calculate percentage deviations to 2 decimal places. No one does that.

Here are the actual exit poll numbers from Simon.
http://www.scoop.co.nz/mason/stories/HL0411/S00142.htm

Critical states (12)
State BUSH KERRY #Resp Time Red Shift
Colorado 49.9 48.1 2515 12:24AM 2.60%
Florida 49.8 49.7 2846 12:21 2.5
*Florida 51.4 47.6 2862 1:01 0.6
Michigan 46.5 51.5 2452 12:21 1
Minnesota 44.5 53.5 2178 12:23 3
Nebraska 62.5 36 785 12:22 4.3
Nevada 47.9 49.2 2116 12:23 2.2
New Hamp. 44.1 54.9 1849 12:24 4.9
New Mex. 47.5 50.1 1951 12:24 1.9
Ohio 47.9 52.1 1963 7:32PM 3.1
*Ohio 50.9 48.6 2020 1:41AM 0.3
Penn 45.4 54.1 1930 12:21 3.4
Wisconsin 48.8 49.2 2223 12:21 (-)0.3
Iowa 48.4 49.7 2502 12:23 2

(Nebraska included because "critical" because of ES&S dominance and history)

Non critical states(35)
State BUSH KERRY #Resp Time Red Shift
Alabama 58.1 40.5 730 12:17AM 4.2
Alaska 57.8 38.8 910 01:00AM 4
Arizona 52.8 46.7 1859 12:19 2.5
Arkansas 52.9 46.1 1402 12:22 1.1
Calif 46.6 54.6 1919 12:23 (-)1.5
CT 40.9 57.7 872 12:22 3.4
(CT 44.4 54.7 872 12:53) 0.2
DC 8.2 89.8 795 12:22 0.3
Delaware 40.7 57.3 770 12:22 4.8
Georgia 56.6 42.9 1536 12:22 2.2
Hawaii 46.7 53.3 499 12:22 (-)1.2
Idaho 65.7 32.9 559 12:22 2.6
Illinois 42.4 56.6 1392 12:23 1.6
Indiana 58.4 40.6 926 12:22 1.6
Kansas 64.5 34.1 654 12:22 (-)2.7
Kentucky 58.4 40.2 1034 12:22 0.9
Louisiana 54.7 43.9 1669 12:21 2.1
Maine 44.3 53.8 1968 12:22 0.8
Maryland 42.3 56.2 1000 12:22 0.5
Mass 32.9 65.2 889 12:22 3.7
Miss 56.5 43 798 12:22 3.3
Missouri 52 47 2158 12:21 1.5
Montana 58 37.5 640 12:22 (-)0.3
ND 64.4 32.6 649 12:22 (-)2.4
OK 65 34.6 1539 12:23 0.8
Oregon 47.9 50.3 1064 12:22 (-)1.3
RI 34.9 62.7 809 12:22 3.4
SC 53.4 45.1 1735 12:24 4.4
SD 61 36.5 1495 12:24 (-)1.8
Tenn 58 40.6 1774 12:23 (-)1.7
Texas 62.2 36.3 1671 12:22 (-)2.0
Utah 68.1 29.1 798 12:22 2.5
Vermont 33.3 63.7 685 12:22 5.2
Wash 44 54.1 2123 12:38 1.6
WV 54 44.5 1722 12:24 1.8
Wyoming 65.5 30.9 684 12:22 2.7

[berniew1]

55. Documentation that Kerry Won Ohio; recount will confirm it

Documentation that Kerry won Ohio
http://web.northnet.org/minstrel/supreme.htm


Strong Evidence of fraud in Miami County, Ohio
http://web.northnet.org/minstrel/miami.htm

High undercounts in Kerry precincts and low undercounts in Bush precincts in Cincinati
http://web.northnet.org/minstrel/cincinnati.htm

Indications of Fraud in several Ohio counties
www.flcv.vom/fraudioh.html

Indications of possible ballot box stuffing in S.W. Ohio counties
http://www.flcv.com/swohio.html
http://web.northnet.org/minstrel/warren.htm

Favoritism in the Suburbs http://web.northnet.org/minstrel/suburbs.htm

Stealing Votes in Cleveland http://web.northnet.org/minstrel/cleveland.htm

(there is more documentation) http://web.northnet.org/minstrel/alpage.htm
(this is without the huge amount of voter suppression that has been previously documented) http://web.northnet.org/minstrel/columbus.htm

[bruised]

28. FULL ORIGINAL EXIT POLLS FOUND

See www.exitpollz.org for FULL ORIGINAL EXIT-POLL for some states.
I would have put this on breaking news -but I do not have enough posts.
What I would really like would be a copy of the original National exit poll that predicted a lead to Kerry in the popular vote. SOMEBODY MUST HAVE DOWNLOADED those files that were on CNN all evening but have now vanished into thin air.
(Look for update times typically 7 pm).

If you have any other original exit polls, President or Senator, I will add them. I will be improving the site soon and adding an email for this. or you can reply here. The object is to publish the exit polls that have vanished into thin air! The fact of having source docs gives more weight to the figures. By the way the full polls also give detail about how voters voted within different regions of a state –which can help analysis.

I can only see 2 possible explanations:
1/ That 44 people out of 2505 respondents all told the truth as to who they voted for Senator, but then lied about who they voted for president. This needs a pattern of respondents all lying in the same way – otherwise the lies would tend to cancel each other out and the overall effect would be negligible.
2/ That 36,500 (1.8%) of votes were switched from Kerry to Bush in Colorado in order to help build up a popular vote mandate for Bush.

The fact that similar patterns are reported in many states makes me lean to the second theory. –if only we had a full list of documented exit polls.

[Name removed]

48. Deleted message

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

[spooked911]

60. Thanks-- those are interesting links.

Between this and the fact that the NH recount hasn't found anything, I am having doubts about the significance of the exit polls.

[newyawker99]

66. Hi bruised!!

Welcome to DU!! :toast:

[driver8]

54. I am thinking the same thing...

I have a feeling that the crooks that orchestrated this mess also had in place a way to cover what they have done.

I hope that is not the case.

[mbee]

96. They will prove something if they properly count the ballots

that were all mixed up between different princints and counted by the wrong machines.

[Nicholas D Wolfwood]

2. Does this account for voter turn out at all?

See, here's where voter suppression could've come into effect in places like Cuyahoga County, where they had insanely long lines in pro-Kerry counties, making it very difficult for a lot of people to vote. I'm not trying to attack, I'm truly trying to understand if that could make up the difference in polling and reality.

[Quakerfriend]

3. Thank you TIA

Once again you buck us up! Sadly though, I too believe they've had way too much time to cover their tracks. And, unwillingless/inability of Mitofsky to release the final exit polls is just too smelly!

[sportndandy]

13. No.

Exit polling is just that - a poll taken AFTER a voter exits the poll. It poll of how a person has voted; people on line to vote will not be asked anything, so anyone who turned away without voting will not show up in an exit poll.

[Name removed]

4. Deleted message

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

[New Earth]

8. i wish i could hide behind a pot leaf too

[weeve]

53. Right on Faye!

Gives a bad name to pot smokers! ha

[Kellis]

34. Rudeness is surely NOT appreciated at DU

If you have information to back up a claim fine.But we all expect a certain amount of civility at DU.You're rudeness is very apparent and has been noted for future reference.

[rwork]

5. DU needs to hire

TIA to fight for us.

[berner59]

9. great job!!

WHY doesn't Kerry believe this???? Where is the fight?? I'm losing patience...wake up John and ask for the recount right now!!!

[Name removed]

107. Deleted message

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

[lthuedk]

6. Wanted: Republican with a conscience to admit wrongdoing.

Now that's a tall order.

[tridim]

10. It's too bad David Brock didn't wait a few more years

to grow his conscience. He could have helped.

Maybe we need to play a bit dirty, and start blackmailing some key repukes. They're pushovers, it should be easy to get them to squeal.

[renaissanceguy]

42. Yeah

if only John McCain had some balls.


--------------------------------
Kerry REALLY won Ohio and the election. Buy Recount, anti-Bush, and liberal stuff at www.cafepress.com/liberalissues

[Name removed]

108. Deleted message

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

[gingergreen]

7. Thank you

I hope it's true. I feel like we're quickly running out of time to prove our point.

[jmatthan]

11. It is not a question of giving them time

The fact is that the American public behaved like sheeple instead of making the fuss from November 2 like the Ukrainians did.

If the American public wanted fairness they should have protested as a body on election night - surrounded the White House and other major institutions, got lawyers to impound machines, taken the Republicans to court for fraud, etc.

But the American public sat at home hoping someone else would do the job of getting them justice. With no groundswell protest Kerry had no option but to roll over!!!

Jacob Matthan
Oulu, Finland

[il_lilac]

16. Yes, we need to Get in the streets!

If on Sunday you do not show up at a state capital and protest, you have no business complaining. We're already a month late and this very well might be our last chance to get someone to notice that we are PISSED & serious about an investigation. Come on- It'll be great to stand in the cold with like minded people and JUST MAYBE we can make a difference.

[Kellis]

39. Thanks for posting this

about the 51 Capital March!I will be there-red gloves & Orange Scarf blazing!!

Americans Did Not ELect Bush!
DieBold & BlackWell Did!

WELCOME to DU il_lilac :toast:

[sabra]

12. Amen!

And the Zogby rep reaffirmed that to us yesterday. Thanks for the post.

[tinfoil_beret]

14. A quick calculation

A quick calculation shows that Bush mysteriously gained 2.237 million votes and that Kerry somehow lost 2.573 million votes over what the exit polls predicted. That's a large discrepancy.

[helderheid]

15. Alan Walden's article

http://www.orlandoweekly.com/news/Story.asp?ID=4688

<snip>Center for Research on Globalization's Michael Keefer states, "The National Election Pool's own data – as transmitted by CNN on the evening of November 2 and the morning of November 3 – suggest very strongly that the results of the exit polls were themselves fiddled late on November 2 in order to make their numbers conform with the tabulated vote tallies."

How do we know the fix was in? Keefer says the total number of respondents at 9 p.m. was well over 13,000 and at 1:36 a.m. it had risen less than 3 percent – to 13,531 total respondents. Given the small increase in respondents, this 5 percent swing to Bush is mathematically impossible. In Florida, at 8:40 p.m., exit polls showed a near dead heat but the final exit poll update at 1:01 a.m. gave Bush a 4 percent lead. This swing was mathematically impossible, because there were only 16 more respondents in the final tally than in the earlier one. </snip>

[NCvoter]

17. NC preelection poll?

where did you get 52-48? just about every poll i saw in nc leading up to the election showed bush up by much more than that.

[TruthIsAll]

18. That 52-48 is the exit poll. Look again n/t

.

[NCvoter]

21. i should have said

46.7 50.3

still pretty much the same difference in percentage.

here is one link with some nc pre election polls
http://www.electoral-vote.com/states/north-carolina.html

here is a better one:
http://www.race2004.net/states.php?state=nc

nothing shows the closeness you represent.

[BeFree]

19. Really?

I suppose you have a link to your information? The polls I saw were 48K-52B. Now all we are doing is he said/she said.

I trust TIA's numbers until they can be dis-proven by actual data. I don't believe his data can be refuted.

[chapel hill dem]

73. Here is a summation of the NC polls

that I saw just before the election. However, Zogby is not in this listing. Others may be missing, but I do not remember hearing about any outliers. The Mason-Dixon poll was referenced by the Raleigh News and Observer and they endorsed K-E.

http://www.dalythoughts.com/nc/nc.htm

[Nederland]

22. Not true

The increase in respondents is not the only variable. In addition to the increase, the sample makeup was altered to correct mistakes in the earlier samples. This is well covered territory, but people around here refuse to believe that the early exit polls were shit.

[helderheid]

24. Like in the Ukraine, right? n/t

[Nederland]

37. Take time to learn about the issue

The exit polls in Ukraine were not half baked analysis released midday with only half the voting completed. More importantly, the exit poll discrepancies in Ukraine were backed up with video tape and testimony of fraud.

[TruthIsAll]

26. No, True.

The final exit polls adjusted to match the rigged votes were shit.

Some people refuse to face the facts.

[Nederland]

36. Question

The only person that refuses to face the facts is you. Warren Mitofsky has owned up to the fact that he screwed up the early exit polls. What proof do you have that he's wrong? What proof do you have that the samples used in the 4pm polls accurately reflect the total voting population? None. I've asked you over and over to address this issue and all I get from you is tautologies and accusation of being a troll. Sorry TIA, its not enough. If you insist on using the 4pm exit polls as input into your analysis, you need to PROVE that the samples are accurate. Its not good enough to simply assume that they are correct. You need proof and you've got none.

I fully expect you to ignore this post or respond in a way that completely ignores the flawed sample issue. I'm waiting.

[Nederland]

70. Just as I expected

No answers from TIA. Typical.

[Nederland]

20. Question

Are you still using those bullshit 4pm exit polls?

I guess some people never learn.

[bruised]

25. ORIGINAL EXIT POLL DOCUMENTS FOUND

Your figures are a bit differen tpo mine
See www.exitpollz.org for FULL ORIGINAL EXIT-POLL for some states.
I would have put this on breaking news -but I do not have enough posts.
What I would really like would be a copy of the original National exit poll that predicted a lead to Kerry in the popular vote. SOMEBODY MUST HAVE DOWNLOADED those files that were on CNN all evening but have now vanished into thin air.
(Look for update times typically 7 pm).

If you have any other original exit polls, President or Senator, I will add them. I will be improving the site soon and adding an email for this. or you can reply here. The object is to publish the exit polls that have vanished into thin air! The fact of having source docs gives more weight to the figures. By the way the full polls also give detail about how voters voted within different regions of a state –which can help analysis.

I can only see 2 possible explanations:
1/ That 44 people out of 2505 respondents all told the truth as to who they voted for Senator, but then lied about who they voted for president. This needs a pattern of respondents all lying in the same way – otherwise the lies would tend to cancel each other out and the overall effect would be negligible.
2/ That 36,500 (1.8%) of votes were switched from Kerry to Bush in Colorado in order to help build up a popular vote mandate for Bush.

The fact that similar patterns are reported in many states makes me lean to the second theory. –if only we had a full list of documented exit polls.

[helderheid]

29. Zogby

I emailed this to my family on November 2 when it was still posted at http://zogby.com

http://www.udpc.org/httpwww_zogby_com.html

[bruised]

30. RE:zogby

Thanks it's not nearly as good as a FULL (8 page exit poll)
but it's something.
If you remeber any blog about peopl;e donwloading those exit polls that's waht we really need.

[SaveAmerica]

40. Colorado and Nevada match TIA's chart...

It's interesting that the pre-election polls and early exit polls of TIA's chart and Zogby's prediction of Colorado and Nevada being too close to call all match up. I don't know if that has anything to do with anything but it stands out to me.

[Name removed]

49. Deleted message

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

[Name removed]

50. Deleted message

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

[Name removed]

51. Deleted message

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

[Name removed]

52. Deleted message

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

[spooked911]

61. Read his links, though.

He raises a valid point.

This does not rule out fraud, but it just means you need to to treat the exit polls with care.

[txindy]

62. You must be kidding, right?

A freeper/fundie site and a more-than-a-bit-biased pollster's site. Those are used as evidence of what? Poor taste? Bad science? Oh, wait, the first doesn't believe in science. :eyes: The second only believes in his own exit polls. That he didn't do because no one paid him to do any. So, he's not envious or anything. Good to know. :eyes:

[spooked911]

112. sorry-- I didn't realize the bias.

are you sure they're wrong?

[dalloway]

27. kick n/t

[On the Road]

31. Here's a Bubble Chart Showing the Correlation

This is very interesting:

[TruthIsAll]

38. Quite interesting. The bigger the bubble, the higher the frequency.

The cluster around the 45 degree line cuts right through the Repuke fog.

[On the Road]

41. Higher the Frequency?

Meaning Kerry won the big states?

Here's a version showing Ohio and Florida broken out. They both matched pretty well:

[rdmccur]

33. yes, I was following the pre-election polls as well

and I could see the trend toward Kerry growing, then the exit polls just reinforced what I was expecting until the 'counts' started coming in. I was suspicious right away!

[Yupster]

43. Here's the final pre-election polls by all

of the polling companies

http://pollingreport.com/2004.htm#Pollster

Polling report lists 22 polls. Some may be duplicates.

Bush led in 15 (highest lead 6 %)
Kerry led in five (highest lead 2 %)
Two were tied

[TruthIsAll]

45. On edit: those were national polls: You need to look at the state polls.

Look at the 18 national polls of Registered voters.

It was a virtual tie at 47%

[TruthIsAll]

46. Here are the final 18 national polls

[Yupster]

57. You have Pew's 10-30 poll showing

Kerry ahead.

Polling report.com shows Bush + 3 in Pew's 10-30 poll.

Haven't checked any of the others.

[TruthIsAll]

58. Kerry by 1% " 45-46-1-(3) RV 10/27 - 10/30 Pew/PSRAI

http://www.nowchannel.com/national/

You might want to check me out on some others. If you find any inaccuracies, please let me know.

[Yupster]

64. Got it

you're using registered voters. Pew has Kerry up 1 with registered voters and down three with likely voters.

I don't know which is better.

I guess as long as you're consistent and use one or the other.

[tex-wyo-dem]

71. Registered vs. Likely

Most pollsters like using the likely voter numbers to give what they think are more reliable numbers, however, every pollster has a different model for determining who is a "likely voter." Some have better models than others; these models are basically formulated questions they ask people to determine thier likelyhood for actually going to the polls and voting.

This election may have blown the likely voter models out the window since the interest and turn-out were so high, so one might surmise that the registered voters model is more acceptable. Also, these pre-election polls do not take into account the newly registered voter numbers, which were at record levels in many states.

Pre-election polls have a wide MoE due to these models and the fact that undecideds are not calculated into the numbers. In exit polls there are no undecideds.

[Yupster]

92. In this case

we'd certainly want to use registered voters because if we use likely voters then Pew had Bush up three points which is exactly what he won by which would kind of blow the whole thesis that the polls showed Kerry ahead but Bush won. So better in this case to use registered voters.

[Yancey Ward]

74. Here is the real data on the last set of national polls.

http://www.realclearpolitics.com/bush_vs_kerry.html

[Skittles]

156. no, these are the results after all the shennanigans

nt

[Kellis]

35. Good Work TruthIsAll.

Thank you.

[RaulVB]

44. "NOTHING HAPPENED" ON ELECTION DAY...right. (n/t)

[pattyloutwo]

56. delete

[KerryOn]

59. Recount Will Come Out The Same

... if the software is fixed. If the computers were hacked, then who knows.

Nice comparison of numbers.

[newyawker99]

67. Hi KerryOn!!

Welcome to DU!! :toast:

[KerryOn]

99. Hi

I just joined a few days ago. I live in Columbus, Ohio. My wife and I put in hundreds of hours for act, moveon, the democratic party, and another local group. Like many others we are depressed and steaming mad. We were at the protest at the state house last Saturday, and will there be again on Sunday.
Blackwell WILL GO!

[Ira]

69. Fighting the good fight is worth it

I agree with you KerryOn. All of the statistical studies are very interesting but not legally compelling. The recounts and their timing are controlled by the Republicans. Voter suppression and software hacking are criminal offenses that take time to mature to indictments and convictions. I am afraid that W will get his second term. However, I believe that cognitive dissonance will ultimately descend upon the electorate that drank the Bush cool aid. As the truth about this administration and the last election is understood by the American people, the sense of betrayal by the Bush supporters will become palpable, leading many to believe that W is one of Scott Peck’s “People of the Lie.” At the very least, the Bush “mandate” and “capital” will evaporate. And depending upon the gravity of the criminal offenses and the drop in approval ratings, an impeachment is certainly within the realm of possibility. Although it is little consolation today, Bush will probably go down in history as one the most incompetent, reckless and dishonest presidents this country has ever had.

[lonestarnot]

102. Not my take after hearing Cliffie the Lawyer for the second time.

I see it like my garbage disposal that had a spoon all ground up in it on Tuesday. I worked and I worked but I finally got it out! The spoon is dead but the disposal still grinds the garbage right out of the sink. Cliffie gona grind on Monday.

[Lisabtrucking]

63. looks like bush won by 51% in a lot of places, sounds odd.

51% has been popping up a lot. Don't you think thats odd, Daschal lost by 51% too. Hmmm very odd to me.

[New Earth]

65. if Kerry would have gotten those 311 electoral votes.....

i'd have been celebrating - 311=my favorite band ever!!!

[davidgmills]

68. exit polls and sampling error and Mystery Pollster

I have challenged the mystery pollster to post his own odds as to whether the exit polls might suggest fraud. He has twice now criticised Freeman's work but has yet to produce any of his own. After Mystery pollster's first criticism, Freeman changed his odds.

Mystery Pollster now has Freeman's third exit poll data (the last data before it was mixed with actual data) and he now has the latest academic numbers for sampling poll error. He could easily do his own analysis and get the credit for it. Why hasn't he?

I told him that Freeman reduced his odds of error (fraud) from 250 million to one to 650,000 to one after revising for sampling error suggested by Mystery Pollster. It should be reduced further (due to better numbers on sampling error than what Mystery Pollster initially supplied Freeman) but, by how much is the question.

I am just a simple lawyer, not a mathematician. But inquiring minds want to know.

I am always suspicious of an expert who points to flaws in other experts' work but will not do his own analysis and publish his own work. This is a tried, but true, courtroom trick. One never wants his own expert to actually do an analysis for fear that his own expert's analysis might prove the other side's case.

I have challenged mystery pollster to do this. Others should do the same to see if he will come up with his own odds. If he won't, then maybe we should suspect a troll.

I think if Mystery Pollster will do his own odds, his odds of fraud will still be high, just far less than 650,000 to one.

[TruthIsAll]

75. Probability of deviations: 1 in 13.5 TRILLION or ZERO........

THESE PROBABILITIES HAVE NOT BEEN REFUTED; IN FACT, THEY HAVE BEEN CONFIRMED BY OTHER MATHEMATICIANS HERE AT DU.

These are the probabilities that Bush's tallies could have deviated as they did from the exit polls based on the MOE and calculated using the binomial distribution:

1) If you assume the calculated MOE, based on sample size (as for standard, non-exit polls), then the odds of Bush's vote tallies exceeding the exit poll MOE in at least 16 states is
1 out of 13.5 TRILLION.

2) If you assume a 2% MOE (more likely for exit polls), then the odds of Bush's vote tallies exceeding the 2% Exit Poll MOE in at least 23 states is ZERO.

If N = number of states exceeding the MOE, then the probability is:
Prob = 1-BINOMDIST(N-1,51, MOE/2, TRUE)

For (1) and (2) above:
P (1) = 1-BINOMDIST(15,51,.015, TRUE) ; the odds are 1/ P(1)
P (2) = 1-BINOMDIST(22,51, .01, TRUE) ; the odds are 1/ P(2)

It's as simple as that.

Try it yourself in Excel. Then show the results to the Mystery Man.

[Name removed]

79. Deleted message

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

[VTHoosierPatriot]

103. They wouldn't have to be decentralized

All ayou have to do is manufacture the machines. What's irrational is to allow democracy to be in control of unproven technology that may or may not be legitimate. Why not have a paper trail to verify the results and end this discussion?

[davidgmills]

80. sampling error

Truth Is All:

I am not a mathemetician. I am a lawyer. Mystery pollster's point is that exit polls are not accurate enough to predict fraud due to a phenomonon called sampling error (whatever that is it is different from random error).

Mystery pollster was very helpful prior to the election in explaining polls and trends etc. and was often on dailykos before he started his own website.

He is read by a lot of people. What bothers me now is that he has not done his own analysis, just done criticism.

I am not capapble mathematically to know whether he is right about this or not.

Initially, people like Tom Hartmann were saying that exit polls in Germany were accurate from one to three tenths of a percent. If so, then it would appear that exit polls are an excellent means of showing probabilities of fraud.

Mystery pollster seems to be suggesting (he has never acutally said this however) that exit polls due to this phenomonon of sampling error may be only accurate to 5% or so, and if such is the case, then they are poor indicators of fraud.

You can read his work at: http://mysterypollster.typepad.com/

Obviously, looking at earlier posts, which have since been erased by the modrerator, he has some people convinced of this.

The bottom line for all us non-techies is whether these late day exit polls (those not infused with actual results) can be used to predict fraud. I began telling people they could be used that way. Now I need to know wheteher I can safely argue this or not.

[KerryOn]

98. Sampling Error

It is true that the exit poll data in it's self could not prove fraud, however... A resonable persosn would not believe that there could be sampling errors in 43 states!

Kerry's exit polls were higher than the actual in 43 states, and Bushe's exit poll data was lower than the actual in the same 43 states. I'm not a lawyer (Nor a mathmatition as TruthIsAll has found), but you would think that there would be more of a balance, even with faulty exit poll data.

[Donailin]

111. 311 -- saw them 4 times here in MD/DC/VA

last time was in July. In the pit. They put on a great show. loud. loud is good.

[New Earth]

113. awesome!

i love meeting random people who even know who 311 is. they've been my favrite band for over 11 years now, seen them 16 or 17 times (bad memory, so many shows). I was RAISED by 311!

[Name removed]

72. Deleted message

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

[TruthIsAll]

76. Bye.

.

[sunnystarr]

77. I sure hope that you're expressing the Republican rationale

cause it's the lamest pile of bs I've heard in a long time. Not that you created it ... I've heard it used since Nov 3rd and it's totally laughable and unrealistic. It wouldn't fly anywhere in the world. According to you the Ukraine election was accurate and should have been supported by the US and the rest of the world. Is it just that you and the rest of *ush controlled freeps really believe that you can feed Americans this unsupported "spin" (ergo lies) and have us willingly lap it up like a thirsty dog on a hot August noon in the middle of a desert?

[slackmaster]

78. HERESY!

:dunce:

[Laurab]

82. Re: Paranoia or Foolishness?

You ARE joking, aren't you? It takes a HUGE stretch of the imagination to think Bush voters would not take the poll, or lie - they are PROUD and LOUD. On the other hand, it's not a stretch of the imagination to think some Kerry voters might be intimidated or afraid, as in BushWorld they may be seen as unpatriotic. By your definition, I guess millions of people are either "mentally ill", or "complete fools" - sign me up! Actually, I think anyone who believes as you do might fit those descriptions better, but I would never stoop to insulting a large group of people in that manner. Brainwashed I think would be the better word. It's happened in most dictatorships. Thankfully there are a lot of people who are not going to blindly follow the leader, as he continues our journey to hell.

[Merlot]

83. "The poll takers were incompetent"

That is so precious I don't even need to read any further.

[genieroze]

81. Judging by the replies I think you hit a nerve. BINGO

[ru49714]

84. Thought I'd chime in

Hi folks,

This is my first post here.

I've been reading DU, and Free Republic, since about a month before the elections, and am finding the political blogosphere to be a fantastic expression of American Democracy in a truly pure and raw form.

I imagine that the pubs where our founding fathers once debated were filled with conversation very much like what is occurring here. It's a great thing.

This one thread, though, caught my eye because I think the premise here is just wrong. The one site that tracked all the battleground state polls and provided an aggregate avarage of them, realclearpolitics.com, had Bush winning Ohio by a few percentage points. If you look at all the aggregate numbers for all the battleground states, they predicted the election nearly spot on.

http://www.realclearpolitics.com/bush_vs_kerry_sbys.html

Why the exit polls were different than the actuals is a debate worth having, but the premise here that the exit polls matched the pre-election polls is simply not true.

I hope you don't mind my joining the discussion!

[tex-wyo-dem]

89. Hi ru49714 and welcome to DU...

Thanks for your input and I understand your argument. I was watching the very same web site before the election since I thought it was one of the better predictors of the election (avg. of non-partisan polling results).

One thing to take into account regarding realclearpolitics.com is that all numbers are based on "likely" voter polling. Most polling companies like using this number over "registered" voter numbers since, historically, "likely" has been the better predictor. In this election, however, with record voter turn-out, a better predictor might have been closer to "registered" rather than "likely."

It's important to note while interpreting pre-election polls that:

1.) Every company, whether it be Gallup or Zogby or Fox News has a different model for identifying who is and who isn't a "likely" voter. This adds error since some models are better than others and none of them are perfect.

2.) Pre-election polls don't take into account undecideds, which for this election was hovering around 6%.

3.) These polls didn't have any way of measuring what the actual voter turnout would be or take into account newly registered voters, which was at an all time high in many places.

4.) And, lastly, note MoE in these polls: all of them between 3-5%. For instance, ARG had Kerry up by 2 in FL with an MoE of +/-4, meaning the final outcome could be anywhere between +6 to -2 for Kerry. A very wide variance.

Pre-election polls are just a barometer, and not a really good one for close races.

[newyawker99]

161. Hi ru49714!!

Welcome to DU!! :toast:

[Name removed]

167. Deleted message

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

[tex-wyo-dem]

85. The "sampling error" excuse is a crock...

See: http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x136599

[davidgmills]

88. sampling error

First, other boards have mentioned the computer blackout and I thought the latest on that issue was that this turned out to be disinformation. Am I wrong?

Second, on the issue of fourth exit poll of the day being "corrupted" by actual polls, and the third poll of the day being "pure," Mystery Pollster and others have made the claim that this is the way exit polls have always been done and that doing it this time was not some nefarious act.

Again, I do not know.

Third, what I have always found interesting is that the difference in number of respondents between the third and fourth polls have been very few, making the difference between the published third and fourth exit poll results mathematically impossible, unless "corrupted" by the actual results.

But if the actual results are supposed to be part of the last phase of the process, then you would expect "mathematically impossible results" if the actual results differed widely from exit polls.

It would seem that somewhere, perhaps Germany or some other European country, we could find out whether final exit poll data should include actual results or should not. Of course Freemans' point was that when they do, they are useless as a check for predicting fraud. Some studies did just that.

My assumption would be that if they are used elsewhere as fraud predictors, they can not be corrutpted with actual results. We need to find out where.

[davidgmills]

91. sampling error

Also, Freeman didn't think it was a crock, because after Mystery Pollster told him about it, Freeman reworked his numbers and reduced the odds of fraud from 250 million to one to 650,000 to one by using the factors for sampling error Mystery Pollster gave him.

Mystery Pollster now says, that after checking some academic articles on the subject, that the off the cuff numbers he gave Freeman (which Freeeman used to rework his first numbers) were off by a substantial amount in the direction that would reduce the odds of fraud by a much greater amount.

By the way, Freeman never uses the word fraud. It is just my way of understanding what he was trying to prove.

[davidgmills]

93. New study

Some really bad news.

Mystery Pollster is now reporting that CalTech/Mit has redone its calculations using Freeman's Unweighted numbers. The result is they conclude that the unweighted numbers also show no signs of fraud. This probably should be breaking news for anyone who wishes to report it as such.

Also state that exit polls are a poor means of discovering fraud. Their conclusion is that exit polls are not accurate enough.

[TruthIsAll]

94. Mystery pollster quoting Caltech - the guys who did that bogus analysis

using CNN's "adjusted" numbers, which they themselves later refuted?

Sure. Right. I Believe. NOT.

Believe the numbers that are out there. Believe Excel. Believe Binomial. Believe MOE. Believe the deviations. Believe the simple truth.

Because soon, it will be established FACT.
The crime was too widespread to cover up.

Systematic, documented incidents.

The disenfranchisement of blacks favored bush.
The programmed touchscreens favored Bush.
The spoiled ballors favored Bush.
The OH Secretary of State favored Bush.
The FL Secretary of State favored Bush.
Jeb Bush favored Bush.
Diebold/ Wally O'Dell favored Bush.
ES&S favored Bush.
The Repugs in Congress favored Bush when they would not mandate a paper trail for the touchscreens.
The Diebold/Saudi vote counters favored Bush.
The Exit Poll Deviations favored Bush in 41 of 51 states.

[davidgmills]

95. what I was hoping to hear

Truth is All:

Proof of a stolen election is now in the hands of mathemeticians and computer programmmers. The rest of us are helpless.

I looked at the addendum and it doesn't seem like they busted their butts coming up with any real numbers, but what do I know.

First of all, I think we need to somehow resolve the issue of whether exit polling can be a valid predictor of fraud, and if so, what are all the variables involved. Every day it seems like there is a new variable that somone hasn't taken into account.

When CalTech and MIT speak, people listen, even if it is BS.

Might as well listen to my broker tell me about a great stock pick.

I wish the academic mathemeticians would get heavily involved in this; same with the academic computer programmers. I would really like to hear some serious noise from them. If they did so, maybe people would have enough confidence to take to the streets. But when the academics are quiet, the rest of us will be quiet as well.

Keep up the good work, but get your academic friends to raise a ruckus.

[c-macdonald]

101. You want the truth...

#1 - You need to take the clustering effect into consideration when considering the MOE of the exit polls. Freeman made an attempt by putting it at introducing an extra 30% error to the results but it actually adds a 50-80% error to the results.

That means, for each Standard Error you calculate, you need to multiply them by at least 1.5 to get the true "clustered" Standard Error.

This will show that using 2% as the MOE is way off. Each state will be different, ranging anywhere from 4-7%.

#2 - You're only using a 95% confidence level. That means 1 in 20 of the exit polls will be wrong. In a nation of 51 separate "clusters" that means you're bound to get 2-3 wrong due to chance alone. Raise your confidence level to 99% or 99.5% to compensate for the innaccuracy.

#3 - Using the 4pm exit poll data is another problem. Use Freeman's numbers to get a better picture of the Final Exit Poll (uncalibrated) data. The 4pm numbers are basically meaningless because they aren't complete.

#4 - Comparing Pre-Election polls with Exit polls really doesn't make any statistical sense because both are, essentially, polls. They both are susceptible to sampling error, non-response error, and many other types of error. A meaningful analysis would compare Exit Polls vs. Actual Results.

Please take these points into consideration before you post your next analysis on DU.

[TruthIsAll]

110. Please take this into consideration next time you post..

I have compared pre-election polls to exit polls to actual votes.
And Bush comes up wanting in each case. Do your homework.

And please, don't use the spin that pre-election exit polls and exit polls are JUST polls. And subject to error. The statistical analysis takes the error into account.

Your arguments about polling error will fall of their own weight.

Why don't you spend some time to analyze the probability of 86 of 91 random touchscreen voters nationwide pressing Bush and coming up Kerry?

A Duer has caclulated this aat ONE in 180 million trillione, in case your interested. Would you call that polling error?

[c-macdonald]

114. But your analysis is WRONG

I'm not sure what you're saying about me doing my homework. I have done my homework and I've done research on clustering and statistical error. I highly doubt you have done that research because, if you did, you'd see that your MOE calculations are incorrect.

Pre-election polls are less accurate than exit polls but you really can't argue that they aren't polls and aren't subject to error. The analysis is supposed to take error into account but your analysis hasn't adequately done that.

My arguments about polling error are backed up by sound statistical theory and research. Read any statistics textbook on cluster sampling and you'll see that the MOE increases as the sample size increases. It results in a 50-80% increase in the standard error (and hence, an increase of the MOE).

As for the probability of 86 of 91 random touchscreen voters pressing Bush and coming up Kerry, that sounds like something worth investigating. Obviously, that wouldn't be polling error - it would be machine error and I would love to see some more data on that.

My point is not to suggest that this election was not stolen by the Repukes - just that you don't mislead the members of DU with your amateur statistical analysis of these exit polls. As far as I see it, blindly believing you're right when the facts show that you are clearly wrong is something you would more likely find from the right-wing noise machine.

All I ask is that you do some research on clustered sampling and rerun your analysis - the more rigorous and statistically sound your analysis will be, the more powerful and meaningful your conclusions will be. That's something I would think you would desire of your analysis - am I wrong?

[Name removed]

168. Deleted message

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

[thesquanderer]

116. TIA, re: 86 of 91 random touchscreen voters

> Why don't you spend some time to analyze the probability of 86 of 91 random
> touchscreen voters nationwide pressing Bush and coming up Kerry?

Did you mean to say pressing Kerry and coming up Bush?

> A Duer has caclulated this aat ONE in 180 million trillione, in case your interested.

I must have missed all of this... Do you have a link for that calculation, or even just to the story of the 86 out of 91 touchscreens? Thanks!

[TruthIsAll]

130. Here is a link

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x126395

[newyawker99]

162. Hi thesquanderer!!

Welcome to DU!! :toast:

[intheflow]

86. Sadly, I don't think a recount will prove Kerry won.

I think there were too many provisional ballots disqualified and then possibly "disappeared" by now. Somehow documentation will be lacking and not all votes that were cast will be able to be counted. Hence, the chimp is in.

I think the Dems know this. That's why they're not leading the pack in the recount thing, that's why they're saying, "We're not looking to overturn the election, we just want to make sure every vote is counted and that this kind of doubt does not happen again." :(

[lonestarnot]

106. you funny one!

:7 Democracy not going down that easily. American's love Democracy and they're just not going to stand by and watch this fraud passively or peacefully. We won't be swept under the rug this time! We plan to fight! Bushitler illegitimate!

[Name removed]

87. Deleted message

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

[davidgmills]

90. freeper talk

Sometimes when their comments are edited, those of us who have made responses to them end up looking like we have made responses to thin air rather than hot air.

[lonestarnot]

105. What side removed?

Doesn't appear to be removed to me....:eyes:

[argh]

97. I posted this before...

but, the subject went off the homepage pretty quick so no one really read it.

The bottom line is I calculated the provisional ballots in Ohio going to Kerry at 55% to Bush's 45%. Someone for the love of god, please prove me wrong on this, because it's been driving me crazy.

---

Something is not right in Ohio unless this AP article Yahoo posted is wrong:
http://story.news.yahoo.com/news?tmpl=story&u=/ap/20041203/ap_on_el_pr/ohio_vote&cid=536&ncid=536

Here's the deal, they don't say how many overseas ballots were counted in the article, but they do say how many provisional ballots were counted. They also say Bush's margin of victory was cut by 17,000 votes. Of these votes 3,893 were from the computer "glitch" that was caught in Columbus.

That would seem to mean that 13,107 votes came from overseas and provisional ballots.

Since the AP didn't say how many overseas ballots were counted, I have to rely on an article which says they weren't expected to go over 10,000:
http://www.ohiocitizen.org/moneypolitics/2004/blackwellletter.html

Adding 121,598 counted provisional ballots and 10,000 overseas ballots brings us to 131,598 total ballots.

To get those 13,107 votes favoring Kerry the votes would have broken down to 59,245 Bush, 72,352 Kerry.

This makes the percentages 55% Kerry, 45% Bush.

See a problem there? That's a massive deviation from the normal ballots: 49% Kerry, 51% Bush. Don't forget that a lot of those overseas ballots are military and supposedly are Bush votes.

Maybe the percentages are skewed because there were more overseas ballots you say? Okay, instead of 10,000 overseas ballots, let's say it's 100,000. That's still 53% Kerry, 47% Bush.

This is a huge red flag. It would seem to me either Democrats were purged from the voter rolls and forced to vote on provisional ballots or the count of Ohio's normal ballots is wrong.

[spooked911]

100. I don't quite get your point.

Everyone expected the provisional ballots to go more heavily for Kerry.

Or are you saying something different?

[Yupster]

109. Agree Spooked

When they first started counting provisionals there were posts on threads here predicting 80-90 % would go for Kerry. Now 55 % to Kerry is supposed be wave a red flag because it's too much?

[argh]

115. My point is exactly what I said.

The election went 49% Kerry, 51% Bush.
The provisional ballots look like they are 55% Kerry, 45% Bush.

Both of these results involve ballots cast statewide. That means they SHOULD come pretty close to matching each other.

You do realize that voting on provisional ballots is not a normal thing? It's something you do when something goes wrong, like when you don't show up on the list as a registered voter.

By having such a wide deviation - and it is a wide deviation - it would indicate - almost prove - that a certain group of voters were removed from the voter rolls. In other words: fraud.

As far as predicting 80%-90%... yeah that would be a flaming red flag. But, I'm not predicting. I'm extrapolating real numbers from real articles.

Just because someone raised the bar too high before we had real numbers doesn't mean the data I have is any less compelling... if it's correct.

That's my problem. I want to know if I'm right. If I am, then there was clearly voter fraud in Ohio and you're blind if you don't see it.

[Yupster]

120. You're assuming people casting

provisional ballots are a random sample of Ohio voters, and they're not.

They are people who never voted before, or haven't voted in a long time, or people who were voting away from where they lived at college or wherever.

These are the exact people the 527 groups supporting Kerry spent yens of millions of dollars registering and trying to get to the polls. Now when they voted for Kerry we're supposed to be surprised?

[argh]

123. I don't understand your point at all.

Provisional ballots SHOULD be a random sampling as far as who the ballot is cast for.

They should be random because they should only be cast when there is an error in the process. If the errors are not random, then there is something going on.

As far as "these are people who never voted" etc. You're just making an assumption and even if you were right it wouldn't matter. Do you really think the Bush campaign ignored these people?

[newyawker99]

Hi argh!!

Welcome to DU!! :toast:

[newyawker99]

163. Hi argh!!

Welcome to DU!! :toast:

[newyawker99]

164. Hi argh!!

Welcome to DU!! :toast:

[lonestarnot]

104. Not purged flipped....

Cliffie got goods!

[TruthIsAll]

117. Check out the Time Zone red shifts

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x141048

[davidgmills]

118. Cluster sampling

Truth Is All:

Last night I was thinking about what a professor had said on another site regarding exit polling several days ago. He used the analogy of putting M&M's in a jar.

Obviously if the jar, say half full of 500 blue M&M"S and half full of 500 red M&M's are sampled, as long as they have been placed in the jar one red one, then one blue one, at a time, and the jar is thoroughly mixed, you can get a very accurate sample when you randomly take out fifty. The key is to make sure the red ones and blue ones are randomly mixed. If you make sure they are randomly mixed, then once you take out fifty at random, you will get pretty close to 25 red ones and 25 blue ones, and can easily make a very accurate projection of how many reds and how many blues in the entire jar.

However, if you fill the jar with a cup of red ones and a cup of blue ones at a time, and do not mix up the jar, you will get "clusters" of red and "clusters" of blue in the jar. In this case, then it would be very possible that in a random sample of fifty of these in the "clustered" jar, it would not be unlikely to get something like 32 reds and 18 blues, making projections of total reds and total blues way off.

What Mystery Pollster and McDonald are saying, I think, is that exit polling in America is much more akin to the "cluster" example than the truly random sample. If so, you must account for this clustering effect. This does make sense because I am sure they poll many mostly red and mostly blue precincts as Republicans and Democrats tend to "cluster" geographically. They also, for financial reasons, cannot poll every precinct.

My understanding of Freeman's work is that he first assumed a truly random sample when he came up with his first analysis which produced 250 million to one odds. Then Mystery Pollster convinced him that he needed to take into account the clustering effect. Using Mystery Pollster's guidance on what that effect might be, Freeman then came up with 650 million to one odds. But then Mystery Pollster, after having done some academic research on cluster effect, decided that he had greatly underestimated the effect. As far as I know, Freeman has not reworked the problem using more accurate numbers on the clustering effect.

On top of that, I think there was also a question of the confidence interval. Mystery Pollster says you must get to a 99.5% confidence interval and (I am guessing here) I surmise Freeman was only able to get to 95%. How you get from 95% to 99.5% I'm not quite sure, other than to have a much larger sample, which may not be possible to get if enough people were not exit polled.

I think what McDonald is saying is that your analysis is not taking into account the clustering effect which is skewing your results. Maybe the confidence interval as well also.

[c-macdonald]

119. well said!

I couldn't have said it better myself.

As for the 95% to 99.5% confidence interval, it has nothing to do with the size of the sample but rather the analysis on the sample. Using a 95% confidence interval means that 5 out of 100 times (or 1 out of 20 times), the actual results will fall outside of our predicted interval. In an election with 51 separate polls, we're bound to get 2-3 results that fall outside of our "predicted" 95% confidence interval just due to chance alone.

A 99.5% interval would eliminate that possibility and would lead to greater accuracy in our results.

You explained everything else better than I could have. Great analysis.

[davidgmills]

121. davidgmills

Thank you. I apologize for misspelling your name.

Unfortunately, while the professor I mentioned helped with the idea of random sampling, he said nothing about clustering. I had to figure out how clustering might occur in a jar.

But anyway, now I get the idea, and more greatly appreciate the difficulty. I also see how people could greatly disagree as to what variable should be used to simulate the clustering effect so I could see how any consensus on the accuracy of exit polling would be highly unlikely.

I suspect in countries like Germany, either they are able to poll every precinct, or they have proven variables on clustering to get the accuracy Thom Hartmann claims they have.

One other thing. Is it possible to get to a 99.5% confidence level and if so how?

Also, would you dare venture a shot at the math?

[davidgmills]

122. davidgmills

One more thought just occurred to me.

Suppose you have 1,000 "clustered" jars. At some point, the clustering effect should begin to cancel itself out and move toward true randomness, wouldn't it?

In other words, 32 reds in one jar should be cancelled out at some point by 32 blues in one jar. Right?

[TruthIsAll]

124. Pollsters ALWAYS use the 95% confidence interval in their work.

"On top of that, I think there was also a question of the confidence interval. Mystery Pollster says you must get to a 99.5% confidence interval and (I am guessing here) I surmise Freeman was only able to get to 95%. How you get from 95% to 99.5% I'm not quite sure, other than to have a much larger sample, which may not be possible to get if enough people were not exit polled"

That is patently untrue. The standard confidence interval used in statistics is 95%. That means that the population mean will be within 2 standard deviations from the mean 95% of the time.

Pollsters NEVER use 99% confidence intervals (3 standard deviations) around the sample mean because the confidence interval around the mean would be too large to be meaningful.

When you have a 95% confidence interval and 5% tails, the probability the actual population mean would deviate from the MOE IN ANY GIVEN STATE is 2.50% for Bush and 2.50% for Kerry.

For Bush the actual vote deviated beyond the MOE 16 times. That means Bush beat 40 to 1 odds in 16 out of the 41 states which deviated in his favor.

For Kerry, this did Not happen even ONCE in the 10 states which deviated in his favor.

To declare that the 99% confidence interval should be used is a canard which has no basis in actual polling. Pollsters ALWAYS use the 95% confidence interval in their work.

Hope that clears it up.

[c-macdonald]

125. they actually do use 99%...

MysteryPollster (who is a professional pollster with years of experience in polling) as well as other professional pollsters have said that they use a 99.5% confidence interval for the EXIT POLLING, not all polls in general. The reason is that using a 95% C.I. for 51 states (incl. D.C.) would result in 2-3 results falling outside the MOE due to chance alone. That's why on election night they rely on a 99.5% confidence interval to call a state for a candidate. This is cold-hard solid fact and, frankly, I'd listen to an expert on polling on this one.

Your discussion of MOE isn't exactly clear though. It all depends on whether or not you have a one-tailed test or a two-tailed test. I've been taking a hard look at Freeman's data (with input from MysteryPollster and another blogger) and I think a one-tailed test would be better to use (the blogger insists on a two-tailed test).

But in either case, you have to look at Bush's and Kerry's results separately, meaning that there's a 2.5% chance Kerry's vote would be above the 95% C.I. and 2.5% chance it would be below (and likewise for Bush). There really is no way (that I know of) that you can combine the Bush and Kerry results to get an MOE as you described.

I've discussed several times on here about your calculations of the MOE and how I think you've underestimated it. I've done my best to get an estimate of the MOE in each state and while the results aren't as overwhelming as they originally were, there are still some very interesting results. Once I do some more checking and verifying, I'll post on my findings. I'll even send you my spreadsheet of the data (full official election returns and Freeman's updated data) if you agree to have a civil conversation about your analysis.

[TruthIsAll]

126. One More Time. I will teach you all about MOE. Step by Step.

"But in either case, you have to look at Bush's and Kerry's results separately, meaning that there's a 2.5% chance Kerry's vote would be above the 95% C.I. and 2.5% chance it would be below (and likewise for Bush). There really is no way (that I know of) that you can combine the Bush and Kerry results to get an MOE as you described".

Now, One more time. Listen close.

1. In any given state, there is a 95% probability that the voting tally will fall within the margin of error.

2. Ergo, there is a 5% probability that it will not.

3. There is a 2.5% probability it will fall outside the MOE for Bush.
4. Ergo, there is a 2.5% probability it will fall outside the MOE for Kerry.

5. FACT. The Bush vote tallies fell outside the MOE in 16 out of 51 states (including DC)
6. FACT. The Kerry vote tallies fell outside the MOE in 0 out of 51 states (including DC)

7. FACT. The probability that the Bush tallies would fall outside the MOE in at least 16 states BY CHANCE is 1 in 13.5 TRILLION.
8. Ergo, since NONE of Kerry's states fell outside the MOE, there is no probability to calculate.

9. FACT. The MOE = 1/sqrt(N), where N is the sample size.
10. FACT. For various N, the MOE is:

N MOE
1000 3.1%
1500 2.5%
2000 2.2%
2500 2.0%
2816 1.8% (this is the FL MOE)

REMINDER. Check the deviations for each state and compare to the calculated MOE and you will see what I'm talking about.

QUESTION I. Is THAT now clear enough for you to understand?

QUESTION II. Do you get it yet?

[davidgmills]

127. MARGIN OF ERROR -- Proposal for New Exit Polls Now

On Stones-Cry-Out website this weekend Rick made it clear to me that the margin of error for Florida was 1.8% if you don't account for THE DESIGN DEFECT CAUSED BY CLUSTERING.

You apparently agree with him on that point and I presume he would agree with you on the other percentages you calculated.

But he insists that the Florida margin of error with "design defect" for clustering is 3.3% not 1.8% and that Ohio and Pennsylvania were 3.5%. Some states, due to low sampling, had margins of error of 4 or 5%.

So the question, to Rick and others like him (Mystery Pollster, Mytofsky, macdonald -- let's call them the "clusterists") at least, is how many of Bush's states are outside these margins of error not how many are outside a 2.5% margin of error.

My point to him was that if mathemeticians can not agree on what the effect of clustering is and how best to calculate it, let's have another National exit poll done right now (a random one to avoid the problems of clustering) of between 20,000 to 40,000 Americans (whatever number it takes to get a random sample that is highly accurate) done by reputable pollsters like Zogby (Democrat), Gallup (Republican) and Harris (middle of the road, I think). Have each pollster poll about 1/3 of the sample and make the results public immediately for academecians everywhere (even outside the US) to have a fair shot at analysis. There is no rule that says that an exit poll has to be done the day of the election.

Maybe do the same for key swing states, but make sure the sample again includes enough voters to be a highly accurate
exit poll (I would hope down to .05% and preferrably less).

Otherwise, I just think we are going to have mathemeticians arguing with each other about the limitations of clustering and sample size. Let's take those arguments off the table.

If we made a demand to have this done, I am sure we could get the funding from places like MoveOn and insist that the Democratic Party help out. We could insist that the Republican Party help out, as a gesture of good will, though I doubt they would.

[TruthIsAll]

128. Clustering. Design defect. Or Spin Defect? Something new

You say:
"On Stones-Cry-Out website this weekend Rick made it clear to me that the margin of error for Florida was 1.8% if you don't account for THE DESIGN DEFECT CAUSED BY CLUSTERING.

You apparently agree with him on that point and I presume he would agree with you on the other percentages you calculated.

But he insists that the Florida margin of error with "design defect" for clustering is 3.3% not 1.8% and that Ohio and Pennsylvania were 3.5%. Some states, due to low sampling, had margins of error of 4 or 5%".

My response:
I don't agree with him because, quite frankly, I have never heard of a system design defect that could not be resolved in 25 years. What has Mitofsky been doing these past 25 years other than refining his craft? You would think that the "clustering" problem (if indeed, it is a problem) would have been solved by now.

How come this "clustering effect" never came up before as an excuse for "bad" polling? Maybe because the polls have always been accurate, after all. I believe Mitofsky knows what he'd doing. Even if he doesn't believe so.

Look, I am NOT a pollster.
I am NOT a statistician.
I DO have three degrees in applied mathematics.
And I DO have a pretty good sense of when I'm being bushit to.

I DO know that exit polls are accurate everywhere else. But the experts caim they aren't here, for some reason.

And the "clustering" is just a excuse for it. Of course, the mass media will just pick it up as another talking point. Another straw man. Another sound bite.

[davidgmills]

129. Maybe you have proven that clustering has cancelled itself out

I'm just a lawyer. I did have calculus in college (perhaps it is better to say it had me). I am pretty good at understanding the concepts but am totally inadequate to do the math.

All I see, as a layman is mathemeticians arguing with each other over the effects of this concept they call clustering. But I figure that if I can't figure who is right or wrong, certainly most journalists can't either. They would much more likely concentrate on the Scott Peterson trial.

So what about my idea of having another National exit poll that rules out the clustering effect?

I think if we had one, it would prove you correct as I am inclined to believe that the national pollsters did not get the polls that wrong the day before the election. Your analysis of last minute pre-election polls vs. late exit polls above certainly indicates that the clustering effect is fiction. If it were real, it should have showed up in your analysis, but it didn't as far as I can tell. I mentioned earlier that if a clustered sample is big enough, the clustering effect should cancel itself out. Your analysis tells me that any clustering effect was cancelled out.

[davidgmills]

131. A random sample and a cluster sample should not be the same

Truth is All:

A random sample and a cluster sample should not produce the same percentage if the cluster phenomon exists.

This is another way to look at it if the cluster phenomonon exists. The pre-election polling is a random sample and the exit polling is purported to be a cluster sample.

If the cluster effect exists, it should be a different percentage. Yet you have shown them to be the same in a nationwide sample.

Ergo, the cluster phenomonon does not exist on a nationwide sample. Am I right or am I wrong?

[TruthIsAll]

132. You may be on to something. Pass it by someone else,

I'm taking a break. And I won't be drawn into what I have already called a strawman argument.

[davidgmills]

134. who?

The world is short of mathemeticians.

[c-macdonald]

133. now let me teach YOU something about MOE...

The clustering effect has ALWAYS been an issue in exit polling. In 1996, the effect was adding a 30% increase to the standard error in every state. I'm not sure what it was in 2000, but this year it adds a 50-80% increase (I don't know why, but that's what professionals who studied the exit polls are saying - I'd assume it's due to the closeness of the election and the fact that clusters of groups were so similar this year - i.e. rural = bush, urban = kerry) but regardless of how many degrees in mathematics you have, you really cannot, I repeat, CANNOT argue that clustring has no effect on the MOE or the standard error. It really does go against standard statistical theory and decades of research. There's just no argument.

Now, to be as succinct and terse (and not to mention obnoxious as you seem to be), I'll move on:

FACT: The MOE is greater because of cluster effect.

FACT: Even with this adjustment, THERE ARE SIGNIFICANT RESULTS.

FACT: I have all my data in a spreadsheet and I am analyzing it as we speak so that I can summarize and hopefully correspond with Dr. Freeman so he can update his numbers.

FACT: I'm not saying the exit polls were right (they obviously weren't). I just saying that you don't know enough about statistics (or MOE) to argue your points about your calculations. If you'd like to see my data with the new MOE calculations, just ask - and be nice about it.

The bigger jerk you are, the less likely I want to even correspond with you but I'm willing to work WITH you, not AGAINST you, to get this message out because it seems you have a good following on DU and you can get a message spread around very rapidly and read by a lot of people.

So please just listen to what I'm saying, do a little bit of research on "clustered sampling" and read a bit on the NEP website (www.exit-poll.net) about their methods and then let me know if you see my point.

My point is not to disprove that there was fraud in the election - my point is to make a statistical analysis of the exit polls more powerful and less susceptible to doubt. We have the same goal here so please, don't jump down my throat when I question your analysis. I'm just trying to be as complete and thorough as possible.

[davidgmills]

135. Truce

macdonald and Truth is All-- I beg for a truce

[c-macdonald]

136. TruceIsAll i want...n/t

[TruthIsAll]

137. You can't convince me that your arguments are not just smoke.

You don't give up, do you? Aren't you dizzy from all that spinning?

You loved the exit polls in the Ukraine.
Why is that?

The exit polls all over the world, including your favorite, Ukraine, have been correct to within 1-2%.
Why is that?

They always "CLUSTER" (I know you love your new mantra) within 1-2% of the vote.
Why is that?

Isn't the bottom line, the ONLY thing that counts, the track record accuracy of the exit polls?
Do you agree with that?

In Germany, they have been accurate to within 1% in each of the last 3 elections.
Why is that?

You never ONCE consider the possibility of FRAUD.
Why is that?

Give it up. You will never be able to logically dispute the essential correctness of my analysis. But you and your cohorts will keep trying.

Take it up with Mitofsky. Ask him to release his exit poll track record going back 25 years.
Do you think he will?

I'm sure he'll tell you what a great job he USED to do as an exit pollster. But now he finds they are no longer accurate.
Why is that?

Ask him why he didn't release exit polls while with VNS in the 2002 mid terms.
Why is that?

Those probability calculations really hit a nerve with you guys, didn't they?
Why is that?

You never thought someone could calculate the odds of fraud based on exit poll anomalies using the classic Normal and Binomial distributions of Probability & Statistics 101.
Did you?

You guys needed the exit polls. But you never considered they would be analyzed as I and others have.
Did you?

Bernoulli and Gauss and Chebyschev must really be enjoying this.
Do you know those guys?

[davidgmills]

138. And one last time

Would an exit poll of 40,000 Americans now satisfy you both?

[TruthIsAll]

139. How about a total revote? By paper. Hand counted.

Keep Diebold and ES&S and Mitofsky out of it.

Just get three people in each precinct to count the votes:
A Dem, Repub and an Independent.

No machines. No exit pollsters.
Just people who can add one and one.

If it's good enough for Oregon, it should be good enough for Ohio.

[c-macdonald]

141. you really are a jackass

Seriously. You're not even listening to what I'm saying. So instead of arguing, I'll show you what MY calculations have uncovered:

In comparing the projected Margin of Victory and the actual Margin of Victory, I have some interesting results in 13 states:

State Final Poll M.o.V.
M.o.V. M.o.V. S.E. Z P-Value
NEW HAMPSHIRE 1.37% 11.00% 2.63% 3.66 0.0126%
NEW YORK 18.04% 27.00% 2.98% 3.00 0.1333%
NORTH CAROLINA -12.43% -5.00% 2.77% 2.68 0.3644%
SOUTH CAROLINA -17.06% -9.00% 3.01% 2.68 0.3731%
VERMONT 20.14% 31.00% 4.23% 2.57 0.5138%
OHIO -2.11% 4.00% 2.55% 2.39 0.8337%
PENNSYLVANIA 2.50% 8.00% 2.41% 2.28 1.1361%
FLORIDA -5.01% 0.00% 2.38% 2.10 1.7816%
NEVADA -2.59% 3.00% 2.77% 2.02 2.1597%
ALABAMA -25.62% -17.00% 4.55% 1.89 2.9230%
MINNESOTA 3.48% 8.00% 2.42% 1.87 3.0925%
DELAWARE 7.59% 16.00% 4.57% 1.84 3.3030%
NEW MEXICO -0.79% 4.00% 2.88% 1.66 4.8151%

These 13 states show that the difference between the Poll margin of victory and the actual margin of victory is NOT due to chance alone. In fact, if our null hypothesis is that Actual M.o.V. <= Poll M.o.V. then we would reject this hypothesis for all 13 states at the 95% level and for 6 of the 13 states at the 99% level.

As Captain Douchebag (aka TruthIsAll) has shown on many occasions, doing a simple binomial test for these results yields:

13 out of 51, 95% confidence: 0.000014%
6 out of 51, 99% confidence: 0.000079%

What does all this mean? Well for one, it means TruthIsAll is a dipshit because he assumed I didn't want his analysis to be true. Second, it shows that something went wrong on election day, either with the polls or with the results (or both). Draw your own conclusions, but that's all I'm going to say on it.

[TruthIsAll]

144. You agree! Results cannot be due to chance. How come it took you so long?

You have just proved my case, only your probability numbers are still too low. But you are agreeing with the math and the approach.

I say AT LEAST 16 states of 51 exceeded the calculated MOE, and AT LEAST 23 of 51 exceeded a 2% Exit Poll MOE, if you believe exit polls are at least that accurate. I do. You look at only 13 states. Why don't you look at all 51?

And as for that 99% Confidence Interval, we can stipulate that it means using 3 standard deviations around the sample mean, which means a higher MOE, which means an increase range of acceptable deviations, which means that fewer states will fall outside the MOE, which means it easier to fool those who are unaware of the statistics. It's very slick way to try to frame the issue, but it's transparent to anyone who knows a little statistics. I'll stick with the tried and true 95% confidence limits as a basis for the probabilty calculations.

I used the SCOOP Exit Poll numbers. I have not updated the analysis the latest Voting tallies - which are raising Kerry's vote. Too little, too late. Just like Gore in 2000. The Media has already said that Bush won. You would expect Kerry's percentage to rise as final votes come in - because he won MORE than 50% of the vote in the first place.

BTW, I don't appreciate being called a dipshit or a douchebag, not even from one who agrees with me. Very unprofessional.

Did I ever resort to name-calling in my posts? Never. Just called you out on the facts - with a few strategic questions.

So this is your response:
"TruthIsAll, you are a dipshit and a douchebag. But you are right."

Well, thank you.

[c-macdonald]

152. you're still a douchebag...

And you know nothing about statistics. As for my analysis above, I did test all 51 states (incl. DC) but the 13 were the only ones that had significant results.

Get a clue, take a statistics course above the 100 level, and then tell me your elementary analysis is better than mine. Whatever, I'm done with you on this topic. You're too close-minded and bullheaded to listen to reason and criticism and I'm just not going to bother with arguing anymore. I have better things to do with my time and hopefully you have better things to do with yours.

To sum, "You're an annoying dipshit and a douchebag, all of your statistics are wrong, but you're on the right track and you have the right idea."

That's all. End of story. I'm done.

[davidgmills]

148. Can You put odds on these numbers?

Like Freeman did. Something like 650,000 to one like Freeman. To us laymen, confidence numbersin the tens of thousands of a percent are difficult to understand. We like odds better.

[davidgmills]

150. Post 148

Was for you macdonald. got it in the wrong place twice.

[Name removed]

142. Deleted message

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

[TruthIsAll]

143. I really touched a nerve, didn't I?

"But hey, at least you care. You're an ass, and a complete obnoxious jerk, but at least you care."


Hey, you don't care. You're an ass, and a complete obnoxious jerk, and you don't care.

Gotcha. You can't spin anymore.

[c-macdonald]

153. great job of moderating...

you delete my post but leave this one? Great job!

[TruthIsAll]

149. Why would you want to work with me? ...

You called me a jerk, douchebag...

Do you really think that I would want to work with you? I'm doing just fine plugging away by myself, but thanks for asking anyway.

I sure wouldn't want to work with a douchebag, so let me commend you on your willingness to work with one. You must be a liberal.

I realize you and your buddies are frustrated, but jeez, to resort to that level of discourse is most revealing.

Besides, and this is even more important: You have no credibility. Your arguments are weak, specious and lack merit. That is my opinion, based on everything I have seen.

Your 99% confidence interval is a straw man I addressed in a few earlier posts. Your attempts at cluster analysis is another one.

If exit polls have been correct to within 2% historically, applying your 50% cluster factor (which I don't buy) brings the MOE to 3%. And you are back where you started. You calculate the probabilities based on the 3% and you will still get

********** ONE OUT OF 13.5 TRILLION ************

Why would you want to work with me when I disbelieve your premise?

I'll stick with my analysis. It must be right on, because since you and others can't argue the math, you attempt to

1) Dilute the number of states which were over the MOE by widening the confidence interval.

2) Downplay historically proven exit poll accuracy by adding in a cluster fudge factor.

3) Avoid the fact that the exit polls were good enough for the Administration and the Media to laud their use in the Ukraine, but to completely disparage them here.

Peter Jennings called us conspiracy freaks. Why would you want to work with a conspiracy freak?

I take great satisfaction in presenting a simple, yet powerful mathematical model. It's not rocket science.

In fact, I give you guys a lot of credit. Your persistence and the ingenious straw men you come up with is truly very creative. I could never do it.

I just plug numbers into simple probability functions using Excel. Nothing too exciting about that. But it works. It's worked so well that now you and others fin that your pants are on fire, so are only left with name calling - I'm a douche bag and a jerk. That says it all.

And you still want to work with me.

I take that as a real complement.

[c-macdonald]

154. I give up

I'm not even going to try. The facts and the statistical theory is out there for everyone to find on their own. You can choose to rely on faulty numbers and poor analysis and lose all credibility with anyone outside of D.U. for spreading your bullshit analysis.

Hey I think Kerry won too, but what good does it do to fudge the numbers so that it looks like there was fraud when you can rely on a better analysis and still prove something went wrong?

Whatever, man, keep posting your silly little numbers and feel good about all the "Wow! Great job TIA!" comments from other people who are intelligent enough to know that they don't know enough about statistics to do an analysis.

But as for me, I'm done with this. I'm going to draft up a paper with my findings and see where it goes. But as for D.U., I'm done. I only found this place a about 2 months ago and I thought it was a great place for likeminded people to gather information and analyze it and discuss it from a progressive perspective but apparently I set my expectations too high. So I'm done. This thread needs to be killed but regardless I'm not going to post on it anymore or read any more responses. That's it. I'm gone. Goodbye.

[davidgmills]

157. Please come back

Please.

[jwmealy]

140. You're Right--Oregon and Wisconsin (Paper Ballot States) Confirm it.

Here's my latest offering--Sent to Conyers via Barbara Lee today.

My study looks at the national pattern in election results, specifically focusing on the phenomenon of “red shift”. Red shift describes Bush getting a greater percentage of the vote in the tabulated results than in the exit poll; negative blue shift describes Kerry getting a smaller percentage of the vote in the tabulated results than in the exit poll. My studies of the pattern of "red shift" (see ) led me to the general hypothesis that there could have been large-scale election tampering. A sense that there were certain anomalies in the distribution of red shift in relation to other factors, such as the spread (the final percentage difference between Bush and Kerry), plus a speculation as to approximately how an automatic election rigging program might be written, led to the following test hypothesis:

For the class of states in which, either according to the last-thing Nov. 2 Mitofsky consortium exit polls or to the official tabulated results, the spread between Bush and Kerry is less than 5.6 percentage points, the following equation, which I have named Formula F, will be true:

Red Shift + Final Spread + 1 > Greatest Recorded Red Shift for the 51 States.

The following six points explain why this equation is very unlikely to be true in all cases unless election tampering has occurred.
Point 1. The master hypothesis being tested is that many of the 13 states in the class described above exceed the margin of error because there is an electronic intervention system in place to fix the result for Bush (a radical hypothesis, but testable). The precise mechanism through which this intervention is accomplished will have to be uncovered by others, but I have appended a general sketch (see Appendix 1) as an illustration.

Point 2. For the purposes of this test I grant, in addition to the master hypothesis in Point 1, the possibility that some or all of the red shift effect in some states may be traceable to Republicans walking out of the polls past the exit pollsters for some as-yet-unknown reason, AND I grant that for some reason this effect may be somewhat exaggerated for states in which the media war leading up to the election is the hottest (especially in the swing states with the most electoral college votes at stake).

Point 3. In the case that Point 2 is true, the red shift due to voter exit poll avoidance behavior will only constitute a kind of noise that partially obscures the pattern that must exist in the final results data if Point 1 is true. It will not be able to hide the pattern altogether.

Point 4. The formula "Red Shift + Final Spread + 1 > Greatest Recorded Red Shift for the 51 States" describes what the election results would look like in every state in which some piece of software at some stage in the system was incrementally altering percentages so as to keep Bush ahead by 1% until one of two conditions applied: (1) Bush was projected to win by 1% or more, or (2) the amount of intervention required to give Bush the win by 1% exceeded a certain parameter, hypothetically 5.6%, which is the amount of red shift of Vermont, the highest recorded for the 50 states and Washington DC.

Point 5. If red shift were solely based on voter behavior as hypothesized in Point 2, and there was not in addition an engineered result being produced by something like a software program or programs, then there would be a general tendency for the equation to be true (i.e. the slimmer the margin, the greater the red shift phenomenon would tend to be). However, there would inevitably be a fair proportion of randomness in the relationship between the two parameters (red shift, spread), because there would not be any simple straight-line relationship between the intensity of the media battle (leading to the red shift), plus whatever the random portion of the red shift (simple error traceable to sample size), on the one hand, and the spread, on the other hand. Thus, one could easily contemplate a red shift of 3 (a relatively high number) and a spread of 1 (a relatively low number) for state X. That would still fail the equation (3+1+1>5.6 is FALSE). One could also contemplate a red shift of 1 (implying a relatively low media battle, or whatever is the supposed mechanism driving voter behavior, lowered yet again by a random sampling error favoring Bush), together with a final spread of 3.5. In that sort of case the equation would also be false. And, of course, one could contemplate a red shift of, say, 2, for whatever reason, and a true spread of 2, which would be more or less in the middle. All of these sorts of combinations of the two parameters would fail the equation. If there is no engineering of the result, every one of the states in the class may individually be somewhat more than 50% likely to pass the equation, but the likelihood that all will pass without exception is very small. In particular, I estimate that the odds against New Hampshire, Pennsylvania, Ohio, Nevada, Michigan, Florida and Colorado all passing the Formula F equation as somewhere between 10 and 30 to 1. The result is equivalent to flipping four or five coins and all of them coming up heads.

Point 6. The equation proved true in 9 of 13 members of the class. Two conspicuous instances of failure were Oregon and Wisconsin, which, despite having close final spreads, had exceedingly low red/blue shift. Under the master hypothesis being tested, the reason for this is not far to seek. Those two states are the only two in the entire Union that still use paper ballots. The current hypothesis concerns an electronic vote rigging system. These two exceptions therefore reinforce the result and tend to disprove the most discussed alternative hypothesis—that in states where the race was close and hard fought there would be a lot of red shift. The equation was notably true without exception among the 4 states of the class that went for Kerry in the final tabulation. I emphasize the 4 states that went for Kerry, because the equation does not actually have to prove true for states that Bush won because it could fail in the case that there was a slim true final spread of 1% or so. Therefore, it only takes a small amount of intervention to put Bush ahead. For example, if Kerry is ahead and has a margin of 2%, the program only needs to swap 2.5% votes to Bush, and now Bush is ahead by 1% rather than behind by 2%, so the software is happy. In that case, 2.5+1+1<5.6. I have implicitly predicated my model on the idea that Bush was in trouble in the swing states, and that a fair amount of intervention will have been needed. But not necessarily in every one of the swing states. For example, there are two swing states, namely Nevada and New Mexico, whose exit poll results show Kerry ahead by between 2% and 3%, but the tabulated result shows Bush winning by a bare 1% in each state. That result is entirely consistent with the hypothesis (Point 1), but each state fails the equation, and each state is within the margin of error. So it might just be chance and Bush won, or it might be that those states really went for Kerry, but the software swung the result to Bush without much intervention because the result was very close to start with. Without a different kind of forensic evidence, there is no way to determine which is the case.

See

Points 1 - 6 explain why I regard the results of the Formula F equation test as revealing the equivalent of a statistical fingerprint of a several-states-wide election tampering scheme.

Seeking a Simple and Irrefutable Proof of the Hypothesis
The hypothesized election rigging scheme, in order to be employed to the full without excessive fear of discovery, should be performed on individual voting machines that have no voter verifiable paper trail. The state of Delaware has (1) a statewide uniformity in using such a machine, and (2) a strong win by Kerry. On the current master hypothesis, the predicted result is that the red shift for Delaware would approximate the maximum (cutoff) parameter. Delaware, indeed, had the second-highest red shift in the nation, with 5.1. If the results of the Delaware race could be subjected to the same type of statistical analysis as shown above, the pattern of manipulation would come out in bold relief. Beyond that, if the actual order of votes cast could be reconstructed in even one machine (keeping blind from the order of voters, to preserve constitutionality), the hypothesis of an engineered result could be decisively proven or ruled out.

Until such proof is accomplished, the current analyses should be treated as provisional and suggestive.

Appendix: Hypothetical Method for Fixing an Election Using an Electronic Voting Machine (EVM) Together with Election Management Software (EMS). Based on the known specifications of a widely used EVM.
1. The firmware for the EVM itself (the EPROMS) is written in machine code so no one can ever figure out what makes it tick except by experience.

2. The firmware is designed to work with proprietary Election Management Software (EMS).

3. Written into the firmware is a program designed to switch presidential votes from one candidate to the other on November 2, 2004 according to a certain algorithm. This program knows to look for the names of the presidential candidates and their positions on the ballot, when that information is plugged into each machine by means of the Memory Cartridge supplied by the local election boards. The EMS, together with the instructions for election board officials and the suggested conventions for ballot layout, is designed in such a way as invisibly to give the firmware election tampering program the information it needs to alter votes under certain conditions.

4. Each machine has a hefty 12-volt rechargeable battery that keeps EPROMS board alive at all times, for years on end. Barring some accident, the EPROMS knows what date and time it is. Any time besides November 2, 2004 between certain hours, the firmware will act normal and will not switch votes.

5. After the end of the day on November 2, or after a certain sequence of events that signals to the machine that this particular election is over, each machine’s firmware is programmed to rewrite itself (reprogram its own EPROMS) to erase all trace of the vote switching instructions. That way, if the machine is tested at any earlier or later date than November 2, 2004, tests will come out normal.

6. There are two ways of switching votes: incrementally, and upon completion of a cartridge.

a. Incrementally. When a vote for candidate A is cast, the program does nothing. When a vote for candidate B is cast, the program asks, “Is candidate A ahead by 1% or more?”. If so, then the program does nothing. If not, the program records a vote for candidate A. The program keeps track of how many votes it switches as it goes along. Generally, before switching a vote, the program also asks a second question. “Have more than 5% (or some similar pre-set parameter, which may be partially randomized) of the so-far-recorded votes been switched?” If so, the program does nothing. Otherwise, if Candidate B is in fact far ahead, so many votes will be switched in order to keep Candidate A appearing to be ahead that it will be obvious that tampering has taken place. So the program has a cutoff parameter that essentially instructs it to do as much vote switching as required to keep Candidate A ahead, but tells it to stop when the situation is hopeless, and further intervention will only lead to inevitable detection.

b. Once per Memory Cartridge. When the cartridge is being prepared to be swapped out, the calculation mentioned in 6.a. is made.

7. The EMS scrambles the order of votes cast in its own records, erasing the inevitable pattern that would arise from manipulation.

[TruthIsAll]

145. Wow! Send it to Mitofsky. I'm sure he will be very interested.

Just kidding. Great work.

All independent, honest analysis comes to the same conclusion:

Kerry is the President.

[davidgmills]

146. Love

All three of you guys even if you can't stand each other.

[davidgmills]

147. Moderator

If Rick Brady tries to say something, please do not erase it. It is true he is a Republican, but he really is a very knowledgeable critic and mathemetician. He has his own website (Stones-Cry-Out)which is mentioned by Mystery Pollster and I have had to go to it to tell him what is going on here and relay some of his points. I am not able to do him justice explaining his ideas nor in relaying these guy's work.

[jwmealy]

155. Funny how every time I publish a friendly expert shows up. . .

FYI,

Dear Bill,

I got your telephone message. I have two questions for you that I will need you to answer before we talk further.

1. Is (name removed to preserve the person's privacy) your birth name?
2. Where did you go to high school or do your A levels, and was your picture in the yearbook at least one year?

I ask these awkward questions because I need to rule out the possibility that you are my babysitter (minder, in UK parlance). Just between you, me, and a whole lot of other people, I really believe that there was election fraud on November 2, and I believe that the people who did it had a budget of many tens of millions of dollars.

These hypothetical people are smart enough to set aside a million or two for follow-up, which implies that someone is watching everything that is being published on the internet, and if anyone shows signs of being on the scent of what actually happened, they are approached by a friendly person who is an expert. This person, the babysitter, monitors what the person is doing, tries to lead them away from the scent as gently as possible, and, if all else fails, tries to stay close enough to the person so that he or she can keep key decision-makers informed of the current state of risk of exposure.

I hope you’ll indulge me on this.

Peace,
Webb

[GOPBasher]

151. The only polls that mattered are the Diebold polls on 11/2.

[Griffy]

158. kick

[rdmccur]

159. This is a bit off the subject

I want to keep in touch with jamboi.
I had his website but lost it. I can give my e-mail to him as well.
I believe he was on the up and up and considered him a very valuable contributor. I'm consternated at the decision to remove his ability to post here! Thanks

Ron

[dlaliberte]

160. jamboi is good

I agree, JamBoi was an asset to us. A little break from his hectic pace at DU is probably a good thing for him too. But we need him back - there is a lot of work ahead of us.

http://jamboi.mydd.com/
http://jamboi.mydd.com/story/2004/12/16/193755/17#readmore

[aloneinva]

165. Exit Polls/Recount

I don't mean to sound pessimistic, I wish what you were say was true about FL:

FL 27 51.55 50 47 3 27 50 49 1 27


The final Score in FL was 52% for the shrub, and 47% for Kerry. I may be naive, but I find it impossible to believe anyone could from fraud could, out of over 7,530,165 they could change about 525,000 votes from one side to the other.

That would require a conspiracy so large it couldn't stay secret. It would have to involve poll workers on the ground to be involved.

A few hundred or a few thousand votes are possible to change an election through fraud, but not a few HUNDRED thousand. Then again, nothing from them would surprise me.

[buddysmellgood]

166. Would you ever have believed that our own aircraft would be used

against us to attack our cities? or that Dubya would continue to school to listen to a children's book after hearing of an attack? Or that gag after gag order would be put on those who try to tell the truth about 9/11? Never say never. I don't want to believe it either, but exit polls don't lie.