To Believe That Bush Won in 2004, You Must Also Believe …
Updated Oct. 8, 2011
Do you believe the Final 2004 National Exit Poll (13,660 respondents)?
Then you must also believe there were 6 million more returning Bush 2000 voters than were alive in 2004 – a 110% turnout.
And you must NOT believe…
- the unadjusted NEP (13,660 respondents) in which Kerry had a 51.7% vote share.
- the unadjusted State Exit Poll aggregate (76,192 respondents) in which Kerry had a 51.1% share.
The Final NEP was forced to match the recorded vote.
Given the 2000 recorded vote, 12:22am NEP (13047 respondents) vote shares and a 98% turnout of living Bush 2000 voters, Kerry needed just a 73% turnout of Gore voters to TIE Bush. For Bush to win his 3.0 million recorded vote "mandate”, there had to be a 64% turnout of Gore voters. If you believe that, there is a great Chinese restaurant in lower Manhattan near a famous old bridge that is for sale. Assuming an equal 98% turnout of Gore and Bush voters, Kerry wins by 9 million votes with a 52.9% vote share.
Final NEP – forced to match the recorded vote: Bush by 3.0 million votes
| Cast | Recorded | Alive | Voted | 2004 | Cast | Vote shares required to match recorded vote | |||
2000 | 2000 | 2000 | 2004 | 2004 | Turnout | Mix | Kerry | Bush | Other | Turnout/Alive |
DNV | - | - | - | - | 20.79 | 17.0% | 54% | 44% | 2% | - |
Gore | 55.06 | 51.00 | 48.45 | 47.48 | 45.25 | 37.0% | 90% | 10% | 0% | 93.4% |
Bush | 51.76 | 50.46 | 47.93 | 46.97 | 52.59 | 43.0% | 9% | 91% | 0% | 109.7% |
Other | 4.01 | 3.96 | 3.76 | 3.69 | 3.67 | 3.0% | 64% | 14% | 22% | 97.5% |
Total | 110.83 | 105.42 | 100.15 | 98.14 | 122.30 | 100% | 48.27% | 50.73% | 1.00% | -2.46% |
| Cast | Recorded | | | | Total | 59.03 | 62.04 | 1.22 | -3.01 |
Gore | 49.68% | 48.38% | | | | Recorded | 48.27% | 50.73% | 1.00% | -2.46% |
Bush | 46.70% | 47.86% | | | | 122.30 | 59.03 | 62.04 | 1.23 | -3.01 |
No, Kerry won in a landslide: 9.0 million votes
True Vote
| | Cast | Recorded | Alive | Voted | 2004 | True Vote | 12:22am NEP Vote shares | | ||
2000 | 2000 | 2000 | 2000 | 2004 | 2004 | Turnout | Mix | Kerry | Bush | Other | Turnout/Alive |
True Vote | DNV | - | - | - | - | 22.56 | 15.6% | 57% | 41% | 2% | - |
49.53% | Gore | 54.89 | 51.00 | 52.14 | 51.10 | 51.10 | 41.8% | 91% | 8% | 1% | 98.0% |
46.71% | Bush | 51.77 | 50.46 | 49.18 | 48.20 | 48.20 | 39.4% | 10% | 90% | 0% | 98.0% |
3.76% | Other | 4.17 | 3.96 | 3.96 | 3.88 | 3.88 | 3.2% | 64% | 17% | 19% | 98.0% |
| Total | 110.83 | 105.42 | 105.28 | 103.18 | 125.74 | 100% | 52.9% | 45.8% | 1.33% | 7.15% |
| | Cast | Recorded | | | | Total | 66.5 | 57.5 | 1.68 | 8.99 |
| Gore | 49.53% | 48.38% | | | | Recorded | 48.27% | 50.73% | 1.00% | -2.46% |
| Bush | 46.71% | 47.86% | | | | 122.30 | 59.03 | 62.04 | 1.23 | -3.01 |
Sensitivity Analysis
Bush | | Kerry Share | | | |
Turnout | | Gore Turnout | | | |
52.9% | 64% | 73% | 80% | 96% | 98% |
90% | 49.5% | 50.8% | 51.8% | 54.1% | 54.4% |
92% | 49.1% | 50.4% | 51.4% | 53.8% | 54.0% |
94% | 48.7% | 50.0% | 51.1% | 53.4% | 53.7% |
96% | 48.4% | 49.7% | 50.7% | 53.0% | 53.3% |
98% | 48.0% | 49.3% | 50.3% | 52.6% | 52.91% |
| | | | | |
Bush | | Kerry Margin | | | |
Turnout | | Gore Turnout | | | |
9.0 | 64% | 73% | 80% | 96% | 98% |
90% | 0.7 | 3.9 | 6.4 | 12.2 | 12.9 |
92% | (0.3) | 2.9 | 5.4 | 11.2 | 11.9 |
94% | (1.3) | 1.9 | 4.5 | 10.2 | 10.9 |
96% | (2.3) | 1.0 | 3.5 | 9.2 | 10.0 |
98% | (3.2) | 0.0 | 2.5 | 8.3 | 9.0 |
And you must also believe that …
1- Myth: The media provided state and national pre-election and/exit poll data and documented evidence of vote suppression and miscounts.
Fact : The pollsters focused on Likely Voter (LV) pre-election polls, a subset of Registered Voter (RV) polls. The LVs failed to account for the tremendous gain in 2004 Democratic new voter registration. There were 22 million new voters (a combination of newly registered and others who did not vote in 2000). The National Election Pool is the mainstream media consortium which funds the exit polls. The FINAL exit polls are always forced to match the recorded vote. It is standard operating procedure. Unadjusted, pristine exit poll precinct data has never been made public.
2- Myth: There are many explanations as to why the exit polls were wrong (see the Edison-Mitofsky Exit Poll Report).
Fact: But they are myths; none are supported by the evidence. Here are a few of the most common: Kerry voters were more approachable to be exit polled than Bush voters who were reluctant to be interviewed; biased interviewers sought out Kerry voters; returning Gore voters misspoke and said that they voted for Bush in 2000; exit polls are not random samples; U.S. exit polls are not designed to catch election fraud; early exit polls overstated Kerry’s vote; women voted early; Republicans voted late; Gore voters defected to Bush at twice the rate that Bush voters defected to Kerry; the exit polls “behaved badly”; there was a massive Christian fundamentalist turnout; if there was a conspiracy, someone would talk (many have). But the myths were not just from the “liberal” mainstream media. Even Democrats such as Bill Clinton said that Bush won “fair and square”.
3- Myth: The votes were fairly counted (according to Bill Clinton, Al Franken, Michael Moore) and other “liberal” media gatekeepers and conservatives).
Fact: There is no way to prove that the Touchscreen (DRE) machines were not rigged; there are no paper ballots. And historical census voting data indicates that 125.7 million votes were cast as opposed to the 122.3m recorded. Investigative reporter Greg Palast presented government data which documented 3 million uncounted votes. And now we know about the Ohio/Chattanooga “man-in-the-middle” – and the unverifiable vote (mis)counts in all the other states.
4- Myth: Democrats failed to attract first-time voters.
Fact: According to the National Exit Poll (NEP), Democrats have won first-time voters in every election by solid margins since 1992. In 2008 Obama won new voters by 71-27%. The 2004 NEP timeline indicated that Kerry had 62% of new voters at 730pm, 59% at 9pm, 57% at 1222am. But the Final NEP is always forced to match the recorded vote. It indicated that Kerry had just 54% of new voters, a massive 8% decline from the earlier share.
5- Myth: Bush’s 48% Election Day approval rating was not a major factor.
Fact: In every election since 1976, presidential incumbents with less than 50% approval lost re-election (Ford, Carter, Bush I) while those above 50% won (Eisenhower, Johnson, Nixon, Reagan, Clinton). There was a near-perfect 0.87 correlation between Bush’s monthly approval rating and the average national pre-election poll. The correlation was confirmed when Kerry won the 12:22am National Exit Poll by 51-48%.
6- Myth: Bush gained 9% over 2000 in heavily Democratic urban locations and lost 3% in highly Republican small towns and rural areas.
Fact: That is an Urban Legend. Bush lost 3% in Republican small towns and rural areas.
7- Myth: There is no evidence that the vote count was corrupted.
Fact: Bush won 51.5% of the first 116m recorded votes. Kerry won 54.6% of the final 5 million.
8- Myth: Final pre-election polls did not match the exit polls.
Fact: After undecided voters were allocated, weighted pre-election state (Kerry 47.9-Bush 46.9%) and national (Kerry 47.2-Bush 46.9%) polls closely matched the national (50.8-48.2%) and state (51.8-47.2%) exit polls. Final Zogby Battleground polls showed Kerry leading in 9 critical states.
9- Myth: Bogus assumptions were used in the pre-election 2004 Election Model.
Fact: The model was based on final state LV pre-election polls; the only base case assumption was that Kerry would capture at least 60% of the undecided vote. The model used a 5000 election trial Monte Carlo simulation. It determined that Kerry would win 337 electoral votes (with a 99.8% probability of winning at least 270 EV). Kerry’s projected 51.3% national vote share and electoral vote were confirmed by the Interactive Election Simulation Model.
10- Myth: There is no evidence that undecided voters break for the challenger.
Fact: Historical evidence indicates that undecided voters break for the challenger over 80% of the time - especially when the incumbent is unpopular - and Bush had a 48% average approval rating. World-class pollsters Harris and Zogby reported their late polling indicated Kerry would win 60-80% of the undecided vote. Gallup gave 90% of undecided voters to Kerry. The National Exit Poll low-balled Kerry’s share: 55-60%.
11- Myth: Bush was leading in the final pre-election polls.
Fact: Kerry led Bush by 1% in the state-weighted pre-election average and by 3% after allocating undecided voters. Kerry and Bush were tied at 47% based on the final national 18-poll average.
12- Myth: Non-response bias in the exit polls was the reason why 43 states red-shifted to Bush.
Fact: Approximately 3 million votes (most in heavily Democratic minority districts) were never counted. Response rates were lowest in Kerry urban strongholds.
13- Myth: It was just a coincidence that Oregon was the only battleground state that shifted to Kerry (in an exit pollster telephone poll).
Fact: Oregon was the only state which voted by mail or hand-delivered paper ballots AND mandated hand counts of randomly selected counties - a powerful election fraud deterrent. Kerry did worse than Gore in the other battleground states because none had an equivalent fraud deterrent. Unverifiable DRE touch screen computers were used to calculate 30% of the votes .
14- Myth: The exit polls did not indicate that electronic voting machines are fraudulent.
Fact: All voting methods had high average discrepancies - except for paper ballots. Lever machine precincts had the highest (11%) discrepancies. Unverifiable touch screen (DRE) and optical scan precincts had 7%. Paper ballots just 2%. There were 88 reported touchscreen vote switching incidents - 86 switched votes from Kerry to Bush (a zero probability).
15- Myth: The exit polls behaved "badly".
Fact: Final state and national exit polls are ALWAYS FORCED TO MATCH the recorded vote. It's standard operating procedure. But the premise is false: a fraud-free election is assumed. Uncounted votes alone prove that elections have been anything but fraud-free. And millions of phantom voters indicated by the Final 2004 National Exit Poll proves that it cannot be correct.
16- Myth: Kerry led in the early exit polls, but Bush passed him in the final.
Fact: Kerry led the National Exit Poll (NEP) by a constant 51-48% from start to end. He led at 4pm (8349 respondents), 730pm (11027) and 1222am (13047). He led the state aggregate unadjusted exit polls by 52-47%. Bush won the Final NEP (13660) by 51-48%. It is a mathematical impossibility that 613 additional exit poll respondents could cause Kerry's 51-48% margin (at 12:22am after the polls closed) to flip to Bush. The Final was forced to match the recorded vote.
17- Myth: The margin of error used to calculate probabilities of exit poll discrepancies were too low.
Fact: Even assuming a 60% “cluster effect”, the probabilities were near zero. Assuming zero cluster, WPE-adjusted state exit poll discrepancies exceeded the MoE in 29 states for Bush and just one for Kerry. For 30%, the MoE was exceeded in 24 states for Bush. Composite (12:22am) exit poll discrepancies exceeded the margin of error in 16 states for Bush; none for Kerry.
18- Myth: There is nothing suspicious about the fact that all 21 Eastern Time Zone states red-shifted from the exit poll in favor of Bush.
Fact: The probability of the shift is 1 in 2 million. But 14 exit polls shifted beyond the MoE - a ZERO probability.
19- Myth: Exit polls are not true random samples.
Fact: Pollsters Edison-Mitofsky stated that respondents were randomly-selected (1% MoE) in National Exit Poll notes and in the NEP Methods Statement.
20- Myth: Bush voters were reluctant to respond to exit pollsters.
Fact: this was contradicted by the Final Exit Poll, which stated that Bush 2000 voters comprised 43% of the respondents, as compared to 37% for Gore voters. It was also contradicted by a linear regression analysis: non-response rates increased going from the strongest Bush states to the strongest Kerry states, indicating that non-responders were Kerry voters.
21- Myth: Ohio, Florida and National exit polls show that Bush won.
Fact: FINAL State and National exit polls are ALWAYS forced to match the recorded vote, even if its is miscounted – as it was in 2004. Unadjusted state and national exit polls showed Kerry winning by 52-47% and 51-48%, respectively.
22- Myth: The Final NEP 43/37 Bush/Gore returning voter mix is possible. After all, it's just a poll.
Fact: The 43/37 mix was not a polling result. It was contrived to force a match to the recorded vote. This is the incontrovertible proof: Kerry had 7074 (51.71%) of the unadjusted 13660 NEP respondents. Bush had 6414 (46.95%). Of the 13660, 3182 were asked who they voted for in 2000: 1257 (39.50%) said Bush, 1221 (38.37%) said Gore. When the 39.5/38.37 mix is applied to the 12:22am NEP vote shares, Kerry has 51.74%, exactly matching the unadjusted NEP. This exposes the Final NEP (Bush 50.7-48.3%) 43/37 returning Bush/Gore voter mix. Note that 13660 is the Final NEP sample size. This is proof that the Final NEP 43/37 mix was forced – not an actual sample.
Fact: The mix could not have from changed from 41/39 at 12:22am to 43/37 with just 613 additional respondents. Bush 2000 voters could not have comprised 43% (52.6 million) of the 122.3 million votes recorded in 2004 since he only had 50.5 million votes in 2000. Approximately 2.5 million Bush 2000 voters died prior to the 2004 election. Therefore, there were at most 48.0 million returning Bush voters in 2004 - assuming an impossible 100% turnout. If 98% turned out, there were 47.0 million returning Bush voters. That means there had to be 5.6 million (52.6-47.0) phantom voters.
23- Myth: The Democratic Underground “Game” thread showed that Bush won with a feasible 39/39% mix.
Fact: In order to force a match to the recorded vote, NEP vote shares also had to be adjusted to implausible levels far beyond the margin of error. The scenario required a) Kerry's share of new voters reduced from 57% to 52.9%, b) Bush's share of Gore voters increased from 8% to 14.6% and c) Bush 2000 returning voter defection reduced from 10% to 7.2%.
24- Myth: The near-zero a correlation between vote swing (from 2000 to 2004) and 2004 exit poll red-shift “kills the fraud argument”.
Fact: “Swing vs. Red-shift” is based on twisted logic by using fraudulent 2000 and 2004 recorded votes to prove no fraud in 2004. There is a strong correlation when the unadjusted state exit polls are used as proxies for the True Vote.
These graphs display Recorded (flat regression line) and True Vote swing vs. Red-shift (sloped line):
Swing Redshift 1992 , Swing Redshift 1996 , Swing Redshift 2000 , Swing Redshift 2004 .
These graphs display Recorded and True Vote Swing vs. Red-shift by partisanship state groupings:
2004 Battleground Swing Redshift , 2004 Democratic States Swing Redshift , 2004 Republican States Swing Redshift
25- Myth: False Recall in the part of Gore voters explains the Final NEP 43/37 Bush/Gore returning voter mix.
Fact: This will put the 43/37 argument to eternal rest and close the book on the False Recall argument. In the unadjusted 2004 NEP (13,660 respondents) Kerry had 7,074 (51.71%) and Bush 6,414 (46.95%). Of the 13,660 respondents, 3,182 were asked who they voted for in 2000: 1,257 (39.50%) said Bush and 1,221 (38.37%) said Gore. When the 39.5/38.37 mix is applied to the 12:22am NEP vote shares, Kerry has 51.74%, exactly matching the unadjusted NEP. This puts the lie to the published Final NEP (Bush 50.7-48.3%) and the 43/37 returning Bush/Gore mix. We have just proved that the Final NEP 43/37 mix is a forced result – not an actual sample.
26- Myth: An NES survey indicates that "a slow drifting fog" caused Gore voters to say they voted for Bush.
Fact: NES used 2000 and 2004 (105.4m and 122.3m) recorded votes as a baseline for the survey, rather than True Votes cast (110.8m and 125.7m). There were nearly 6 million uncounted votes in 2000, approximately 4.5 million were Gore votes. Therefore, Gore won the True Vote by 2-3 million - not the 540,000 recorded. Kerry won 3 of 4 million uncounted votes. Using the True Vote a baseline shows that the NES respondents did not misstate who they voted for in 2000.
Fact: Only 25% of respondents were asked how they voted in 2000. The other 75% 10,000 were asked who they just voted for minutes before. Voters do not forget who they voted for in the previous election, much less how they just voted a few minutes ago. False recall was obviously not a factor in the pre-election polls which matched the exit polls after undecided voters were allocated.
27- Myth: Returning Gore voters misrepresented their 2000 vote because they wanted to be associated with a winner - Bush.
Fact: Bush had a 48% approval rating on Election Day. The majority of new voters were Democrats and Independents who gave him 25-30% approval. It was Gore, not Bush, who was the popular vote winner in 2000 by 2-3 million True Votes. So why would Gore voters want to be associated with Bush? It makes no sense.
28- Myth: Bush gained 12 million new voters in 2004.
Fact: Simple arithmetic shows that Bush needed more than 16 million new voters. He had 50.5m recorded votes in 2000. Approximately 2.5m died and 1.0m did not return to vote in 2004 (assuming 98% turnout). Therefore, approximately 47 million Bush 2000 voters returned to vote in 2004, so he needed 15m (68%) of 22 million new voters to get to 62 million. But according to the 12:22am National Exit Poll, he had just 41% (9 million) of new voters, a full 27% less than what he needed. The probability that he had 68% of new voters is ABSOLUTE ZERO.
29- Myth: Bush won a 3 million vote "mandate".
Fact: Gore won by 540,000 recorded votes over Bush (actually he won by 2-3 million True votes) so Kerry had a head-start. According to the 12:22am National Exit Poll, Kerry had 57% of new (DNV) voters (first-timers and others who did not vote in 2000). He won returning Nader voters by 64-17% and 10% of Bush voters. Just 8% of Gore voters defected to Bush. So how could Bush have won? He needed a massive net defection of Gore voters. No way.
30- Myth: The sensitivity analysis showing that Kerry won all plausible scenarios is not a mathematical proof.
Fact: But it is proof beyond a reasonable doubt when the WORST CASE (implausible) scenario indicates a Kerry win probability GREATER THAN 90% and the BASE CASE (plausible) scenario indicates a win probability GREATER THAN 99%.
31- Myth: Bush’s share of females increased by 4.2% over his 2000 share.
Fact: That's implausible; his share of male voters declined by 0.2%. In the 12:22am NEP, females voted 54-45% for Kerry.
32- Myth: Bush won Ohio.
Fact: There is plenty of documented evidence of uncounted and switched votes and massive voter disenfranchisement. Two election workers were convicted of rigging the recount. In direct violation of standing federal election law, 56 of 88 Ohio counties destroyed all or part of their 2004 election data. The final Zogby Election Day poll had Kerry leading by 50-47%. Kerry led the pristine Ohio exit poll by 54.1-45.9% (based on the average 10.9% precinct WPE). He even led the adjusted 12:22am Composite (which used final pre-election polls) by 52.1-47.9%.
33- Myth: Bush won Florida by 52-47%, a 368,000 vote margin.
Fact: The Democrats had a 41-37% registration advantage in Touch Screen (TS) counties and a 42-39% edge in Optical Scan (OS) counties. Florida pre-election polls were trending to Kerry. After allocating the undecided vote, Kerry led the final 10-poll average by 51.1-48.8%. Kerry was on track to a 200,000 vote win. But Kerry won TS counties (3.86mm votes) by 51-47% and Bush won OS counties (3.43mm votes) by a whopping 57-42%. The final Zogby election day poll had Kerry leading by 50-47%. Kerry won the exit poll by 50.9-48.3%. In 2000, Gore won 70% of 180,000 uncounted under/over votes. If counted, he would have won by at least 60,000 votes. Dan Rather's expose on voting machines proved that poor-quality paper used in Florida punch card machines was a major cause of spoilage in heavily Democratic precincts.
34- Myth: NY pre-election and final exit polls (Kerry 58.5-40.2%) were correct. The unadjusted exit poll (Kerry 64.1-34.4%) was wrong.
Fact: New York and California were rigged to inflate Bush’s popular vote margin and provided at least 2.0 million of his 3.0 million vote "mandate".
1- In 2000, NY voted 60.5% for Gore, 35.4% for Bush and 4.1% for Nader. This is just one example of the impossible scenarios that are required to match the NY vote: a) 100% of returning Nader voters had to break for Bush (he had 17% nationally), b) Bush needed 50% of new voters (he had just 41% nationally), c) Bush needed 11% of returning Gore voters (he had 8% nationally). The clincher: Kerry's NY share was 10% higher than his national share. How could Bush have done so much better in heavily Democratic NY with returning Gore, Nader and new voters than he did nationally? It is extremely counter-intuitive and makes no sense.
2- Pre-election likely voter (LV) polls did not factor in the heavy turnout of new Kerry voters. The pre-election LV poll had a 4% margin of error (MoE); there was a 95% probability that Kerry ‘s share would be in the 54.5-62.5% range.
3- The NY exit poll 3.2% MoE indicates a 95% probability that Kerry's vote was between 60.9-67.3% and also within the MoE of the LV poll (which low-balled projected Kerry turnout).
4- Bush gains from 2000 to 2004 in the 15 largest (Democratic) New York City and suburban counties is an Urban Legend.
35- Myth: Unlike touchscreens and optical scanners, NY lever voting machines cannot be programmed to switch votes.
Fact: Levers are highly vulnerable to fraud. Like touch screen DREs, there is no verifiable paper ballot. Levers did not produce paper ballots; vote counts could not be verified. Defective levers were placed in urban precincts; many voters left the precincts without voting. NY lever advocates refuse to consider this fact: Votes CAST on levers are COUNTED on unverifiable central tabulators. There were many incidents of Lever voting machine failure.
36- Myth: Polling data was cherry-picked and assumptions biased for Kerry.
Fact: The following models all used the exit poll data from 1250 precincts provided by exit pollsters Edison-Mitofsky. The data included partisan response rates and corresponding average within precinct discrepancies. The models produced equivalent results, confirming the USCV simulation which had previously debunked the reluctant Bush responder (rBr) hypothesis.
a) 1250 precinct Exit Poll Response Optimizer: Strong Bush, Bush, Even, Kerry, Strong Kerry.
Kerry won the 2-party share: 52.1-47.9%.
b) Location-size response: Big Cities, Small Cities, Suburban, Small Towns, Rural.
Kerry also won this category by 52.1-47.9%.
c) State exit poll response: Kerry 52.3-47.7%
d) 12:22am NEP “Voted in 2000” demographic: Kerry by 51.9-48.1%
e) Unadjusted state exit poll aggregate (70,000 respondents): Kerry by 52.5-47.5%.
