Avoiding the Systemic Election Fraud Factor

[Richard Charnin]

It is about time that the so-called experts who promote overly complex or overly simplistic pre and post election models started to apply the scientific method and utilize a robust probability and statistical analysis- and factor in the election fraud variable.

Conventional media pundit post-election analysis is based on a faulty premise: that it is a perfectly acceptable practice to force the exit polls to match the recorded vote. It is Standard Operating Procedure. It is also unscientific. And it is faith-based.

Election forecasters and political scientists implicitly assume that the recorded vote is equal to the True Vote and that there is no such thing as Election Fraud. But the recorded vote has never been equal to the True Vote because of uncounted votes. And now, since HAVA was enacted in 2002, the massive State and National UNADJUSTED exit poll discrepancies have been caused by proprietary, unverifiable voting machines and central tabulators programmed to miscount votes.

It is often stated that exit polls were very accurate in elections prior to 2004, but have deviated sharply from the recorded vote since then. But it's like comparing apples and oranges: FINAL exit polls in elections prior to 2004 are being compared to UNADJUSTED exit polls since.

It is not generally known that all FINAL exit polls published in the media have ALWAYS been FORCED to match the RECORDED vote. That's why they APPEAR to have been accurate. On the other hand, the RECORDED votes have deviated sharply from UNADJUSTED exit polls (and the TRUE VOTE) in EVERY election since 1968.

It is the UNADJUSTED exit polls that have ALWAYS been accurate. They closely matched the True Vote in 2000, 2004, 2006, 2008 and 2010. FINAL exit polls have exactly matched the fraudulent RECORDED vote because they have been forced to do so.

This is a link to a 1988-2008 comparative analysis workbook of the unadjusted exit polls and recorded vote:
https://docs.google.com/spreadsheet/ccc?key=0AjAk1JUWDMyRdFIzSTJtMTJZekNBWUdtbWp3bHlpWGc#gid=15


This is a link to the 1988-2008 State and National Presidential True Vote Model:
https://docs.google.com/spreadsheet/ccc?key=0AjAk1JUWDMyRdGN3WEZNTUFaR0tfOHVXTzA1VGRsdHc#gid=0


The experts and pundits claim that likely voter (LV) pre-election polls have been very accurate in matching the recorded vote. That is true. But they won’t tell you that votes are miscounted in every election. Or that their predictions failed to account for the great majority of newly registered Democrats who did not pass the Likely Voter Cutoff Model (LVCM) screen.

They also claim that registered voter polls (RV) don’t reflect actual voter turnout. That is only partially true; not all registered voters turn out. But they don’t tell you that predictions based on RV polls (after allocating undecided voters) closely matched the unadjusted exit polls in 2004, 2006 and 2008.

They don't mention the 11 million uncounted votes in 1988 which may have cost Dukakis the election. Or the six million uncounted votes which definitely cost Al Gore the 2000 election. Or that over 80 million votes (mostly Democratic) were uncounted in the 11 elections since 1968.

The experts and pundits should consider the following factors:
- use votes cast - not recorded – in their analysis.
- employ reasonable assumptions based on both Registered (RV) and Likely Voter (LV) polls.
- indicate that LV polls are a subset of RV polls
- note that the Likely Voter Cutoff Model (LVCM) has a built-in bias against new, Democratic voters.
- allocation of undecided voters and uncounted votes
- apply voter mortality rates before estimating new and returning voter turnout.
- discuss the strong correlation between unadjusted state exit poll shares, approval ratings and Party-ID as shown in this graph:

https://docs.google.com/spreadsheet/ccc?key=0AjAk1JUWDMyRdFIzSTJtMTJZekNBWUdtbWp3bHlpWGc#gid=21

These are just some of the facts which the experts never mention:
- In 1988, Dukakis won the unadjusted battleground state exit polls by 51.6-47.3%, He lost by 7 million recorded votes. There were 11 million uncounted votes.

- In 1992, Clinton won the unadjusted exit polls by 16 million votes (5.6 million were recorded). The Final National Exit Poll (NEP) was forced to match the recorded vote by implying there was a 119% turnout of living 1988 Bush voters. There were 10 million uncounted votes.

- In 2000, Gore won the unadjusted state exit polls (58,000 respondents) by 50.8-44.4%, a 6 million vote margin. It was matched by the True Vote Model (TVM). He won the recorded vote by just 540,000. There were 6 million uncounted votes.

- In 2004, Kerry won the unadjusted National Exit Poll by 51.7-47.9%. He won the True Vote Model with 53.5% - a 10 million vote margin. He lost by 3.0 million recorded votes.

- In 2008, Obama won the unadjusted state exit polls by 58.0-45.4%, a 22 million vote margin. It was matched by the TVM. He had a 9.5 million recorded margin.

- The Democrats won the 1988-2008 unadjusted presidential exit polls by a massive 51.6-41.8% margin. The average was matched by the TVM.

- In the 1988-2008 presidential elections, 148 unadjusted state exit polls exceeded a conservative 3.0% margin of error: 1988:14, 1992:30, 1996:21, 2000:21, 2004:25, 2008:37

- Of the 148 state exit polls that exceeded the 3.0% MoE, 138 shifted to the Republican. The probability that this was a random occurrence is ZERO. The movement from the exit polls to the vote has been in ONE direction.

Election forecasters, political scientists, academics and media pundits never consider the overwhelming evidence that the 2000 and 2004 elections were stolen and that Democratic landslides were denied in 2006 and 2008. Or that Senate and Governor races were likely stolen in 2010. They never mention the massive documented evidence of Election Fraud, yet never stop talking about non-existent Voter Fraud. They talk about voter suppression as if it is the only problem that needs to be addressed.

They refer to the final exit polls for demographic trends. But they won’t tell you that the finals are always forced to match the recorded vote – and therefore all demographic category crosstabs understate the Democratic vote.

They claim that the 2004 pre-election poll average predicted a Bush win. But they don’t tell you that the weighted average predicted a Kerry win. Or that they were using LV polls before allocating undecided voters. Or that the RV polls, adjusted for undecided voters, predicted a 51-48% Kerry win.

They claim that Bush won by 3 million votes. But they won’t tell you that in order to match the recorded vote, the Final 2004 NEP required 6 million more returning Bush 2000 voters than were still living in 2004 -a mathematically impossible 110% turnout.

They claim that in the 2006 midterms, the Democrats won the House by 52-46% (230-205 seats). But they won’t mention that the Democrats won all 120 pre-election Generic Polls and that the trend line gave them a 56.4% share - exactly matching the unadjusted National Exit Poll. Or that approximately 20 House seats were stolen (primarily in FL, OH, NM and IL). The landslide was denied.

They claim that the 2008 pre-election LV polls predicted Obama’s 52.9-45.6% recorded share - a 9.5 million vote margin. But they don’t tell you that the RV polls projected that he would win by 57-41%. Or that Obama had a 58.0% unadjusted state exit poll aggregate share and a 22 million vote margin.

They don't mention that in order to match the 2008 recorded vote, the Final NEP required a 103% turnout of living Bush 2004 voters - which meant there had to be 12 million more returning Bush voters than Kerry voters. Or that the Final indicated there were 5 million returning third party voters even though only 1.2 million were recorded in 2004.

The pundits failed to question why in the 2010 midterms, the Democrats easily won the unadjusted Governor exit polls in Florida and Ohio – but lost the election. Or why Giannoulias easily won the Illinois Senate exit poll – and lost the election. Or why Sestak lost the Pennsylvania Senate race after leading in the exit polls.

You can be sure that election forecasters, political scientists, media pundits and the Democrats won’t bring up the fact that Obama's 2008 True Vote was reduced by a 5% fraud factor. Considering that the 1988-2008 average Democratic True Vote margin was reduced from 10% to 2% by election fraud, Obama needs 55% just to break even. He needs another landslide to overcome the fraud factor.