I meant to put this up Wednesday. It's from November 3, 2004.

Exit polls and ‘actual’ results don’t match; Evoting states show greater discrepancy
By The Raw Story.

An analysis of the original AP exit polling, which showed Kerry with a tighter margin and leading in myriad states, raises serious questions about the authenticity of the popular vote in several key states, RAW STORY has learned.

Since the actual outcome of the votes have been called, AP has changed nearly all of their exit polling to tighten the margin. A reason has not been given.

The analysis, first conducted by a poster at the popular Democratic Underground, suggests possible voter fraud in states that do not have electronic voting receipts, and those that limit the media’s access to polls.

Two inquiries placed by RAW STORY with the media contact for the six-network exit polling consortium at NBC News has received no response.

The curious result comes after the head of Diebold, which produces much of the nation’s electronic voting machines, told Republicans in a recent fund-raising letter that he is “committed to helping Ohio deliver its electoral votes to the president next year.”

Here is the full text of the article in case the link goes bad:

http://www.bluelemur.com/index.php?p=388

11/3/2004
Exit polls and ‘actual’ results don’t match; Evoting states show greater discrepancy
Filed under:

* General

— site admin @ 11:04 am Email This

LATEST: Officials admit machines gave Bush extra votes in two states

An analysis of the original AP exit polling, which showed Kerry with a tighter margin and leading in myriad states, raises serious questions about the authenticity of the popular vote in several key states, RAW STORY has learned.

Since the actual outcome of the votes have been called, AP has changed nearly all of their exit polling to tighten the margin. A reason has not been given.

The analysis, first conducted by a poster at the popular Democratic Underground, suggests possible voter fraud in states that do not have electronic voting receipts, and those that limit the media’s access to polls.

Two inquiries placed by RAW STORY with the media contact for the six-network exit polling consortium at NBC News has received no response.

The curious result comes after the head of Diebold, which produces much of the nation’s electronic voting machines, told Republicans in a recent fund-raising letter that he is “committed to helping Ohio deliver its electoral votes to the president next year.”

An exit poll involves asking someone after they walk out of the election booth who they voted for. While not a guide for proving results, it can be a mechanism for ensuring voting accuracy and flagging potential fraud. Exit polls were recently used in Venezuela to ensure the vote was accurate and legitimate.

Perhaps more importantly, while exit polling is unreliable, the odds of President Bush having gaining an advantage from every exit poll in swing states is an extremely improbable coincidence.

In Florida, Bush led exit polling by CNN’s exit polling consortium by just 5355 votes (when the exit polling information is multiplied by the actual vote). Yet he led by 326,000 in the end result. On Wednesday morning, CNN changed their exit polling to favor Bush, saying that had overweighted African American voters.

In Wisconsin, where exit polls put Kerry up seven percent, Bush has a lead of one percent, an unexplained difference of eight percent.

In New Mexico, Kerry led Bush by 3.8 percent, yet Bush leads Kerry by 3 percent in actual reported voting.

In Minnesota, where a new law sharply restricts reporters’ access to polls, Kerry led 9.6 percent in exit polling. Actual voting counts found that Bush trailed by 5 percent, with a 5 percent discrepancy favoring Bush.

Ohio, which does have paper trail capability but does not mandate receipts, had exits showed Kerry and Bush in a dead heat; in the near-final results, Bush led by three percent.

Exit polls put Kerry up by 8 percent in Michigan; actual results show Bush trailing by just 3 percent.

Nevada, which also has electronic voting – though should have mandated paper trails, had a variance of 4.2 percent. Kerry led the exit polls by 1.2 percent, while Bush led reported votes by 3 percent.

Two states with paper trails for voting were within 0.5 percent margin of error.

New Hampshire, which has electronic voting but provides verified receipts, exit polling is within 0.1 percent of the actual vote. Kerry led by 3 percent in exit polling, and 2.9 percent in the actual vote.

Maine, the final state for which analysis of exit polling was conducted before the AP “resampled” their data, was in the Kerry column by 7.5 percent; the end result put Kerry up 8 percent, a variance of 0.5 percent. Maine has no electronic voting.

Kerry does not gain by any significant margin in actual voting in any state for which analysis has been conducted, RAW STORY found.

Exit polling accurately predicted the results in most states with very little error. Where there were discrepancies, they were significant in the +5 percent range, and always favored Bush.

Allegations of voter fraud is not new to President Bush. On November 12, 2000, the Washington Post ran an article suggesting anomalies in the hotly constested state of Florida.

Something very strange happened on election night to Deborah Tannenbaum, a Democratic Party official of Volusia County. At 10 p.m., she called the county elections department and found that Al Gore was leading George W. Bush 83,000 votes to 62,000 votes. But when she checked the county’s Web site for an update half an hour later, she found a startling development: Gore’s count had dropped by 16,000 votes, while an obscure Socialist candidate had picked up 10,000 … all because of a single precinct with only 600 voters.

Early returns from Florida showed the Green Party candidate leading President Bush and Sen. Kerry in two Ohio counties. They later appeared corrected, but raised eyebrows among liberal bloggers.

Another site created a graph comparing the exit poll percentages and the end result; click to enlarge.

Correction: An earlier version of this article said CNN had polled more than 3 million voters, based on information RAW STORY received from the original analysis. The actual polling was of several thousand voters, but was multiplied by the actual vote using percentages supporting each candidate to compare the statistical results. The voting difference, 5355, is what the difference would be if the exit polling queried as many voters as actually voted.