Partial Summary of Evidence for Electronic Vote Switching that Swung US Elections from 2002 to 2006

[Time for change]

With the proliferation of electronic voting machines in the United States, concern over their accuracy increased steeply in 2004, especially with awareness of substantial discrepancies between national exit polls and the official vote count in the 2004 Presidential election. In that election, John Kerry led in the final Edison-Mitofsky national exit polls by 3.0 %, despite the fact that George Bush won the official vote count by 2.5% – an exit poll discrepancy (henceforth referred to as “red shift”, to indicate official vote counts in favor of the Republican candidate, as compared to exit polls) of 5.5%, a result that could happen only one in a million times by chance. The red shift was especially large in the crucial swing states of Florida (4.9%), Ohio (6.7%) and Pennsylvania (6.5%). All in all, 22 states demonstrated red shifts in excess of the statistical margin of error and not a single state demonstrated an exit poll discrepancy in the other direction (blue shift) beyond the statistical margin of error. Consequently, these surprising results generated much concern about the possibility of systemic nation-wide election fraud.

That concern was magnified by the fact that votes were counted electronically in the good majority of precincts throughout the United States in 2004, as well as by the fact that for many of those precincts no paper trail existed to potentially verify the electronically produced official vote count. In essence, voters simply had to accept on faith the assertion that the machines that counted their votes were programmed to ensure the accuracy of the vote counts and that those machines were not subsequently hacked to corrupt the vote counts. However, subsequent events and analyses seriously threw those assertions into question.

With crucial 2008 elections approaching and with methods of voting in most jurisdictions yet to be decided, I believe it is important to be able to argue the point that electronic voting presents more than just a theoretical danger to our democracy. With that in mind, here is a partial accounting of substantial evidence of stolen national elections, mediated by inaccurate (and probably fraudulent) electronic voting machines, from 2002 to 2006:


2002 Senate and Governor race in Georgia

Andrew Gumbel, in “Steal this Vote”, describes the 2002 Senate and Governor races in Georgia:

On June 10, 2002… six tabulation machines and a touch-screen voting terminal were stolen…. The theft was also an extremely serious security breach, because a technically adept hacker who gained access to the tabulation machines and the associated GEMS election management software could effortlessly – and undetectably – alter the outcome of an election not only in Georgia but anywhere in the United States where Diebold machines were used.

The November 2002 elections in Georgia were screwy in more ways than one. The state had its share of machine malfunctions … Most troublesome, however, were the results of the races for governor and U.S. Senate, which suggested wild double-digit swings in favor of the Republican candidates from the final pre-election opinion polls. Sonny Perdue became the first Republican governor to be elected since Reconstruction, thanks to a sixteen point swing away from the Democratic incumbent, Roy Barnes. And Saxby Chambliss, the colorless Republican Senate candidate, pulled off an upset victory against the popular Vietnam War veteran Max Cleland, representing a nine- to twelve-point swing… But it wasn’t just the opinion polls that were at variance with the result. The voting pattern was also drastically different from Georgia’s open primary … in 74 counties in the Democrat-heavy south of the state, Chambliss improved on his own standing by a whopping 22 points. Were these statistical anomalies, or was something fishier going on? In the absence of a paper backup, or of any hint of transparency from state officials, the question was for the most part unanswerable.

Testimony of Clint Curtis

In October 2000 Clint Curtis was a computer programmer and life-long Republican who worked for the Florida based Yang Enterprises, Inc. (YEI). According to Curtis’ sworn testimony to House Judiciary Committee Democrats in December 2004, while working for YEI he wrote a prototype for a computer program that would switch votes from one candidate to another, at the request of Republican Congressman Tom Feeney in October of 2000. Believing at the time that the purpose of Feeney’s request was to better understand how Democrats might plan to commit election fraud, Curtis complied with the request to write the program and presented it to his employer, Mrs. Li Woan Yang.

According to Curtis’ sworn affidavit Ms. Yang responded to his presentation of the program by saying “You don’t understand, in order to get the contract we have to hide the manipulation in the source code. This program is needed to control the vote in south Florida.” Curtis testified that he believed that the computer program he wrote, or a similar one, was used in the 2004 presidential election to switch votes.

In his affidavit, Curtis also describes a June 2003 meeting with Raymond Lemme, an official from the Florida Inspector General’s Office who was charged with investigating Curtis’ earlier allegations. Lemme told Curtis that he (Lemme) “had tracked the corruption all the way to the top”, and that the story would break shortly. But we will probably never know what information Lemme had obtained, for he was found dead in a Valdosta, Georgia, Knights Inn motel room two weeks later, July 1, 2003, his arm slashed twice with a razor blade near his left elbow.


Reports of vote switching from the 2004 national Election Incident Reporting System (EIRS)

On Election Day 2004 the national Election Incident Reporting System (EIRS) received numerous reports from voters from all over the country with the complaint of attempting to vote for one presidential candidate on an electronic voting machine, where their vote was switched to another candidate. An analysis of these reports showed that complaints of electronic “vote switching” incidents that favored Bush outnumbered those that favored Kerry by a ratio of 12 to 1. Furthermore, complaints from the 13 swing states outnumbered complaints from all the other states by a ratio of 9 to 1. And, 48% of the reports that favored Bush originated from South Florida – the part of the country that Clint Curtis noted in his sworn testimony.


Undercounted votes in the 2004 New Mexico Presidential race

An extensive analysis of the 2004 Presidential vote in New Mexico demonstrated serious problems with electronic voting, with much higher undervote rates in Hispanic and Native American precincts than in Anglo precincts. In precincts using direct record electronic (DRE) voting in 2004, Hispanic and Native American precincts demonstrated undervote rates (for President) of 6.3% and 7.6%, respectively, compared to only 2.2% in Anglo precincts. In marked contrast, the average undervote rate was under 2% for each of the three ethnic groups in precincts using paper ballots in 2004, with no substantial difference by ethnic group. And, the undervote rate for governor in New Mexico in 2006, when all precincts used paper ballots, was under 2% for each of the three ethnic groups, again with no substantial difference by ethnic group. It is also important to note that if the DRE precincts in 2004 had had undervote rates comparable to the paper ballot precincts, John Kerry would have won New Mexico in 2004.


Nation-wide red shifts in the 2006 House of Representatives races

The official nation-wide result for House races on Election Day 2006 showed a Democratic margin over their Republican opponents of 7.6%. Comparing that to a nationwide exit poll democratic margin of 11.5%, the difference between the official count and the results predicted by the exit polls was nearly 4%, just a little bit less than the red shift of 5.5% in the 2004 Presidential election.

The 4% red shift in the 2006 mid-terms – if that represents election fraud – was not nearly enough for Republicans to maintain control in either the House or the Senate. But in 2004, it would have been enough to throw the 2004 Presidential election to George Bush – in the nation-wide popular vote, as well as in Ohio, where Bush “won” his electoral vote victory. As described by Jonathon Simon in “Landslide Denied”, the 2006 exit poll discrepancy was far beyond the margin of statistical error. And, it is also important to note that the calculated red shift when comparing official 2006 House results with exit polls is almost identical to the calculated red shift when comparing official House results with pre-election polls.


Undercounted votes in the 2006 Florida House District 13 race

If the 4% red shift noted above is due to electronic vote switching, that does not mean that any significant portion of that vote switching would necessarily have been detected. The vote counts in precincts that use DRE machines are determined by secret computer programs that produce vote counts that are unverifiable.

Nevertheless, as explained by Paul Krugman, the election in Florida Congressional District 13, which was “won” by the Republican candidate by 369 votes, was almost certainly determined by faulty (whether intentional or not) DRE voting machines. In Sarasota County, which used ES & S voting machines, 15% of voters did not register a vote for the House race, compared to 2.2% to 5.3% of voters who did not register a vote for the House race in neighboring counties. That amounted to almost 18,000 ballots that did not register a vote for the House race in Sarasota County. Furthermore, those who failed to cast a vote in the House race were shown by their other votes to strongly favor Democrats.

Why did an excess of 15% of the voters voting on ES & S machines in Sarasota County fail to vote for a House candidate? The answer to that question can be ascertained from an interview of voters by the Sarasota Herald-Tribune, which found that one third of voters couldn’t find the House race on their ballot, and that 60% said that they did vote for a House candidate, but their vote didn’t show up on their summary page.


Apparent hugely inaccurate vote count in the 2006 Florida House District 24 race

In the 2006 U.S. House election in Florida’s 24th District, Tom Feeney defeated Clint Curtis by 16% of the vote, despite election eve polls that showed Feeney and Curtis to be in a statistical dead heat. An investigation into that election using door-to-door canvassing to interview voters demonstrated that Curtis received 12 to 24% more votes in every precinct canvassed than stated in the official election results. Sworn affidavits were used to document Curtis’ votes. In other words, Curtis’ research team was able to garner substantially more sworn affidavits – far more than needed to swing the election – from voters attesting that they voted for him than actual votes as counted by the electronic voting machines in the election.


What does this all mean?

It is very important to note that electronic vote switching on individual voting machines is by no means the only mechanism that Republicans have used to steal elections in recent years. Other mechanisms include miscounting of votes by central tabulators, voter registration fraud (i.e. illegal purging of voters), and the use of dirty tricks to suppress voter turnout. This post covers only electronic vote switching by individual voting machines.

Though electronic vote switching on individual voting machines is not the only mechanism for stealing elections it is certainly a very important one. Having to trust that electronic voting machines will count our votes accurately and fairly, even when there is no means of verifying whether they have done so, is not something that should be tolerated in a democracy – period. Even so, with the future of our democracy on the line, it is very important, I believe, to be able to point to demonstrable evidence, beyond the potential for election fraud, that crucial national elections in our country actually have been stolen with the aid of electronic voting machines. The above noted thoroughly documented instances, though only a partial list, make a very strong case for that assertion. And it is worth noting that all documented instances of this type during the 21st Century have favored the Republican candidate.