There are strong indications that election technology was used to rig the 2004 Ohio presidential race by a variety of means—from the unfair allocation of voting machines (leading to long lines and frustrated voters), all the way through a subverted recount.

  • Voters Disenfranchised: Purposefully unfair distribution of voting machines
  • Corruptible Machines: Vendors had installed fraud-prone voting systems in Ohio
  • Partisan Companies: Voting machine vendors had allegiance to the Republican Party
  • Biased Results: Voting machine "glitches" tended to favor the Republican candidate
  • Evidence: The tabulators in at least one county showed suspicious vote patterns
  • Fake Recount: Voting machine contractor subverted the recount by giving cheat sheets

One story stands out as the kingpin of electronic rigging in Ohio 2004. An out-of-state technology firm was hired to provide backup servers for Ohio. That firm also hosted an array of Republican Party websites. Evidence compiled by researchers shows that the firm was uniquely positioned to guide ballot box stuffing and may have had control over the vote tabulators.

  • Republican Secretary of State hired a deeply partisan technology firm: SmarTech
  • SmarTech intercepted vote totals late on Election Night
  • Counties reporting after the intercept all had anomalous election results
  • Network architecture shows SmarTech may have been able to electronically alter votes
  • SmarTech's other clients included officials in the White House and throughout the GOP

Private secrets have no place in public elections and, in fact, engender a system that is vulnerable to tampering, covert manipulation and fraud.
— Donna Brazile

Intercepting Results

Partisan Technology Firm SMARTECH Was Perfectly Placed to Alter Results on Election Night

The architecture map (CLICK GRAPHIC TO EXPAND) of the backup server used to reroute Ohio's election night results to Chattanooga, Tennessee. The partisan firm SmarTech potentially had access to rewrite the election results on the Ohio Secretary of State's website. SmarTech was hired by Kenneth Blackwell and also hosted a long list of Republican websites, including the character assassination site Swift Boat Veterans for Truth and Karl Rove's email.

In 2004, Ohio Secretary of State Kenneth Blackwell hired Michael Connell to design a website that would display Ohio's election results to the public. Connell was a zealous anti-abortion activist whose companies built websites and email systems for the Republican National Committee, Swift Boat Veterans for Truth, and top GOP figures including Jeb Bush and Jack Abramoff. Connell was also the Bush campaign’s chief IT strategist.

Connell’s contract with the Ohio Secretary of State included plans for a “mirror site” that would kick in to display election results in case servers got overwhelmed on Election Day.

For development of the mirror site, Connell subcontracted SmarTech, a partisan IT company based in Chattanooga, Tennessee. SmarTech had deep connections like Connell—the company hosted dozens of high-profile Republican websites, as well as Karl Rove's email.

Election results were routed through SmarTech at 11:14 p.m. on the night of the election. As reported by Craig Unger, author of Boss Rove, and corroborated by election investigator Richard Hayes Phillips, counties reporting after that time had anomalous results favoring George W. Bush.

The votes of the people of Ohio were in the control of a fiercely partisan IT company.
— Stephen Spoonamore

With real-time access to the vote tallies, SmarTech was well situated to tamper with the election results. “The SmarTech people may have been guiding the manipulation of paper ballots in places like Warren County,” said Ohio attorney Cliff Arnebeck in an interview for Harper's Magazine.

Others argue that SmarTech's access to the Ohio election system included the ability to modify vote totals. Cybersecurity expert (and lifelong Republican) Stephen Spoonamore studied the electronic “architecture map” used by Ohio during the 2004 election and concluded that the SmarTech mirror site was purposefully built as the kingpin in a man-in-the-middle fraud attack. As Spoonamore told Craig Unger:

The only purpose I can conceive for sending all county vote tabulations to a GOP-managed Man in the Middle site in Chattanooga BEFORE sending the results onward to the Sec. of State, would be to hack the vote.

Arnebeck subpoenaed Connell as a key witness in a case he was building against Karl Rove for manipulating the 2004 election. He took a sworn deposition from Connell and hoped to have him testify in court against Rove, but the prospective witness died when his private plane crashed en route from Maryland to Ohio. The circumstances of Michael Connell's death were viewed with suspicion by his family, but no criminal investigation was launched.



Distribution of voting machines greatly favored affluent white suburban Republican districts, and led to unprecedented long lines in urban minority neighborhoods. Thousands of citizens were discouraged and disenfranchised in this manner. The problem was particularly acute in Franklin County: At least 124 machines were undelivered, including a truckload intended for Columbus precincts that trend Democratic.



At least 20–30 touchscreen voting machines in Youngstown (Mahoning County) were programmed to default to George W. Bush. Other machines froze, or erased votes. In Franklin County, voters reported that votes for Kerry would literally “fade” from the screen. As many as 2,100 Kerry votes became "undervotes" in this manner.



A technician from the voting machine vendor ES&S reprogrammed all six tabulators in Butler County on Election Day. He had not been invited by the Board of Elections. Suspicious errors were revealed in a subsequent citizen audit:

  • In 9 of 13 precincts, the count of ballots did not match the tabulated results.
  • In 8 of 13 precincts, the ballots were in a nonrandom sequence.
  • In Monroe City 4CA, there were 212 consecutive ballots for Bush.



Voter turnout was improbably high in several Miami County precincts (98.55 percent in one case) and 19,000 extra votes were added after 100 percent of the precincts had reported their results.

Subsequent citizen hand counts revealed a 5 percent discrepancy in 8 percent of of the county's precincts, and voter registration books that did not match the vote counts.

When challenged on this issue, the outside contractor who had managed the vote counting replied that "the additional votes were inadvertently missing from the early returns."

Connecting the dots here, election defender Robert Fitrakis says this is the "first strong evidence suggesting the county's computer central vote tabulator was corrupted."

Bad Voting Machines

Researchers Declare Ohio's Election Technology Wide Open to Insider Manipulation and Hacking

Prompted by mounting evidence of problems with electronic voting machines across the country, Ohio Secretary of State Jennifer Brunner released a report on Ohio’s electronic voting technology in December 2007.

Known as Project EVEREST, the security review aimed to identify problems that might occur under real-world election conditions. Three teams of researchers examined touchscreen and optical-scan electronic voting systems from Elections Systems and Software (ES&S), Hart InterCivic, and Premier Election Systems (formerly Diebold). The researchers had access to voting machines and software source code from the three vendors.

The study found exploitable security weaknesses in all three vendors’ systems, representing practical threats to the integrity of Ohio elections.

The structural flaws they identified were so profound that "reliably correcting them would require re-engineering and redesign of the equipment and software." Some of the specific vulnerabilities they found included:

  • Voters or poll workers able to place multiple votes.
  • Voters or poll workers able to infect the precinct with virus software.
  • Voters or poll workers able to corrupt previously cast votes.

The vendor systems lack basic technical protections necessary to guarantee a trustworthy election.

They also identified vulnerabilities at election headquarters that could "expose precinct or county-wide ballots and tallies to widespread manipulation." Election software running on Microsoft Windows 2000 or XP machines could be compromised by "viruses arriving from precincts, or by an attacker with seconds at the controller terminal."

Their findings posed an enormous challenge for election security. As the study authors conclude, the integrity of the election relies almost entirely on the physical procedures used to carry out the election:

Unmonitored poll-workers and election officials could exploit security failures to circumvent protections or misuse software features to manipulate voting equipment, vote counts, and audit information.

Failures were present in almost every device and software module they investigated. They called it a "pervasive lack of quality" in the coding and manufacturing.

The systems exhibited a near universal lack of effective protections against insiders.

No Accountability

A Corrupt Election Technology Contractor Subverted the Recount BY Providing Cheat Sheets

Ohio Secretary of State Kenneth Blackwell violated the Due Process and Equal Protection clauses of the U.S. Constitution by failing to conduct a valid recount. Several major breaches of election administration stand out among the misdeeds overseen by his office:

  • Counties did not select precinct samples randomly.
  • A full hand count was not conducted when the partial counts did not match.
  • Counties failed to secure and store ballots and machinery.
  • Counties prevented witnesses from observing various aspects of the recount.

Blackwell's office was abetted in this subversion of democracy by voting computer company Triad Governmental Systems, Inc. Triad had provided vote-counting software to 41 counties in Ohio 2004. It also provided the election results to Pike and Madison counties.

During the recount, Triad provided “cheat sheets” informing election officials how many votes they should find for each candidate, and how many over- and under-votes they needed to match the machine count. They also reprogrammed the machines to recount only the presidential race. These violations made a true and legitimate recount impossible.