Why is Mitt Romney so confident? In states where the winner will be decided by less than 10% of the vote, he already knows he will win.

This is no tinfoil hat conspiracy. It’s a maths problem. And mathematics has showed changes in actual raw voting data that had no statistical correlation other than programmable computer fraud. This computer fraud resulted in votes being flipped from Democrat to Republican in every federal, senatorial, congressional and gubernatorial election since 2008 (thus far) and in the 2012 primary contests from other Republicans to Mitt Romney.

This goes well beyond Romney’s investment control in voting machine maker Hart Intercivic and Diebold’s close ties to George W. Bush. Indeed, all five voting machine companies have very strong GOP fundraising ties, yet executives (including the candidate’s son Tagg Romney) insist there is no conflict between financially supporting one party whilst controlling the machines that record and count the votes.

A retired NSA analyst has spent several sleepless nights applying a simple formula for past election results across Arizona. His results showed across-the-board systemic election fraud on a coordinated and massive scale. Interestingly, the analysis demonstrated that this only happens in larger precincts because anomalies in small precincts can be more easily detected.

"Easy to Cheat"

Retired NSA analyst Michael Duniho has worked for nearly seven years trying to understand voting anomalies in his home state of Arizona and Pima County. This publication has written extensively about apparent vote machine manipulation in a 2006 RTA Bond issue election that is still being fought in the courts. Said Duniho, "It is really easy to cheat using computers to count votes, because you can't see what is going on in the machine."

When Duniho applied a mathematical model to actual voting results in the largest voting precincts, he saw that only the large precincts suddenly trended towards Mitt Romney in the Arizona primary – and indeed all Republicans in every election since 2008 – by a factor of 8% to 10%. The Republican candidate in every race saw an 8 to 10% gain in his totals, whilst the Democrat lost by 8 to 10%. This is a swing of up to 20 points, enough to win an election – unless a candidate was losing very badly.

Since sifting through and decoding massive amounts of data on behalf of the National Security Agency was his work for decades, he wanted to understand why this was only happening in large precincts.

Nose Counting

The idea of examining large precinct results came via a link to a report written by Francois Choquette and James Johnson. Choquette became curious about South Carolina primary results in the February Republican contest, in which a poll observer noted an unusually sizable gain of votes for Mitt Romney in large precincts. Choquette wanted to know why.

He examined and applied all of the usual statistical markers to see where a variance might occur: income level, population density, race, urban vs. rural, even party registration numbers. He found no correlation to explain why Romney votes trended upward while Paul and Santorum votes trended downward - except in large precincts.

Choquette then looked at all 50 states and found roughly a 10% switch in votes from GOP to Democrat everywhere except Utah, where the presumption was, because it's Mitt's home state and populated by religious conservatives, there was no chance of Romney losing.

Choquette found that in Maricopa County, which consists of Phoenix and its suburbs, Romney used this technique against John McCain in 2008. But McCain beat him by too much for an ill-gotten 10% gain to matter. McCain tried to do the same thing in the general election, but nine million votes nationally were too many to make up.

Examining every county across America was too massive an undertaking for any one person, so he included a simple set of instructions and encouraged others to do the same with raw vote totals in their county/state:

Download the text files of all raw actual vote results by precinct from the Secretary of State's Office. Arrange them in precinct order. Put in all of the candidate totals for each precinct. Sort the data by total vote smallest on the top.

Here it gets a bit dense: He needed to add columns that show cumulative totals by candidate then compare them by candidate to establish trend lines. That reveals trends which should remain statistically constant throughout an election.

Stealing Votes

But as the spreadsheet shows, the larger the precinct, the more dramatically the numbers start to change.

"If percentages did not change from one precinct to the next, we would see a flat line, but what we are seeing is sloped lines downward for Democrats and upward for Republicans - or, in the case of the Presidential primary, upward for Romney and downward for his opponents," said Duniho.

In every election contest, the trend lines dramatically crossed for no apparent reason. It was revealed that votes were being systemically bled from Rick Santorum and Ron Paul and diverted to Mitt Romney.

Once Duniho completed the spreadsheet, he pumped in actual vote totals from other Arizona election contests.

He looked at every 2010 race in Arizona, from Governor Jan Brewer to Senator McCain and Congresswoman Gabby Giffords. The trends lines all did the exact same thing. Someone had manipulated the election outcome, most likely one person inserting a programme inside the system's central computer that flipped votes.

The results were astounding.

They showed that Governor Brewer actually lost her election and Gabby Gifford's razor thin less than 1% point re-election victory over Tea Party Conservative Jesse Kelly was closer to a 20 point victory for her.

Duniho stressed, "We need to have strong hand count audits to confirm the integrity of these elections. This means comparing hand counts with official reports of the election."

Ohio Precedent

This isn't the first time Republicans have been charged with vote theft. It happened in the 2004 presidential election, in Ohio and Florida. In Ohio, GOP consultant Michael Connell claimed that the vote count computer program he had created for the state had a trap door that shifted Democratic votes to the GOP.

He was subpoenaed as a witness in a lawsuit against then-Secretary of State Ken Blackwell, and lawyers for the plaintiff asked the Justice Department to provide him with security because of threats made against Connell's life by people associated with Karl Rove. But in December 2008, before the trial began, Connell was killed in a plane crash outside Akron, Ohio.

There were problems in Florida, as well.

A study by the Quantitative Methods Research Team at Berkeley found that anomalies between Florida counties using touchscreen voting and other methods could not be explained statistically. Noting the higher-than-expected votes for Bush in three large Democratic counties - Miami-Dade, Broward and Palm Beach - Michael Hout, a Berkeley professor who did the study said there were strong suspicions of vote-rigging.

"No matter how many factors and variables we took into consideration, the significant correlation in the votes for President Bush and electronic voting cannot be explained," Hout said. "The study shows that a county's use of electronic voting resulted in a disproportionate increase in votes for President Bush. There is just a trivial probability of evidence like this appearing in a population where the true difference is zero - less than once in a thousand chances."

Don't Trust, Verify

Indeed the only way to 100% verify this election fraud would be through hand counts of ballots by precinct, matching those results to the reported totals. But as was mentioned earlier, a group in Pima County has been trying unsuccessfully to get access to ballots to conduct such a count for almost five years, since anomalies first surfaced in voting machines in 2006.

Is there a judge in Arizona likely to suddenly reverse past trends and allow access to conduct such a hand count of ballots 12 days before a national election? And if not, why not? Maybe someone needs to commission Anonymous to level the playing field because the courts are not going to do it. Welcome to Zimbabwe.

More than 100 million Americans will cast their ballots thinking their vote will be fairly counted. It should be. Yet the crooks know they can safely flip up to 10% of votes without consequence. Anything more than that is statistically suspect.

President Obama won by such a huge margin in 2008 that even with this anomaly built into the system, he cruised to victory. This year the election is much closer. Can American democracy afford yet another election crisis placing three of the four last national Presidential election results in question or worse: The outcome was stolen, the outcome a victim of election theft?

Don't Take Our Word

Use the spreadsheet above to do the maths in your own state, county or precinct. The results are compelling. Then demand that the Justice Department stop this insane view that results need to be reported by the television networks. Demand hand-counted ballots!

We use paper ballots in the U.K. and results do not even begin to trickle in until 3 am. The final outcome can take up to three days to finalize. But voters in Britain know the count is accurate because every ballot is transparently hand-counted. When I read an article that Serbia, Belarus and Kazakhstan were sending election monitors to watch the U.S. Election, I knew we'd jumped the shark.

We are already being victimized by vote fraud on a scale that, in another country, would lead to calls for international election monitors. It is time for Americans to stop being victims of ghosts in the machine.

Denis G. Campbell is the author of six books, including "Billionaire Boys Election Freak Show," "The Vagina Wars" and "Egypt Unsh@ckled." He is the editor of UK Progressive Magazine and provides commentary to the BBC, ITV, al-Jazeera English, CNN and MSNBC. You can follow him on Twitter via @UKProgressive.

Charley James is a longtime independent journalist who covers social justice, politics and economic issues. He's worked in print and broadcast media for national magazines, large newspapers and major market radio and television outlets. Follow him on Twitter @SuddenlyHomeles.