Half the country will say it was stolen

Something we can do now to offer confidence in next month's election result

No matter who is announced as our new President on the morning of November 6, half the country will suspect there was cheating. 

One thing that Democrats and Republicans and every mainstream news source can agree on — don’t peek into the kitchen where the sausage is made. No one in power wants to explore the black boxes that count our votes. The newspapers and broadcast stations all agree — election theft is not a thing. It DOES NOT happen.  It CANNOT happen. Not in America. Only a tinfoil hat conspiracy theorist could even imagine such a thing. And don’t you dare talk about it. 

Since January 6, 2001, it’s not just gauche to ask questions about how our votes are counted — it can land you in jail. Protesters were mislabeled
“Insurrectionists”, denied habeas corpus, and jailed without trial for over a year.

But in the meantime, trust in America’s electoral system is at an all-time low (Pew research, Christian Science Monitor). It turns out  that there is a well-documented history of election theft in America, and the game took a quantum leap into the 21st century when computerized voting was introduced, beginning 2002. 

Cheating in elections is as old as voting and as American as Billy the Kid.  In 1792, the first scandal rocked a Congressional vote in Georgia, and as a result, no one was sent to Congress from that district. Andrew Jackson won the popular vote in 1824, but he was cheated out of the electoral college vote by a back-room deal between John Quincy Adams and Henry Clay. Tammany Hall has a history going back at least 150 years of fixing New York City elections on behalf of the Democratic machine.  In the 1960s, Mayor Daley was famous for getting out more votes than there were voters in Chicago. 

But election theft has been an unmentionable topic since at least the 1980s, when James and Kenneth Collier wrote about their experience in a book called VoteScam. They witnessed ballot box fraud first-hand, but as they brought their story to one newspaper after another, they repeatedly found that the editors refused to get involved.

The Help America Vote Act was passed after the 2000 “hanging chad” debacle in Florida, and ever since, our votes have been counted in the pitch black of cyberspace. US Courts, in their wisdom, have consistently ruled that the software that counts votes is a trade secret belonging to the companies that produce voting machines, and not even the counties that buy the equipment are entitled to inspect the software.

Leading up to the 2004 election, NYTimes columnist Paul Krugman wrote a series of columns [one, two, three, four] detailing the vulnerabilities of the electronic system, and the insanity of trusting our democracy to three software companies, all with close ties to the Republican right. Then the 2004 election was stolen in Ohio when a White House computer guru created a “backup” for the official count by the Secretary of State in Columbus. (I describe details here.)

Then Krugman was “on leave” from his post at the Times for three months, and he never broached the subject again.

The Carter Center monitors elections in many countries around the world, but when Jimmy Carter was asked why he doesn’t check on American elections, he replied it was not possible here — public access is unavailable. 

Jonathan Simon coined the term “red shift” when, year after year, there was a disparity between exit polls and the announced vote that almost always seemed to favor Republicans. Still, the Democrats kept mum — Not the DNC nor the Democratic Congressional leadership nor the DC liberal think tanks nor the Liberal Press were willing to talk about this issue. (This has been my personal experience as an election integrity statistician, shared by many others I know who are passionate about the issue.) Why don’t the Democrats cry “foul” when Republicans steal votes wholesale? Maybe because the Democrats have a long tradition of ballot harvesting and ballot box stuffing at the retail level.

This is a problem with a solution

For a long while, exit polls were our best check on the integrity of American elections. Edison Research was hired by a consortium of print and broadcast media to poll thousands of representative precincts around the country. As people exit the polling station, a pollster hands them a clipboard and asks them to check a box indicating for whom they voted. This has been a gold standard for checking up on election integrity.

But when a “redshift” began opening up, beginning in 2004, Edison was cited as evidence of stolen elections. Warren Mitofsky, long time head of Edison Research, saw that his business model was threatened. The establishment media hired Edison to predict the winner a few hours in advance, not to second guess America’s democracy. 

In a bizarre role reversal, Mitofsky argued vigorously that his polls were all wrong, as Steve Freeman, a professor at University of Pennsylvania turned election integrity activist, cited statistical indications that the polls were accurate, and it was the official count that was questionable. After this, Edison and other polling companies began adjusting their methodology, progressively with each new election, because they wanted to “get it right”. This meant sampling more Republicans, fewer Democrats, in the name of “accuracy” in their predictions.

Warren Mitofsky passed away in 2006, but his legacy lives on. We can no longer count on Edison or any of the news media to conduct an honest exit poll, but We the People can do it ourselves. In fact, the Trump Campaign and the Harris Campaign might both see that they have an interest in teaming up to fund an independent exit poll.  

Here’s how it will work. A polling company quietly chooses about 2,000 precincts around the country, maybe concentrated in swing states, maybe chosen to be a representative sampling of urban and rural, black and white and hispanic and Asian, male and female, old and young, Republican and Democratic. They keep their choices secret, while sending a poll organizer to each site a week or two in advance, hiring pollsters from a classified ad, training them to be neutral, polite, inviting, and even-handed in approaching each person who emerges from the polling station. 

When the day is over, we will have “within-precinct differences”.  For each of these 2,000 precincts, we will have an official count and we will have a polling result. Do the polling results look like the announced result for that same precinct? Is there a consistent difference across the country, either a “redshift” or a “blueshift”? 

This is a methodology used by the Carter Center and our own State Department abroad, and it is statistically ultra-reliable. This method has a great deal of historic credibility behind it. 

We might ask both candidates to agree in advance to trust a neutral polling company, and to pool their resources to pay the company jointly. It’s not expensive. Total spent on this campaign is north of $8 billion and counting; this exit poll will give America confidence in the result for less than $10 million.

Footnote

But are Republicans or Democrats consistently more likely to take the proffered clipboard and submit to the exit poll? Is there a possible bias there?

For properly trained pollsters, this has not been a problem in the past, time and time again, when exit polls predicted election outcomes with a 1 or 2% margin of error. But we can do even better. Jonathan Simon, mentioned above, invented the following two check in 2006 and 2010.

Choose precincts where there is a lopsided local race, handicapped at 70-30 before election day, alongside a close Presidential race. The cheaters don’t try to rig these lopsided elections — they can’t steal enough votes to make a difference, so they don’t try. But these lopsided races can serve as a control group for the close race. 

If in a given precinct, the same voters, filling out the same clipboard, are a fair sample of the 70-30 race, but there is a shift in the 50-50 race, this is a red flag that the official count has been corrupted. If, on the other hand, we see a consistent bias in both the 70-30 race and the 50-50 race, it’s an indication that the exit poll, not the official count, was in error.

One more check we can do is to choose some rural precincts that are still using hand-counted paper ballots. If in the hand-counted precincts the polls are accurate, but in places where there is computerized voting the numbers are skewed, we suspect trouble.

We can do this

The time is now. The two parties can come together. We the people can demand it. A truly independent poll to give us confidence that we can trust the results we see Wednesday morning, and it will cost less than a nickel per voter.