Mail-In Ballot Fraud Study Finds Trump ‘Almost Certainly’ Won in 2020

Empty envelopes of opened vote-by-mail ballots for the presidential primary are stacked on a table at King County Elections in Renton, Wash., on March 10, 2020. Jason Redmond/AFP via Getty Images

A new study examining the likely impact that fraudulent mail-in ballots had in the 2020 election concludes that the outcome would “almost certainly” have been different without the massive expansion of voting by mail.

The Heartland Institute study tried to gauge the probable impact that fraudulent mail-in ballots cast for both then-candidate Joe Biden and his opponent, President Donald Trump, would have had on the overall 2020 election results.

The study was based on data obtained from a Heartland/Rasmussen survey in December that revealed that roughly one in five mail-in voters admitted to potentially fraudulent actions in the presidential election.

After the researchers carried out additional analyses of the data, they concluded that mail-in ballot fraud “significantly” impacted the 2020 presidential election.

They also found that, absent the huge expansion of mail-in ballots during the pandemic, which was often done without legislative approval, President Trump would most likely have won.

“Had the 2020 election been conducted like every national election has been over the past two centuries, wherein the vast majority of voters cast ballots in-person rather than by mail, Donald Trump would have almost certainly been re-elected,” the report’s authors wrote.

Over 43 percent of 2020 votes were cast by mail, the highest percentage in U.S. history.

‘Biggest Story of the Year’

The new study examined raw data from the December survey carried out jointly between Heartland Institute and Rasmussen Reports, which tried to assess the level of fraudulent voting that took place in 2020.

The December survey, which President Trump called “the biggest story of the year,” suggested that roughly 20 percent of mail-in voters engaged in at least one potentially fraudulent action in the 2020 election, such as voting in a state where they’re no longer permanent residents.

Former President Donald Trump speaks during a press conference held at Mar-a-Lago in Palm Beach, Fla., on Feb. 8, 2024. Joe Raedle/Getty Images

In the new study, Heartland analysts say that, after reviewing the raw survey data, subjecting it to additional statistical treatment and more thorough analysis, they now believe they can conclude that 28.2 percent of respondents who voted by mail committed at least one type of behavior that is “under most circumstances, illegal” and so potentially amounts to voter fraud.

“This means that more than one-in-four ballots cast by mail in 2020 were likely cast fraudulently, and thus should not have been counted,” the researchers wrote.

A Heartland Institute research editor and research fellow who was involved in the study explained to The Epoch Times in a telephone interview that there are narrow exceptions where a surveyed behavior may be legal, like filling out a mail-in ballot on behalf of another voter if that person is blind, illiterate, or disabled, and requests assistance.

However, the research fellow, Jack McPherrin, said such cases were within the margin of error and not statistically significant.

What Are the Implications?

In addition to reassessing the likely overall degree of fraudulent mail-in ballots in the 2020 election, Heartland analysts calculated the potential impact that fraudulent mail-in ballots might have produced in the six key swing states that President Trump officially lost.

This, then, was used to determine the impact of potentially fraudulent mail-in ballots on the overall 2020 election result.

First, the researchers analyzed the electoral results for the six swing states—Arizona, Georgia, Michigan, Nevada, Pennsylvania, and Wisconsin—under the 28.2 percent fraudulent mail-in ballot scenario that they estimated based on the raw survey data.

Then they calculated the electoral results in the six states under the different scenarios, each with a lower assumed percentage of fraudulent ballots, ranging from 28.2 percent all the way down to 1 percent.

For each of the 29 scenarios that they assesses, the researchers calculated the estimated number of fraudulent ballots, which were then subtracted from overall 2020 vote totals to generate a new estimate for vote totals.

Overall, of the 29 different scenarios presented in the study, the researchers concluded that President Trump would have won the 2020 election in all but three.

Specifically, they calculated that the only scenarios that would affirm the official 2020 election result, namely that candidate Biden won, were mail-in ballot fraud levels between 1 and 3 percent of ballots cast.

Mail-in ballot fraud rates higher than 3 percent would, according to the study, mean more fraudulent Biden votes that should be subtracted from the total, putting President Trump ahead.

For example, the adjustment to the vote tallies under fraud percentage rates between 13 and 6 percent would mean President Trump would have won Arizona, Georgia, Pennsylvania, and Wisconsin, though he would have still lost in Michigan and Nevada.

Under such a scenario, President Trump would have won 289 Electoral College votes compared to candidate Biden’s 249.

In scenarios of 5–4 percent fraud, each candidate would have received 269 Electoral College votes, but President Trump would likely still have won because Republicans controlled more state delegations and, under a tie scenario, Congress would have voted based on the number of delegates.

However, the researchers expressed confidence in their overall assessment that the level of mail-in ballot fraud was over 25 percent, indicative of an actual Trump win.

“We have no reason to believe that our survey overstated voter fraud by more than 25 percentage points, and thus, we must conclude that the best available evidence suggests that mail-in ballot fraud significantly impacted the 2020 presidential election, in favor of Joe Biden,” the paper’s authors wrote.

This combination of pictures created on October 22, 2020 shows President Donald Trump, left, and Democratic presidential candidate Joe Biden during the final presidential debate at Belmont University in Nashville, Tenn., on Oct. 22, 2020. Brendan Smialowski and Jim Watson/AFP via Getty Images

Survey Criticism

Jim Womack, president of the North Carolina Election Integrity Team, told The Epoch Times in an earlier interview and in additional written comments in response to the new study, that he believes the survey questions were flawed and make the survey statistically meaningless, though not without value.

“We know there was fraud in the 2020 election, but you can’t conclude that it was 20 percent or 10 percent or even 5 percent based on the survey because the questions that could lead to such conclusions were unclear,” Mr. Womack said.

However, he said that the survey questions on which Heartland based its research were unclear. He argued that the questions comingled legal and illegal activity and that this made it impossible to conclude specific percentages of mail-in ballot fraud with certainty.

For instance, Mr. Womack pointed out that it’s legal and permissible in all states for people who by reason of blindness, disability, or illiteracy request or require assistance in filling out mail-in ballots to get such assistance.

However, the wording of one of the survey questions—“During the 2020 election, did you fill out a ballot, in part or in full, on behalf of a friend or family member, such as a spouse or child?”—did not differentiate between legal and illegal forms of filling out a mail ballot on behalf of someone.

Therefore, 21 percent of people responding “yes” to this question does not necessarily mean that this percentage of people actually committed voter fraud, Mr. Womack argued.

Mr. Womack also said that another survey question–“During the 2020 election, did you cast a mail-in ballot in a state where you were no longer a permanent resident?”—to which 17 percent replied yes—also does not support the conclusion that all such cases were illegal. That’s because, as Mr. Womack pointed out, federal and state laws allow some voters (such as UOCAVA registered citizens) to cast a ballot in a state where they are no longer permanent residents under certain circumstances.

“We'd need to dive deeper into these responses to determine if these were fraudulent or not,” Mr. Womack said.

Regardless, he praised the Heartland Institute for engaging with the topic of mail-in ballot fraud and raising public awareness about what he said is an important problem.

Response to Criticism

When asked to comment on Mr. Womack’s objections, Mr. McPherrin, of the Heartland Institute, told The Epoch Times that he stands by the findings.

For instance, Mr. McPherrin acknowledged that it’s legal for people who are blind, disabled, or illiterate to get help from someone in filling out a ballot.

However, he argued that the number of such individuals responding to the Heartland/Rasmussen survey (which was based on a representative sample of 1,085 likely voters) would likely have been tiny.

“It would be difficult to imagine that dozens of blind people or those that are illiterate or disabled are answering this poll,” he said, adding that the presumably tiny fraction of survey respondents who fall into this category would be statistically insignificant and not impact the overall survey results.

But even if that particular question is left out due to concerns about its clarity, the percentage of people who admitted to potentially fraudulent voter activity would still be about one in five, he said.

Mr. McPherrin said he and his team have received and reviewed Mr. Womack’s criticism and they believe the points he makes have some validity but not enough to affect their findings in a meaningful way.

He maintains the study clearly shows that if the 2020 election had been as fair and secure as prior elections, President Trump would “almost certainly” have been re-elected to a second term.

Mr. Womack continues to stand by his criticism of the survey question design, providing The Epoch Times with a written statement on Feb. 8 that calls the survey “very poorly constructed, failing to capture even a single instance of probable voter fraud.”

He argued that the survey questions were “vague and ambiguous, commingling permissible with impermissible behaviors, thus diminishing the quality and usefulness of responses.”

Further, Mr. Womack argued that propagating the contents of the survey does more harm than good and potentially undermines the work and reputation of “legitimate election integrity organizations like EIN,” referring to the Election Integrity Network, a project of the Conservative Partnership Institute.

Meanwhile, the authors of the Heartland study call for state legislatures to do all in their power to ensure the 2024 presidential election is as secure as possible, mostly by severely limiting mail-in voting and adopting other commonsense policies to prevent mail-in voter fraud.