We’re rounding up all of the claims of voter fraud or other impropriety, along with their associated fact-checks, so that you can access them all in one place.
CLAIM: The number of votes cast in key battleground states exceeded the number of registered voters, thereby indicating fraud.
FACT: This claim was based on outdated data that hadn’t been updated since August 7th. Updated numbers reflect the following:
- Nevada reported 1,821,864 registered voters and 1,280,639 ballots cast for a 70.29% turnout;
- Pennsylvania reported 9,091,371 registered voters and 6,685,865 ballots cast for 73.54% turnout;
- Minnesota reported 3,589,785 registered voters and 3,278,548 ballots cast for a 91.32% turnout;
- Wisconsin reported 3,684,726 registered voters and 3,289,472 ballots cast for a 89.27% turnout;
- Michigan reported 8,128,430 registered voters and 5,519,348 ballots cast for a 67.9% turnout;
- Arizona reported 4,281,152 registered voters and 3,154,794 ballots casts for a 73.69% turnout; and
- Georgia reported 7,587,625 registered voters and 4,960,605 ballots cast for a 65.38% turnout.
CLAIM: A farmer found ballots on the side of the road that had been stolen from mailboxes in Arizona.
FACT: This is true. Eighteen sealed ballots were discovered by a farm worker in a suburb of Phoenix on Oct. 30th. Thankfully, replacement ballots were delivered prior to Election Day, ensuring that no Arizona voter would be disenfranchised. An investigation into who took the ballots and why is ongoing.
CLAIM: A software glitch in Michigan gave Biden the win in a county that Trump actually won. This software is also being used in other counties and states, and thus, similar glitches could be happening elsewhere.
FACT: This did happen, but it was a result of user error, not a software glitch. The mistake led to incorrect unofficial vote tallies being reported but was never reflected on official tabulator tapes.
CLAIM: After reviewing a provisional ballot report, Fulton County, Georgia noticed an error that had caused some provisionals not to be included in the results and others not to be scanned.
FACT: This is true. To fix the error, Fulton County recanvassed all of Friday’s ballots, ultimately helping Biden to widen his lead.
CLAIM: The Trump campaign identified at least 450,000 ballots that had a vote only for Biden and no other candidates, thereby suggesting fraud.
FACT: This is called “undervoting” and is a common phenomenon in all elections. In 2016, for example, in states with either a Senate race or a House race, there were 1.8 million more votes cast for president than for the Senate and 6.3 million more cast for president than for the House. And the inverse is often true, too. In 2016, 1.7 million people in 33 states and D.C. cast a ballot without voting in the presidential race. All that to say, undervoting is relatively common and is not evidence of voter fraud.
CLAIM: An Atlanta poll worker was caught on video destroying an absentee ballot.
FACT: This is false. The piece of paper that was discarded was a set of instructions, not a ballot.
CLAIM: The fact that counting carried on past election night is irregular and is prima facie evidence of fraud.
FACT: Counting was always projected to carry on past election night and has in years past.
For reference, FiveThirtyEight predicted that Arizona wouldn’t be finished counting until Thursday at the earliest. They estimated that a winner in Michigan would be declared by Friday. In Wisconsin, certain counties were set to release their results between 4 and 7 a.m. the morning after election night, leading most to conclude that they would call the race Wednesday morning. And Pennsylvania was always expected to take the longest, due in large part to the fact that military ballots postmarked by election day would be accepted until November 6th. As FiveThirtyEight said of Pennsylvania prior to the election, “It’ll be slow going.”
That being said, there is nothing about the timing of results that would suggest any kind of malfeasance, as they seemed to follow projected timelines. Not only that, they followed projected trends, too.
Depending upon when mail-in ballots (which favored Democrats) would be counted relative to in-person ballots (which favored Republicans), FiveThirtyEight was able to predict that states like Arizona, Texas, and Ohio would start out blue early on in the night and then turn more red over time. Conversely, due to the fact that in-person ballots would be counted prior to mail-in ballots in these states, FiveThirtyEight was able to predict that Michigan, Pennsylvania, and Wisconsin would start the night off red but turn more blue in the days that followed.
All that to say, there is no evidence of impropriety, as the events of Tuesday night and the days that followed were all anticipated by election experts.