The distorted claim that 86,391 voters in Maricopa ‘don’t appear to exist’

A GOP-led review of ballots cast in Maricopa County in the presidential election confirmed that Joe Biden won, but some supporters of Donald Trump have used the outcome of the ballot review to portray thousands of votes as suspicious.

"AZ audit could not find the identity of 86,391 voters – they don’t appear to exist," said a headline by Unite America First, a website that posts news critical of Democrats and favorable toward Republicans.

The post was flagged as part of Facebook’s efforts to combat false news and misinformation on its News Feed. (Read more about our partnership with Facebook.)

The Facebook post distorts what the ballot review had to say about those 86,391 voters.

Arizona state Senate Republicans ordered a review of 2 million ballots in Maricopa County. The review found that Biden beat Trump in Maricopa County by about 45,000 votes. That number was virtually the same as the county’s official canvass.

The Unite America First article refers to one of the findings in the report related to a database used by the Cyber Ninjas, the contractors hired to review the ballots. The Cyber Ninjas checked the file of voters against a commercial database called Melissa.

The Cyber Ninjas found that 86,391 voters were not found in the Melissa database.

But the report did not state that the voters don’t exist. In fact, it said: "it is expected that most if not all of these individuals are in fact real people with a limited public record and commercial presence." 

In an earlier section of the report, the Cyber Ninjas acknowledged that the database was incomplete, writing "some error is expected within these results."

Greg Brown, vice president of global marketing at Melissa, told PolitiFact that the website uses many sources, including names from newspaper and magazine subscriptions, credit cards, and mortuary records to compile its Personator data tool. 

"Being absent from the Personator data tool is not an indicator that an individual does not exist," Brown said.  

Brown estimated that their database covers about 80% of U.S. adults. Some people may have opted out of such services, such as law enforcement officials, or just kept a low profile.

Election experts say that official government voter registration databases offer a better set of data for verifying voters  than commercial databases. 

Tammy Patrick, an expert on elections at the Democracy Fund and a former Maricopa County elections official, said some individuals could have many entries in a commercial database while others have none. 

 "Some individuals may use names or information that does not align with their official (name) — a subscription to Cat Fancy for Princess Kitty Jefferson rather than the cat’s owner," Patrick said. "Commercial data is not government data."

Arizona election officials take steps to update voter registration information by checking it against government databases. Arizona is a member of ERIC, a partnership of states that allows election officials to access data showing voters who have moved within their state or out of state, have died or have duplicate registrations. 

Featured Fact-check

Donald Trump

stated on November 22, 2020 in a tweet

“In certain swing states, there were more votes than people who voted, and in big numbers.”

Benny White, a Republican and longtime volunteer data analyst for the state Republican Party, previously told us that commercial databases are a flawed source for voter information.

"You have to use the official voter registration records maintained by county officials and the Secretary of State," White said. "I have worked with the RNC database for a number of years, and that database, which they spend millions of dollars trying to keep current and complete, is almost unusable because it is not current and not complete."

A Facebook post said the ballot review in Maricopa County found 86,391 voters who "don’t appear to exist."

The GOP-orchestrated review of ballots in Maricopa County cast in the 2020 presidential election found that 86,391 voters were not found in a commercial database known as Melissa. 

But the ballot review report did not declare that these voters don’t exist. The report said, "It is expected that most if not all of these individuals are in fact real people with a limited public record and commercial presence." An official for Melissa estimated that their database covers 80% of adults and said that being absent from their data tool doesn’t mean someone doesn’t exist.

We rate this statement False. 

RELATED: Arizona "could decertify" the 2020 presidential election

RELATED: Trump falsely describes Arizona audit findings

RELATED: No evidence Maricopa County audit found 17,000 "duplicate" votes

RELATED: No evidence for claim that Maricopa County ‘purged machine records before audit’