New Analysis Suggests Link Between Race And Odd Vote Count In Georgia’s 2018 Midterms

People cast their ballots ahead of the Nov. 6, general election at Jim Miller Park, in Marietta, Ga.

Mike Stewart / Associated Press

A new data analysis suggests the sharp drop in votes in the lieutenant governor’s election last year may be connected to the race of voters.

The finding raised more questions about the results of the down-ballot contest in which Republican Geoff Duncan handily defeated Democrat Sarah Riggs Amico.

Compared to the governor’s race between Democrat Stacey Abrams and Republican Brian Kemp, about 160,000 fewer votes were cast in the lieutenant governor’s contest, while just 82,000 fewer votes were cast in the attorney general’s contest.



There’s no clear evidence to suggest a problem with the vote count, but it’s virtually impossible to prove that vote totals in Georgia are accurate, because the state’s voting system lacks a paper trail that can be audited.

Chris Brill has been analyzing the results for the Coalition for Good Governance, an organization that advocates for paper ballots, which is challenging the results of the lieutenant governor’s election.

A judge threw out the challenge last month, but the Coalition says it plans to appeal. Amico is not challenging the election results.

Brill works for Target Smart, a data firm he said usually helps liberal candidates.

Brill said he looked at the state’s voting precincts, and compared the vote drop off in each precinct with other characteristics of the precinct, like the racial makeup of voters, and the percentage of first-time voters.

“When you look at the 2,634 or so statewide precincts,” wrote Brill in a report, “the under vote rate for lieutenant governor does tend to increases as the electorate becomes more African American.”

For example, Brill found that when 50-75 percent of voters in a precinct were African-American, the under vote rate was 5.5 percent. In Georgia’s precincts where 0-25 percent of voters were African-American, the under vote rate was 3.5 percent.

Brill said he did not find a correlation between the vote drop off, and the percentage of first-time voters in a precinct.

Statistics experts say racial demographics aside, the big under-vote in the lieutenant governor’s contest alone suggests something went wrong with the voting system. But without a paper trail, or other hard evidence, that’s hard to prove.

After a judge rejected the election challenge last month, Ed Lindsey, a lawyer for Republican Lt. Governor Duncan said “The people of Georgia should be able to rest assured on the evidence…that the election was accurate not only in this race, but in all the other races cast on November 6th.”