“There were a couple different mathematical methods used," Swope said. "One of them said right from the beginning that we should sample 600 ballots, and based on the amount of time we only did about 250.”

The city’s goal of at least 90 percent confidence in the accuracy of the ballots was met, according to Swope, who said he was happy with the results.

“Being a pilot, we went ahead and stopped where we were with our plan," Swope said. "It was definitely a great learning experience, and if it weren’t a pilot situation, we would have continued with additional sampling.”

The city clerk previously explained the separate processes used for absentee ballots and ballots cast at voting booths. Absentee ballots were compared with the certified cast-vote record, which is not possible for in-person ballots whose orders are randomized in machines.

“Unfortunately, today’s realities require election administrators to be even more vigilant and I’m delighted to pioneer a new tool that may be helpful in our ongoing efforts to protect elections," Swope said in a statement. "Pairing our election security measures already in use with a robust post-election audit to check election results seems like a common-sense step forward.”