Overview

The intent of conducting pilots in municipal elections is to place the new equipment into real election conditions to evaluate the equipment, the processes and importantly the people interacting with the new verifiable paper trail voting system.

As the office of Secretary of State has been working with the new system, we have seen that it isn’t simply changing one type of mechanism, a Direct Recording Electronic (DRE) voting, machine for another machine, the Ballot Marking Device (BMD). The new paper ballot voting system fundamentally changes the way that the state of Georgia has conducted voting for nearly two decades.

With that in mind, the implementation team understood that piloting the new system in real world conditions, with the introduction of poll workers, polling places, and most importantly voters, would be vital to bring potential issues to the forefront.

With that in mind, we picked nine counties to do the initial pilots:

Bacon, Bartow, Carroll, Catoosa, Decatur, Evans, Lowndes, Paulding, and Treutlen. All elections officials were brought in for three days of training to our Center For Elections with Dominion and SOS Staff. Bacon, Evans, and Treutlen ended up having no municipal elections, so the remaining six counties conducted their November elections on the new system. Cobb County volunteered to conduct a pilot election with the new system, except using hand-marked paper ballots instead of ballot marking devices. That pilot still included the Poll Pad, Polling Place Scanner, and new election management system. The Secretary of State’s office is extremely grateful to the counties who volunteered to be the first to test this new system. Their work and lessons learned is irreplaceably valuable as the State continues this implementation.

Initial Findings

There were 27,482 votes cast in the Municipal (and County) elections in the six pilot counties. That means there were 27,482 interactions by voters with the new voting machines that were completed successfully.

o overall in service there were 576 BMDs and 83 Polling Place Scanners deployed.

As you will see in the subsequent appendices there were 45 total “incidents” reported. Some of those were not problems, they were reports of the system functioning properly, but it being new, and wanting to be thorough, the poll workers called them into the technicians from Dominion.

By that measure, we had 45 incidents out of 27,482 votes or an incident rate of 0.164%. There were 4 touchscreens out of 576 (0.69%) and 1 scanner out of 83 (1.2%) were taken from service out of an abundance of caution. Further, nearly all issues were caused by human error or interaction which can be mitigated through training or identified through testing.

The report is available online at: https://sos.ga.gov/admin/uploads/Executive_Summary_Initial_Findings_Pilots_11-14-19.pdf