Election results may take longer than usual to be reported Tuesday night, and part of that could be deliberate.
“There is this intent to muck up the election and set it up for failure,” said Natalie Tennant from the Brennan Center for Justice at New York University law school during a webinar Wednesday discussing concerns about the Nov. 8 midterm voting.
Lisa Danetz, an elections advisor for the center, discussed what she said were loosely organized online efforts to get voters to slow down the counting of ballots on election night. Delay is the goal of people who want to increase skepticism about elections, which makes it easier to overturn results that they don’t like, Danetz said.
“The longer a count takes, the less people trust the outcome. Especially in larger communities, intentional overvoting could undermine confidence in election outcomes,” she said. She cautioned people to not get alarmed if results don’t get reported by media or online as quickly as in past elections.
An overvote is when a person votes too many times for a race, such as voting for a candidate and then also marking a write-in for the same person in the same position, or voting for three persons in a state representative race with only two seats.
Danetz described how this can small step can mire counting votes on election night.
“They are advising people to intentionally overvote … which makes the machine eject the ballot, (causing) a manual review and hand count,” she said. “This can cause a huge slowdown of the count, like a denial-of-service attack on a computer, but in a physical way.”
The Accuvote machines that optically scan New Hampshire’s ballots in most communities are programmed to eject ballots with even a single overvote. Those ballots must then be hand counted by volunteers or local election officials after the polls close, which can add hours to the process of declaring winners.
Liz Wester, director of a group called New Hampshire Voter Empowerment Task Force, said the plan was part of attempts to discredit the Accuvote machines, which have been used in all New Hampshire elections since the late 1980s.
“We’ve seen it pushed through by folks spreading misinformation about ballot-counting devices” on various social media sites, she said. The efforts are mostly seen in the state’s southern tier and the Seacoast, she said, and vary among communities, usually because of the action of one or a small number of individuals in that community.
State law requires voting to be done on paper ballots. Most New Hampshire communities with fewer than 1,000 people count such ballots by hand because of the expense of optical scanners, but virtually all larger communities use the machines, both to save time on election night and because it is difficult to enroll enough volunteers or staffers to count large numbers of ballots.
Studies consistently find that optical scanners are more accurate than counting by hand.
A Monitor analysis of recounts of three 2020 races confirmed this. The three races each involved multi-town state representative seats in which some towns counted the ballots by hand and some towns counted the same ballots using Accuvote machines. In all three cases, the recount – done days later, by hand in state offices – found more errors in the hand-count towns than in the machine-count towns, although none had more than a few errors.
Concern about the state’s voting machines has increased since a 2020 election in the town of Windham in which several hundred improperly folded ballots were misread by the town’s four Accuvote machines. An audit found that some 300 votes were denied to the four Republicans in the state representative race – all of whom had won anyway – and about 100 votes were added to a Democratic challenger. That political split in the errors fueled conspiracy theories about the machines.
This year’s town meeting season saw requests to ban ballot-counting machines put on the warrant of about a dozen towns – all of them failed. A bill to require hand-counting of all elections was killed by the legislature this year.
New Hampshire is looking to replace the Accuvote machines, which are so old that parts are no longer available. The process has been going on for at least three years.
Noting the “may” and “could be” in paragraph one above: Now that it’s over, is there any evidence that this actually happened to any significant extent?