PHOENIX β Democrat Katie Hobbs was confirmed Saturday as winner of the governorβs race as a judge rejected claims by Republican Kari Lake that there was misconduct in the election.
Maricopa County Superior Court Judge Peter Thompson issued a detailed report saying testimony on behalf of Lake, who lost by 17,117 votes, provided no evidence she should have been declared the winner. He said some of her theories about what went wrong, and why, are not backed by facts.
βA court setting such a margin aside, as far as the court is able to determine, has never been done in the history of the United States,ββ the judge said, noting Hobbsβ margin of victory was outside the margin for an automatic recount under state law.
Hobbs, the secretary of state, will be inaugurated as governor on Jan. 2.
Thompson said he is aware that tabulators at some of Maricopa Countyβs 223 vote centers were unable to read the ballots printed out by on-site printers. That, in turn, led to long lines at some sites.
βThis court acknowledges the anger and frustration of voters who were subjected to inconvenience and confusion at voter centers as technical problems arose during the 2022 general election,ββ he wrote in his 10-page ruling.
βBut this courtβs duty is not solely to incline an ear to public outcry,ββ the judge continued. βIt is to subject plaintiffβs claims and defendantsβ actions (which included both Hobbs and Maricopa County) to the light of the courtroom and scrutiny of the law.ββ
His review of the evidence shows Lake did not prove her case, he wrote.
Lake says she will appeal
Lake responded in a Twitter post. βMy election case provided the world with evidence that proves our elections are run outside the law,ββ she wrote.
She said she will appeal βfor the sake of restoring faith and honesty in our elections.ββ
To do so, she will have to convince the Arizona Supreme Court that Thompsonβs detailed findings were flawed.
Lake relied heavily on the testimony of Clay Parikh, who opined that a problem with some ballot printers β spitting out undersized images β must have been done intentionally.
The judge said Parikh has βan impressive technical background as a cybersecurity expert for Northrup Grumman.ββ But he said Parikhβs conclusion is undermined by the actual evidence.
βIf the ballot definitions were changed, it stands to reason that every ballot for that particular definition printed on every machine so affected would be printed incorrectly,ββ and would put out a 19-inch ballot on 20-inch ballot paper, Thompson said. He said that wasnβt the evidence.
The judge also said David Betencourt, another witness called by Lake, undermined Parikhβs theory.
Betencourt, a temporary technical worker for the county, testified that problems with printers could be solved by several means, including taking out and shaking the printer cartridge, cleaning the corona wire, letting the printers arm up, and adjusting settings on the printer. Apart from the last, none of those suggest an intent to produce an image unreadable to the tabulators at each polling place, the judge said.
In fact, Thompson said, Betencourt not only said he had no knowledge of anyone engaged in intentional misconduct but that the techs he worked with βdiligently and expeditiously trouble-shot each problem as they arose, and they did so in a frenetic Election Day environment.ββ
No evidence anyoneβs vote wasnβt counted
The judge also said there was no evidence anyoneβs vote was not counted, even if it was cast on one of the 19-inch ballots. He said any undersize ballot that could not be immediately read by the tabulator could be placed by the voter in a sealed box to be taken to a central location at the end of the day where it would be βduplicated by a bipartisan board onto a readable ballot, and β in the final analysis β counted.ββ
The judge said the testimony of Richard Baris was no more helpful to Lakeβs claim the election was rigged.
Baris, director of Big Data Poll, testified that the printer failures on Election Day disenfranchised a number of voters who, he opined, simply gave up. His basis for that opinion, he said, was that he had contacted a certain number of voters prior to the election but many did not check back in with him afterwards for an exit poll.
But Thompson, even giving credit to Barisβ polling techniques, said nothing in his testimony showed an actual effect on the election.
That is critical: Thompson said anyone seeking to overturn an election must show not only that there was an intentional act, but that it was committed by someone involved in the election, that it was intended to change the result, and that it did alter the results.
βMr. Baris admitted at trial that βnobody can give a specific numberβ of voters who were put off from voting on Election Day,ββ the judge said.
Baris suggested the number of ballots cast was anywhere from 25,000 to 40,000 less than would have been expected without those printer and tabulator problems. But Thompson said even Baris acknowledged that could have created anywhere from a 4,000-vote margin for Lake β to adding 2,000 votes for Hobbs.
Thompson said he is not saying statistical evidence can never be used to demonstrate a direct effect on the outcome of an election. But he said βa statistical analysis that shows the current winner had a good chance of winning anyway is decidedly insufficient.ββ
Moreover, he said, βNo election in Arizona has ever been set aside, no result modified, because of a statistical estimate.β
He said it would be one thing if this had been a case where someone proved malfeasance to the point where the effect could be quantified. But the judge called it a βquantum leapββ for a court to set aside an election where the effects are unknown.
Elections decided by votes, not polls
Nor was Thompson willing to accept Barisβ estimates. βElection contests are decided by votes, not by polling responses,ββ he wrote.
βThis court has found no authority suggesting that exit polling ought to be used in this manner,ββ the judge continued. βGiven that exit polling is done after a vote has been cast, the weight of authority seems to be contrary to this proposition.ββ
Thompson also said Lakeβs claims got no help from testimony by Mark Sonnenklar, a roving attorney with the Republican National Committee, who went to eight polling sites on Election Day.
He testified about the failure of tabulators at multiple locations, his own personal estimate on the rate of failure, the efforts of county election workers to fix the problems, and the frustration and anger of voters who had to wait in longer lines due to the failures. Sonnenklar called it βcommon senseββ that such widespread failure must have been the result of intentional conduct.
βBut this intuition does not square with Mr. Sonnenklarβs own observations,ββ the judge said. βFor one thing, county T-Techs (temporary technical workers) being sent to troubleshoot and fix the issues with tabulators are not consistent with a scheme by a person or persons to alter the result of an election.β
He also said Sonnenklar admitted he did not observe anything supporting his intuition that someone engaged in intentional misconduct. Sonnenklar also admitted he had no personal knowledge of any voter being turned away from polls as a result of printer failures, Thompson said.
The judge also brushed aside testimony from Heather Honey, a supply chain auditor and consultant, who claimed the county violated chain-of-custody rules for ballots. He said that was based on the county failing to give her copies of βdelivery receipt forms.ββ
But Thompson said Honey testified under cross examination that she knows these forms do exist. What that leaves, the judge said, is a claim the county did no comply with law for producing records β but no basis for a claim of violations of election laws.
Finally, the judge noted there was testimony the county did not know on election night how many ballots it had received, a fact that Lakeβs attorney Kurt Olsen said βdoes not make sense.ββ That formed the basis of Lakeβs claim that perhaps 25,000 extra ballots had been injected into the system after the vote.
County officials said the numbers released that night were simply estimates.
βIt was not Maricopa Countyβs burden to establish that its process or procedure was reasonable, or that it had an accurate unofficial count on election night,ββ Thompson wrote. βEven if the county did bear that burden, failing to carry it would not be enough to set aside election returns.ββ