10

Loading video...

Lars Hanson

By Lars Hanson

Associate News Editor

0

Share

Newsweek is a Trust Project member

See more of our trusted coverage when you search.Prefer Newsweek on Googleto see more of our trusted coverage when you search.


Alaska counts its ballots the way it does most things: slowly, across enormous distances, one envelope at a time. In the race for the state's U.S. Senate seat, Democrat Mary Peltola is now just 157 votes short of an outright majority.

The Division of Elections' unofficial summary report, released Saturday afternoon, shows Democrat Mary Peltola with 140,637 first-choice votes, or 49.94 percent, to Republican Senator Dan Sullivan's 129,424, or 45.96 percent. Republican Gerald L. Heikes drew 6,002 votes.

Daniel J. Sullivan Jr., a retired Petersburg teacher who shares the senator's name, drew 4,652, and write-ins accounted for 872. As of 8:29 a.m. Eastern on Sunday, The Associated Press estimated that 97 percent of the vote had been counted and had not called the race.

Read More on News

The Division's first-choice count includes the 872 unresolved write-ins, which its published rules treat as a single group in the first round.

That number matters because of how Alaska settles close contests. A candidate who wins more than half of first-choice votes is elected outright. If no one does, the Division of Elections begins a ranked-choice tabulation on November 18, eliminating the last-place candidate and moving those ballots to voters' next choices.

A candidate who passes 50 percent in any round wins, though the Division continues counting until two candidates remain.

The Arithmetic of a Majority

The 157-vote gap will not hold still. Each newly counted ballot raises the bar for a majority, so Peltola cannot simply collect a handful of votes; she has to win a share of everything that remains.

AP does not publish a count of outstanding ballots. Its 97 percent is a rounded estimate of the expected vote. Assuming AP rounds to the nearest point, treating that figure as anything from 96.5 to just under 97.5 percent and applying it to the Division’s total, Newsweek’s analysis estimates that several thousand Senate votes remain. To clear 50 percent on first choices alone, Peltola would need roughly 52 percent of what is left.

The estimate rests on assumptions: that AP's figure applies to the Senate contest, and that every remaining ballot carries a valid Senate vote, though some voters skip races. The pool is not yet sealed, either. Ballots mailed from overseas can still arrive through November 18.

Alaska Democratic Senate candidate Mary Peltola walks up to speak at an election night event Tuesday, Nov. 3, 2026, in Anchorage, Alaska.

...

Sullivan's Two Roads

For the senator, the first-choice count offers little room. He trails by 11,213 votes. Under AP's estimate, Newsweek's analysis found that deficit is larger than the number of Senate votes that appear to remain. Even if every outstanding vote broke his way, Peltola would still hold more first choices, so long as the estimates hold.

His other road runs through ranked choice, and it opens only if Peltola finishes below a majority. Then the race turns on something the public count does not show: how voters ordered the candidates behind their first pick.

Newsweek independently ran the Division's ballot-level cast vote record**,** a record of how each scanned ballot ranked the candidates, which covers 282,698 ballots, through RCTab 2.0.1, open-source software built to tabulate ranked-choice elections, with settings matching the Division’s published rules for write-ins, overvotes and skipped rankings.

Of those ballots, 1,059 ranked no one in the Senate race. Another 598 were inactive from the first round: 428 were overvotes, with more than one candidate marked at the same ranking, and 170 were set aside under the state's rules on skipped rankings**.** Alaska's tabulation removes one candidate per round. Each ballot of the eliminated candidate moves to that voter's next-ranked candidate still in the race; a ballot with no such ranking goes inactive and stops counting.

In the tabulation, write-ins fall first. Of those 870 ballots, 165 move to Peltola, 117 to Sullivan, 71 to Sullivan Jr., and 125 to Heikes, while 392 go inactive.

Sullivan Jr., who starts the record with 4,687 first choices, goes next, and his 4,758 ballots, including 71 picked up from write-ins, lean toward Peltola: 1,395 move to her, 753 to the senator and 861 to Heikes, while 1,749 go inactive.

Heikes, who starts the record with 6,046 first choices, is eliminated last. His 7,032 ballots, including 125 from write-ins and 861 from Sullivan Jr., break the other way: 2,893 to Sullivan and 1,351 to Peltola, with 2,788 going inactive.

These counts come from the November 14 cast vote record, not from the Division's first-choice totals cited above. The two don’t match exactly, and some candidates’ totals run lower in the cast vote record while others run slightly higher. One reason is documented: the record leaves out ballots counted only by hand. Differences in timing and in how first choices are tallied may also contribute; how much each accounts for has not been determined.

With two candidates left, 276,112 ballots were still active; 5,527 had gone inactive along the way, including the 598 inactive from the first round, while the 1,059 blank ballots never entered the count. Newsweek's RCTab analysis of the scanned ballot records shows Peltola with 143,127 votes, or 51.83 percent, to Sullivan's 132,985, or 48.16 percent, a lead of 10,142 votes.

That is Newsweek's independent model of the scanned ballots, not the state's official result. The Division says its cast vote record contains scanned ballots and leaves out those counted only by hand. The file also does not include ballots still to be tallied. The state's official tabulation will be conducted separately by the Division, using its own records and procedures as additional eligible ballots—including hand-counted ballots and mail ballots received by the deadline—are included.

The 10,142-vote margin reflects only the ballots included in Newsweek's analysis; how the remaining and hand-counted ballots would affect the final comparison is not yet known. If no candidate finishes the count with a majority of first-choice votes, the official tabulation, run on the state's software, begins November 18.

Earlier projections pointed in opposite directions. A memo from Sullivan's campaign, circulated soon after Election Day, projected him narrowly winning a ranked-choice count of the ballots then in hand, 105,037 to 104,709. A preliminary FairVote analysis released Thursday had Peltola winning the final round 50.1 percent to 49.9 percent.

Competing Certainties

Peltola's camp considers the matter closed. Her campaign has said Sullivan has "no mathematical path" to victory, and in a post on X it called her win "all but guaranteed." The figures support a narrower claim. On first choices, under AP's estimate, Sullivan trails by more than appears to remain. In Newsweek's ranked-choice tabulation**,** Peltola leads, but the picture is incomplete.

View this post on X

Sullivan's campaign has not conceded. In a statement Wednesday, spokesperson Nate Adams said the campaign would "look forward to a complete and accurate count."

The Other Sullivan

The rankings also speak to the question that has shadowed this race for months. The New York Times reported, according to AP's race summary, that Peltola's campaign was involved in efforts to recruit Sullivan Jr., allegedly to siphon votes from the incumbent. Peltola's campaign has said it had no involvement with either Sullivan campaign, and Sullivan Jr. has said he ran on his own.

The Justice Department sued to block certification if the senator lost but would have won had every vote for Sullivan Jr. gone to him instead, pending an investigation into whether those voters meant to choose the senator. The department’s amended complaint also sought to prevent certification while that investigation took place. A federal judge dismissed the suit November 1 for lack of standing. She also wrote that segregating ballots ranking Sullivan Jr. so voters could be investigated would violate Alaskans’ right to a secret ballot. The department has appealed.

President Donald Trump also weighed in on the unusual race, posting on Truth Social on November 5 that Sullivan Jr. was "Fake Dan Sullivan." About a week after Election Day, Trump posted again, writing "Dan Sullivan PLUS Dan Sullivan" alongside a reference to the Alaska race.

Wednesday

Democrats have already won control of Congress. Alaska will help set the size of their Senate majority, and a Peltola victory would give the party its first Alaska Senate seat since Democrat Mark Begich lost to Sullivan in 2014.

On Wednesday, the last ballots mailed from overseas are due. If no candidate has a first-round majority, the Division begins its ranked-choice tabulation, and certification is targeted for November 25. For months, this race was argued over names: which Dan, whose recruit, whose confusion.

If no candidate wins outright, voters' rankings will determine which candidate wins a majority of the active ballots.

View 0 Comments

Request Reprint & Licensing

Submit Correction

View Editorial & AI Guidelines