Preliminary ranked choice voting results in Alaska’s 2026 elections
On November 3, Alaskans went to the polls in critical ranked choice voting (RCV) elections for Congress, governor, and state legislature. On election night, the Division of Elections released a preliminary cast vote record (CVR) – an anonymized digital record of most ballots counted so far – which can be used to create unofficial RCV tallies.
Notably, the cast vote record includes about 203,000 ballots, while the Division of Elections has reported first-choice results for about 215,000 ballots. The 215,000 ballots represent about 75% of the estimated count. We are releasing these RCV tallies for transparency, and for a sense of how RCV will impact the outcome as more ballots are counted; however, they do not match one-to-one with all first-choice results released by the DOE. As the most significant example, Senator Dan S. Sullivan has the most first-choice support in the current tranche of ballots in the CVR; however, he trails Mary Peltola in the reported first-choice results.
In these preliminary RCV tallies, both Democrats and Republicans lead statewide races. Democrat Jonathan Kreiss-Tomkins leads the race for governor 55%-45%, with his margin shrinking about 9.5 points in the RCV count. Incumbent Republican Nick Begich leads the U.S. House race 54%-46%, with his margin shrinking about 2.5 points in the RCV count. Incumbent Republican Dan S. Sullivan leads the U.S. Senate race 51%-49%, with his margin growing about 0.5 points in the RCV count; as noted, Sullivan’s first-choice lead does not match the full results and the margin change in the RCV count is the most relevant finding at this point.
Two additional notes as more ballots are counted:
- Ballot Measure 2 – which would repeal Alaska’s election system – is trailing 52%-48%.
- In prior elections, late-arriving ballots in Alaska have tended to favor Democrats more than those counted early.
Governor
In the preliminary results, Jonathan Kreiss-Tomkins leads with 48.5% of first choices. In the RCV tally, he defeats Bernadette Wilson 55-45%. It is possible that Kreiss-Tomkins earns over 50% of voters’ first choices, once all ballots have arrived.
U.S. House
In the preliminary results, Nick Begich leads with 50.3% of first choices. In the hypothetical RCV tally, he defeats Bill Hill 54-46%. It is possible that Begich stays over 50% of voters’ first choices, once all ballots have arrived.
U.S. Senate
In the preliminary results, Dan S. Sullivan leads with 48.9% of first choices. In the RCV tally, he defeats Mary Peltola 51-49%. It is possible that either Sullivan or Peltola earns over 50% of voters’ first choices, once all ballots have arrived.
How do voters use their backup choices?
In the governor, House, and Senate races, a majority of voters ranking a lower-performing candidate as their first choice ranked one of the two finalists on their ballot — giving them greater voice in a crowded field.
This includes 26,747 voters for Treg Taylor and Dave Bronson in the governor’s race, 11,003 voters for Eric Hafner and Jim McDermott in the House race, and 4,481 voters for Gerald Heikes and Daniel J. Sullivan in the Senate race. Compare this to a multi-candidate race without ranked choice voting, like Tuesday’s Ohio Senate race – 222,781 voters (4.9%) cast a ballot for the Libertarian or independent socialist, and had no opportunity to weigh in between the candidates who actually had a chance to win.
In the Senate race, 44% of voters ranking Republican Gerald Heikes first had their vote count for Dan S. Sullivan, compared to 16% for Mary Peltola. 33% of voters ranking Daniel J. Sullivan below had their vote count for Mary Peltola, compared to 22% for Dan S. Sullivan. The table below shows a large number of Democrat-leaning voters who rank Daniel J. Sullivan first and Peltola second, suggesting some form of protest vote. Republican-leaning voters who rank Daniel J. Sullivan first are more likely to rank Senator Dan Sullivan second.
Ballots that ranked Daniel J. Sullivan first voting broken down by governor vote
| First choice in governor’s race | Transfers to Peltola | Transfers to Senator Sullivan | Inactive | Total |
|---|---|---|---|---|
| Republican candidate (Wilson, Bronson or Taylor) | 270 | 593 | 952 | 1815 |
| Kreiss-Tomkins | 818 | 166 | 387 | 1371 |
| Blank first ranking for governor | 35 | 19 | 90 | 144 |
| Total | 1123 | 778 | 1429 | 3330 |
What are the preliminary results for Ballot Measure 2?
Alaskans also voted yesterday on whether to keep ranked choice voting for future elections. The full reform package at stake includes RCV, Alaska’s top four primary, and an anti-dark money rule requiring the public disclosure of contributions over $2,000.
Currently, “No” on repeal leads with 52% of the vote. By comparison, in a failed 2024 effort to repeal Alaska’s election system, “No” on repeal trailed by 2 points. It did not take the lead in the polls until Day 13.
Why can election results take 15 days in Alaska?
As noted above, these tallies are preliminary, ballots are still being counted, and the leaders could change in the coming days – completely unrelated to RCV. In prior elections, late-arriving ballots in Alaska have tended to favor Democrats more than those counted early.
Alaska accepts absentee ballots up to 15 days after the election, and the state has a long history of extended vote-counting that predates its adoption of ranked choice voting. Many ballots arrive late in Alaska’s 15 day window because of the state’s size; tens of thousands of voters live in areas reachable only by boat or plane.
In two recent close U.S. Senate races, the winner was declared at the end of the 15-day period – Sen. Lisa Murkowski in 2010, and Mark Begich in 2008. More recently, in 2024, Ballot Measure 2 – a “Yes/No” vote regarding Alaska’s election system – also took all 15 days.
We appreciate the Alaska Division of Elections’s decision to release preliminary cast vote record data, which is also available for state legislative races. This increases transparency, and offers a reminder that RCV results can be produced as quickly as those in non-RCV elections.
