Alaska ballot data exposes uneven voting patterns as results remain unfinished
Select a version of the text written from a presumed ideological perspective. This is not the original text, but a hypothetical version — how someone with that viewpoint might have phrased it. Tapping the current version again will return to the original or select cleaned version.
Preliminary cast-vote records from about 204,000 Alaska ballots reveal split support in the Senate race, mixed ranking behavior in the governor's contest, and broader voter patterning, while still leaving out rural precincts and many absentee ballots.
Alaska officials will release additional vote tallies on Monday and Tuesday, and the U.S. Senate and governor races remain uncalled for now. The Division of Elections has put out a "cast vote record" covering roughly 204,000 scanned ballots, but the dataset still leaves out votes from rural precincts counted by hand and tens of thousands of early and absentee ballots that have yet to be processed.
Analysis of the partial data shows that the challenger sharing the name Dan Sullivan, a Petersburg resident, drew most of his support from voters who also favored Democratic Senate candidate Mary Peltola. About 30,000 Peltola voters ranked the Petersburg Sullivan second, while only around 3,000 of Senator Dan Sullivan’s voters did the same. Charles Stewart, director of the MIT Election Lab, pointed to a modest level of ballot confusion, with roughly 16% of the challenger’s voters ranking the incumbent senator second and a quarter of the challenger’s supporters ranking Peltola second, suggesting some voters saw the two as politically aligned.
In the governor’s race, three Republican candidates urged supporters to rank all three to counter Democrat Jonathan Kreiss-Tomkins. Of the 103,000 ballots that placed a Republican first, 56% also ranked the other two Republicans, while the rest selected only one candidate. Data analyst Burke Croft of Ship Creek Group said the strategy appears to have resonated, though he cautioned that its effect on the final result is still uncertain.
The record also allows a granular view of individual voting behavior. While many voters followed expected partisan patterns, some deviated—for example, a voter in Willow ranked a conservative longshot candidate first, Peltola second, and the incumbent senator third. Additionally, voters who supported repealing ranked-choice voting were more likely to rank three or more candidates in the governor’s race (58%) compared with those who favored keeping the system (22%).
Because the current sample excludes a significant portion of the electorate, especially from remote areas, analysts stress that the final outcomes could still shift once all ballots are counted.