Axios reported yesterday that the chair of the Democratic National Committee, Ken Martin, is mulling the use of ranked-choice voting (RCV) for the 2028 presidential primary. Interest in RCV has exploded since New York City Democrats nominated, and subsequently elected, Zohran Mamdani. Newsmax notes that “In addition to New York City, other states and localities that have adopted ranked-choice voting include Alaska, Maine, San Francisco, and Minneapolis.”
As an Alaskan-by-birth who voted in the state’s first RCV election — I wrote about it in 2022 for the Wall Street Journal — this would be the wrong move for the Democratic Party and the nation. RCV invites voters to rank candidates in order of preference (up to three, four, or five choices, depending on which locality). If no candidate wins more than 50 percent of first-choice votes, tabulation occurs in rounds, distributing second-choice votes from eliminated candidates. It doesn’t have to be paired with the elimination of the traditional party primary, but it usually is, such as in Alaska and Minneapolis.
In 2020 Alaskans approved, by a margin of about 3,700 votes, a 25-page initiative promising to improve elections and eliminate “dark money” influence in campaigns. In the fine print, the initiative instead eliminated the party primary and instituted RCV. The 2022 election featured 48 primary candidates on the ballot and the top four advanced to the general election (though only after some legal questions about eligibility). The two Republicans, Sarah Palin and now-Representative Nick Begich, split their first-choice votes, tallying 60 percent together, and the so-called moderate candidate, Mary Peltola (D-Alaska), won.
Consider the claim that RCV leads to more moderate electoral outcomes. Mathematically, Begich would have been preferred to either Palin or Peltola in a head-to-head contest (a concept called the Condorcet winner), but by coming in third in the initial round of voting, he was eliminated and his second-choice votes redistributed. Voters shouldn’t have to figure out the “least bad” strategy to support the candidate they want.
In the 2025 mayoral election in Minneapolis, RCV retained Mayor Jacob Frey over Omar Fateh. Yet eighteen percent of the ballots cast ranked Frey as first preference, and no one else. A quarter of all ballots left a second choice blank, and 37 percent left their third choice blank. That just sounds like one-person, one-vote, but with extra steps.
In fact, a University of Illinois Law Review research paper analyzing Alaska’s 2022 results suggests that RCV elects candidates that are “more divergent ideologically” than the state’s median voter, with exacerbated effects for more polarized states. For the DNC, that means an RCV primary will embolden the further-left wing of the party.
It would be one thing if voters were spontaneously clamoring for changes to one-person, one-vote systems. But in each state that sees ballot initiatives pop up, they’re sponsored by a nonprofit funded with national advocacy dollars. Opponents to Alaska’s 2024 ballot measure to repeal the initiative raised $14 million from its top three contributors, Unite America, Article IV, and the Action Now Initiative. Axios reports that pushing for the change at the DNC is another usual suspect, FairVote Action.
As I detail in the Spring 2025 issue of Thinking Minnesota, ranked-choice voting mostly lost at the ballot box in 2024. Nevada, for example, had originally approved a ballot initiative in 2022 to institute RCV, but due to a quirk in their state constitution, they needed to approve it again in 2024. Yet RCV was rejected that year. Colorado, Idaho, Montana, and Oregon all rejected RCV last year, with blue-voting Colorado and Oregon rejecting with solid majorities.
Adopting RCV for the 2028 Democratic primaries would only serve to drag the party further to its left, worsening polarization and creating candidates that are farther apart on the issues that the average Joe Voter cares about. Maybe that helps Red Team (or maybe not), but it will damage the culture for all.
