In a highly unusual move, the Trump administration asked a federal court to block certification of the Senate election in Alaska if the Republican incumbent loses.

The Justice Department argued that confusion over another candidate on the ballot with a name very similar to the incumbent, Senator Dan S. Sullivan, could taint the election result. Mr. Sullivan is in a close race against Mary Peltola, a Democrat, but the ballot also features another Daniel Sullivan, but this one is Daniel J. Sullivan, Jr.

The lawsuit argues, “There is a realistic chance that, even if a majority of Alaskans wish to vote for Senator Sullivan, Daniel J. Sullivan’s presence on the ballot will result in a victory for Peltola.”

To guard against voter confusion, the suit contends, Alaska should “not certify any election result that may be impacted by voter confusion.” The possibility of such confusion exists, the suit continued, if “(1) Senator Sullivan does not win the election; and (2) had all votes cast for Daniel J. Sullivan been cast instead for Senator Sullivan and Alaska’s rank-choice process played out accordingly, Senator Sullivan would have won the election.” The vote should not be certified until “an investigation to determine whether votes cast for Daniel J. Sullivan were meant to be cast for Senator Sullivan and any further procedures or relief this court deems necessary and appropriate.” It is unclear how such an investigation would work since ballots are cast anonymously.

I read the complaint several times and still do not understand the legal basis for the claim. The Constitution gives the states and Congress oversight over elections. As far as state oversight goes, this issue has already been litigated in Alaska. Its state Supreme Court affirmed in June of 2026 that both Sullivans could be on the ballot. Congress has not acted on the issue. This means that neither entity designated by the Constitution to oversee the election has found this to be significant enough to affect the election. In fact, the ballot takes care to avoid the confusion by explicitly listing Senator Sullivan as the Republican incumbent and listing the other Sullivan without a party affiliation.

Alaska will use ranked-choice voting. Ranked-choice voting will only come into play if no candidate receives an initial outright majority. If neither Peltola nor the incumbent has a majority on the first-place ballots, only then does ranked-choice come into play. At that point the candidate with the lowest vote total is dropped, and the second choices on those ballots are reallocated to that second choice. (See post of October 14 for an explanation of ranked-choice voting.) If the voters for the “fake” Sullivan list the incumbent as their second choice, then the Senator will be reelected.

Part of the reason for ranked-choice voting is to prevent a candidate from winning with a plurality when most of the voters were actually voting against the plurality winner.

With national ranked-choice voting, Trump may not be president. In 2016, he won crucial swing states not with majorities but with pluralities. In those states third- and fourth-party candidates received enough votes to deprive both Trump and Clinton of majorities. If these voters had the opportunities to list second choices, Clinton may have won. For whatever the reason, many voters who did not vote for Trump did not vote for Hillary Clinton either. They voted for third party candidates instead, and in key states those votes for Libertarians or Greens gave Trump pluralities and the decisive electoral votes.

In the swing state of Pennsylvania in 2016, for example, Trump won the state with the plurality of 48.8 percent (Clinton got 47.6 percent). Trump and Clinton together garnered 96.4 percent of the total ballots, while in 2012, the major presidential candidates had received 98.8 percent in Pennsylvania. In the intervening four years, third parties nearly trebled their votes from 69,000 to 192,000. A sizeable number of voters apparently wanted to vote for neither Trump nor Clinton, and as a result, Trump won.

A similar scenario played out in Michigan where Trump won in 2016 with only 47.3 percent of the vote. Again “others” made the difference. In 2012, only 1.4 percent of the ballots were not cast for the major parties, while in 2016 that number rose to 5.2 percent for minor parties, with vote totals increasing from 65,000 to 250,000. Trump’s plurality (again not a majority) was a mere 10,000 votes. The move to third parties again allowed him to win a plurality and get all of Michigan’s electoral votes.

And the beat went on. In Wisconsin, Trump won the plurality at 47.9 percent, but again, third parties swung the state. In 2012, third parties got 28,000 votes and 0.9 percent of the total. In 2016, third parties garnered 137,000 votes accounting for 5.4%.

We do not know who those third-party voters disfavored more, Trump or Clinton. Ranked-choice voting tends to answer that by allowing a second-place vote. If the second choices had primarily gone to Clinton, the results might have been different in 2016. Moreover, had Trump not been elected in 2016, I doubt that he would even have been nominated in 2024, much less elected. And even in 2024, although he came close to a nationwide majority, he still got only a plurality with 49.8% of the vote.

Perhaps this explains the Alaskan lawsuit, which Trump proudly proclaimed he ordered.