Consensus Voting Shows the Voters Preferred Nick Begich in Alaska's Special Election

In Alaska's special election, the voters spoke clearly. But they weren't heard. They strongly preferred Republican Nick Begich to Democrat Mary Peltola by 52.5 to 47.5 percent to represent them in Congress. Instead, Instant Runoff Voting (IRV) delivered the victory to Peltola.

This is not a theory. It's a fact easily proven by looking at the ballots. 

The Alaska board of elections has released electronic voting records for the special election. A few hand-counted ballots aren’t in the electronic record, but not enough to alter the results. This detailed information has the anonymous contents of every ballot cast in the special election and allows a more detailed examination of the results. That’s where things get interesting.  

A simple consensus check prior to conducting the Instant Runoff would produce a result that may have made a majority of Alaskan voters much happier than they are now. 

The ranked-choice ballots from the August 16 special election were intended for Instant Runoff, but they can also be used to see which candidate the voters prefer given any two candidates.  By listing Begich before Peltola, a voter says they prefer Begich over Peltola and would vote for Begich in a one-on-one contest between the two candidates.

Using that principle, I went through the ballots from the Alaska special election. There were 87,889 ballots where the voter preferred Begich over Peltola, but only 79,449 ballots where the voter preferred Peltola over Begich. I did the same for Begich and Palin and finally for Peltola and Palin.

The voters preferred Begich to Peltola, and by an even higher margin preferred Begich to his fellow Republican Sarah Palin. Begich defeated each of the other candidates in the race when compared one-on-one. Begich should be the winner because any other winner overrides majority rule, the most basic rule in a democracy.

Nick Begich was also first or second on 94.2% of the ballots that listed more than just a first choice; he was the consensus candidate.  

It’s impossible to give every voter exactly what they want and no candidate will ever be every voter’s favorite, but when one candidate is preferred to every other candidate, they should win. It’s like a round-robin sports tournament; if one team is undefeated, they are the winner.  If no candidate is undefeated – about one time in 1000 – then the Instant Runoff can be used as a tiebreaker.

Begich wasn’t the first choice of either party’s voters, but that’s exactly what we need.  Electing one party’s favorite candidate without regard for what the other party’s voters want only creates more polarization and division.  We need candidates who unite instead of divide and a voting system that rewards those candidates.

The conservative outcry blaming ranked-choice voting for Peltola’s victory is ironic. Palin would likely have won a Republican Primary in the same way that voters who listed a Republican first chose Palin over Begich in the first round of Alaska’s IRV. That would have set up a general election between Palin and Peltola, just like the one that happened under IRV.  Palin lost to Peltola in Republican-leaning Alaska because Alaska voters preferred Peltola to Palin and not because of any voting rule. Reverting to partisan primaries would not solve the problem.

Palin’s voters might well have believed that if their first choice didn’t win, their vote would count for their second choice.  Other than a few ballots that listed write-in candidates, the only second choices that mattered were the 28.5% of voters that listed Begich first.  Mary Peltola won the election because the second choice for Palin voters never mattered.

I helped write California Assembly Bill 2161; it proposes using Consensus Voting to calm the chaos of the California gubernatorial recall system. AB 2161 uses pairwise comparisons to see if there is a consensus winner; for the rare instance when there isn’t a consensus winner, it uses Instant Runoff to select the winner. These election rules would have prevented the problem we have today in Alaska.

The Alaska election reform got so much right. It did away with partisan primaries that test candidates for partisan purity and opened the field to more representative candidates. It allows the voters to use a more expressive ballot where they can indicate their second and third choices. But it needs to do more. It needs to listen to the voters.

Rather than moving backward to partisan primaries and pluralities that often produce divisive and polarizing results, reform needs to move forward by adding a consensus check to ensure that when the voters speak, the voting system hears them.

Robbie Robinette is a data scientist and cofounder of RGM Advisors, a quantitative trading firm. He has spent the last five years working on voting reform.

This piece was reviewed and edited by Henry A. Brechter, AllSides Managing Editor (Center bias), Joseph Ratliff, AllSides News Editor (Lean Left bias) and AllSides CEO John Gable (Lean Right bias).