On the Alaska Special election

OK, I promised to revisit this, so here I am revisiting this.

Alaska’s current system is as follows:

The “primary” is an at-large race with all candidates on one ballot, irrespective of party. Voters get a single vote. All but the top four candidates are eliminated. Then, in the “general”, voters rank choice the top four candidates, and the ranked choice ballots are tabluated by reassigning the votes of the last place candidate down their rank order.

Primary results:
Palin 27.1
Beigich 19.1
Gross 12.8
Peltola 10.1
Sweeney 5.9

First round ranked choice:
Peltola 40.2
Palin 31.3
Beigich 28.5

Second round ranked choice:
Peltola 51.5
Palin 48.5

Doing the math on the totals showed the Beigich vote broke
50.3 Palin, 28.7 Peltola,, 21.0 No one

But, looking at these results, you might immediately ask “what about Al Gross?”

He did qualify for the second round, but being a Democratically aligned independent, opted instead to drop out. What would happen if he had stayed in? Let’s just scale up the top four from the first round so that the sum total is 100%

Primary with only top four:

Palin 39.2
Beigich 27.6
Gross 18.5
Peltola 14.6

These totals still look badly out of line with the first round results, with Palin doing better, and Gross + Peltola doing much worse than Peltola did in the first round. It is pretty well documented that, since the primary, the Dobbs decision and all of the Democratic congressional legislative successes happened, and Democrats, across the board, are doing better than they were two months ago. To account for this, I’m going to go and make the Gross + Peltola total match the Peltola total and the Palin + Beigich total match it’s actual result. I’m also going to flip Peltola’s and Gross’s position in the general, both for a better apples to apples comparison with the real result, and assuming that “Dobbs momentum” would have broken toward a pro-choice woman

First Round General, with gross, rescaled to match real general

Palin 35.1
Beigich 24.7
Peltola 22.5
Gross 17.7

Now, this is the key point. How the ranked choice breaks down depends a lot on how those Al Gross votes break. If they 100% all go to Peltola, then Peltola goes to first place in the second round, and we play out the same result. But what happens if they do something like break something like 70% for Peltola, 20% for Beigich, and the rest spoiled?

Second round general, after gross votes break
Palin 35.1
Peltola 34.9
Beigich 28.24

apply the same fraction vote break to Beigich from the real first round, and rescaling to 100 to account for the spoiled ballots:

Third round general, hypothetical
Palin 52.9
Peltola 47.1

There are a lot of hypotheticals in this arrangement, and this is not meant to be a demonstration of “this is definitely how the race would have turned out”, but the key point here is that the outcome ranked ordered balloting depends a lot on the order in which candidates are eliminated from the ballot, and it opens up races to strategic decisionmaking. Here, the decision by Al Gross to drop out of the election is critically important. We got to this conclusion by making not-wild assumptions.

Also, none of this is to discount the accomplishement Peltola made. Here, we handwaved the real campaigning and work that was done as “Dobbs momentum”, but that’s almost certainly only half of the story. I’m not trying to talk about her real campaign and real victory, but just trying to demonstrate how ranked-order balloting can play out, and how important strategic concerns and party coordination remain in these systems.

Thank you for reading.