Republicans have learned from what happened last time, when they remained split. Looking only at first-choice votes in the 2022 special election that Ms. Peltola won, she had 40.2%. The GOP’s contenders, Sarah Palin and Mr. Begich, had 31.3% and 28.5%, respectively. Under the RCV rules, Mr. Begich was eliminated, and his supporters were reshuffled to their subsequent preferences. Enough of them didn’t like Ms. Palin that the victory went to Ms. Peltola, with 51.5%.
Strangely, though, this result was sensitive to the order of elimination, meaning that the final No. 1 depended on who was the initial No. 3. If Ms. Palin had been dropped instead, a strong majority of her ballots would have gone to Mr. Begich, who would have beaten Ms. Peltola.
Not only that, he’d have won about 52.5%, a bigger victory than Ms. Peltola’s ranked-choice majority. Doesn’t that seem . . . odd? What if some Democrats ranked Ms. Palin first on their ballots to ensure that the most polarizing GOP opponent made the final round?
This is actually a really interesting real-life instance showing that, in this particular case, RCV didn't achieve an optimal outcome.
But what's a better voting system that would have achieve the presumably optimal outcome here (the less-polarizing GOP candidate Begich getting elected)?
If we just compare to simply plurality voting, the outcome seems the same: Peltola got the highest vote of the 3, so in regular FPTP voting, she would have won. Or if we compare to FPTP party-based elections as is normally done in the US, I think again it would have been the same: the GOP vote was split between Palin and Begich, and presumably Palin would have narrowly won that party's primary, and then enough GOP voters hated her enough to list Peltola 2nd-choice in the RCV, so presumably they would have also voted for Peltola in a regular party-based election, so again the outcome would have been the same.
If anything, I think this shows that these multi-candidate elections have odd-looking results when there's uneven numbers of candidates for the different parties. If there had been 2 DNC candidates, this might have played out quite differently. But maybe not: in this particular election, one of the candidates (Palin) was extremely polarizing and many voters truly hated her, but she had enough support from the right-leaning voters to get a majority on that side, over the more moderate Begich. But that was unique to the GOP side; there's no reason to think that the left-leaning voters of Alaska would have similarly voted in high numbers for an extremely polarizing candidate like AOC: Alaska just doesn't have many far-left voters the way some other states do. So it seems like the right outcome may have been reached here, because so many GOP voters in Alaska are willing to support a very extreme and polarizing candidate, rather than a more moderate one: they ended up making themselves irrelevant.
> But what's a better voting system that would have achieve the presumably optimal outcome here (the less-polarizing GOP candidate Begich getting elected)?
One I don't think gets talked about enough is approval voting. Vote for as many candidates as you like, whoever gets the most votes wins. Voting for all of them is effectively the same as voting for none of them. It also has the massive benefit for the general population that it's simple to explain. Even gives third-party candidates and moderate candidates a real chance at winning, since no one would feel like they're throwing their vote away by including them.
i like this. i think it even works for electing multiple candidates. we need 5 people in a committee. everyone list their preferred 5 or 10 or more choices. add up all the votes and the 5 top get chosen.
I think "begging the question" describes your answer, in the classical sense of "assuming what you need to prove."
You are assuming that a multi-party, multi-candidate election is a desirable norm. But that needs to be proven by you, not assumed.
In a pre-RCV system, the parties would choose their candidates, however they do it, and then the general election would pit one candidate of each party against each other, plus whatever other parties qualify for the ballot. The public has not accepted that that must be eliminated. It did make things hard for third parties, but so it goes. Prove that that's a bad thing.
So you'll take your arguments and go home, and not even try to convince me otherwise and potentially win me over with your candor and intellect and personality?
That seems like the problem with politics as a whole these days.
No, citing one example of the current system as "proof" that the whole system must be thrown out is the problem with "politics as a whole."
Also: thinking that a few paragraphs on Hacker News constitutes sufficient wisdom is another problem with "politics as a whole." It is not going to convince you no matter what I or anyone else says.
Do some research. That probably means reading some books, hate to warn you.
> We have a two-party system, and that's just the way it is.
Some argue that our current voting system creates the two party system we all know and hate. Derek from Veritasium did a video about this the other day. Really good stuff.
No need to come to Palin's or the Republican's defense here. Candidates are more than capable of trying to campaign for the second place votes.
Check out this IPSOS poll that says otherwise. That was run in January. Yet somehow by May both of the upopular candidates won their national primaries.
https://alaskapublic.org/2024/05/29/ranked-choice-voting-tha...
Gamesmanship: Alaska is a red state. Only RCV would result in a Dem being elected.
https://www.wsj.com/opinion/alaska-ranked-choice-voting-mary...
(not on archive, sorry. Here's the relevant part)
Republicans have learned from what happened last time, when they remained split. Looking only at first-choice votes in the 2022 special election that Ms. Peltola won, she had 40.2%. The GOP’s contenders, Sarah Palin and Mr. Begich, had 31.3% and 28.5%, respectively. Under the RCV rules, Mr. Begich was eliminated, and his supporters were reshuffled to their subsequent preferences. Enough of them didn’t like Ms. Palin that the victory went to Ms. Peltola, with 51.5%.
Strangely, though, this result was sensitive to the order of elimination, meaning that the final No. 1 depended on who was the initial No. 3. If Ms. Palin had been dropped instead, a strong majority of her ballots would have gone to Mr. Begich, who would have beaten Ms. Peltola.
Not only that, he’d have won about 52.5%, a bigger victory than Ms. Peltola’s ranked-choice majority. Doesn’t that seem . . . odd? What if some Democrats ranked Ms. Palin first on their ballots to ensure that the most polarizing GOP opponent made the final round?