feddit.org

1 point

5 months ago

That linked data is collected from local American races. The winner is overwhelmingly the person who won the first round, which is the only round the majority of the time. When people claim RCV will break the two party system they are trying to claim it will change the winners. The evidence largely shows that no voting system can take a single-winner duopoly and break it.

Any new voting system would require only a simple bill from the legislature. “Ballots instructions for every election at every level shall direct voters to select any number of candidates. The candidates with the most votes wins their respective election.”

report

[ - ]

Maggoty@lemmy.world

1 point

5 months ago

Proportional representation specifically refers to how parties divide the available seats in a parliamentary body. Not how you cast your vote.

RCV allows for changes that FPTP doesn’t but that has never meant this would be shaken up right away. Mostly it’s a way to avoid vote splitting. So you can run a progressive, moderate, conservative, and an alt right candidate without the traditional alliances worrying about vote splitting.

report

[ - ]

2 points

5 months ago

Proportional representation specifically refers to how parties divide the available seats

I apologize for not addressing that, but I didn’t think it required expanding on. Yes, that’s correct. I feel the preferred proportional method is Sequential Proportional Approval Voting

RCV doesn’t eliminate vote splitting, it only mitigates it. If two candidates have similar support in a non-final round, one can act as a spoiler for the other. The problem is that it’s harder to understand and FairVote used to lie about it, so a lot of people think it’s not a problem. The Alaska special election from a few years back is an example of a spoiler election. If Palin hadn’t run (or fewer Palin voters voted) the other Republican would have won. If you want to completely eliminate vote splitting you have to move to a cardinal voting method that satisfies the independence of irrelevant alternatives criterion, which is most of them, including approval voting.

report

[ - ]

Maggoty@lemmy.world

0 points

5 months ago

SPAV is specifically constructed to work with proportional representation. It iterates until all seats are filled. But in the US, by Constitutional law, it’s one seat per geographical district.

About RCV though it’s still head and shoulders above FPTP, and easy to understand. About Alaska specifically, I don’t understand why you would call the party backed candidate who got more votes a spoiler?

Palin lost in the second round because roughly half of Begich’s voters did not want Palin. If the less popular Republican candidate wasn’t in the race then Peltola still wins. This was a case of RCV working exactly as advertised. A traditional party primary would have nominated Palin, not Begich, and she would have lost anyways.

report

[ - ]

1 point

5 months ago

SPAV is specifically constructed to work with proportional representation. It iterates until all seats are filled.

Yes, that’s how it works. The first round is functionally identical to regular approval which is why I like using the two. Approval for single-winner, SPAV for multi-winner.

But in the US, by Constitutional law, it’s one seat per geographical district.

I’m pretty sure it’s just federal law, but I would have to double check. Not like Congress would change it anyway.

A traditional party primary would have nominated Palin, not Begich, and she would have lost anyways.

That’s pure speculation. But using the voting data from the general, we Begich was preferred to both Palin and Peltola in head-to-head matchups. Palin pulled enough votes from Begich to eliminate him in the first round and he lost to Peltola in the second. If Palin hadn’t run Begich would have won.

You can read more about it from the linked sources here.

Here’s the most relevant section:

Some social choice and election scientists criticized the election in published opinion pieces, saying it had several perceived flaws, which they technically term pathologies. They cited Begich’s elimination as an example of a center squeeze, a scenario in which the candidate closest to the center of public opinion is eliminated due to failing to receive enough first choice votes. More voters ranked Begich above Peltola, but Palin played the role of spoiler by knocking Begich out of contention in the first round of the run-off. Specialists also said the election was notable as a negative vote weight event, as those who voted for Palin first and Begich second instead helped Peltola win by pushing Palin ahead of Begich in the first round.

Elections scientists were careful to note that such flaws (which in technical terms they call pathologies) likely would have occurred under Alaska’s previous primary system as well. In that binary system, winners of each party primary run against each other in the general election. Several suggested alternative systems that could replace either of these systems.

You have to be careful analysing RCV results, because people tend to only look at what the election did, and fail to look at what it didn’t do. One of the good things about RCV is that it collects a fair bit of information, but then it usually ignores a fair bit of it. When trying to understand whether a candidate was a spoiler or not, you have to ask what would have happened if they didn’t run at all, which requires considering collected information the “unaltered” election didn’t take into account. If removing them from the election changes the winner of the race, then they were a spoiler. We know that removing Palin would have resulted in a Begich win over Peltola, so that makes Palin a spoiler. She’s a losing candidate that changed the winner of the race simply by entering, assuming voter preferences are stable.

report

[ - ]

Maggoty@lemmy.world

-1 points

5 months ago

Yeah no. That’s a lot of noise to ignore that the party and Republican voters preferred Palin. Begich wouldn’t even have been there in traditional FPTP. Calling the most popular candidate from a party “a spoiler” is a rhetorical device republicans came up with to go after RCV.

Peltola is also hardly some far left representative. So calling it a center squeeze is a bit rich. This entire write up screams, “I can’t approve of the Alaskan RCV election because I’m paid not to.”

To be clear, not you, the author of that Wikipedia article.

Edit to add- and sure enough if you go to the talk page there’s a partisan group defending it from any changes to bring it towards Wikipedia objectivity standards. This is why your teachers told you Wikipedia is a bad source.

report

[ - ]

1 point

5 months ago

I didn’t tell you to trust Wikipedia, I told you to follow through to the linked sources in that section. Also, that talk page suffers from the same problem you’re having, which is assuming that the RCV results are the same thing as the public opinion. The entire point of analysing the data is to look past the voting system used and try to understand what people’s preferences are. Here’s another (very long) source that summarizes the full ballot data and explains that, yes, Palin was a spoiler. Justifying this as acceptable by saying that RCV followed its own rules (which it must do, by definition) is the same as saying Ralph Nader spoiling the 2000 election was the correct outcome because those people had Nader as their favorite.

Look I don’t hate RCV. It’s certainly better than FPTP. I just don’t want people to have false ideas about its function. Spoilers can and do happen, they just behave differently than FPTP. And, I will add, they behave in a much more acceptable way, with RCV spoilers being much more likely to be competitive candidates compared to FPTP. Plus, RCV has less center-squeeze than FPTP. Mathematically, Approval doesn’t have spoilers nor does it have a center-squeeze effect, and I would argue that it’s better than both RCV and FPTP for this and other reasons, but I do want to re-confirm that FPTP is the worst.

report