So you tell me.

OK, now that I have more time to look at it:

“The bad apple sort” seems to be identical to Coombs’ method? (Not a “variation”, but equivalent?)

“Balanced voting” I haven’t seen before. I don’t think it’s the same as Nanson/Baldwin because they take all the rankings into account, while you only consider top and bottom in each round.

Note that there are other systems with “Balanced” in the name, such as “Balanced Plurality” and “Balanced Approval”, both of which feature explicitly voting for or against candidates.

Are you aware of any scale for measuring polarization?

I like showing people these graphs of voter vs representative ideology distribution:

Because of First Past The Post voting, party primaries, and other polarizing effects, moderate/centrist voters are almost completely unrepresented in congress.

These are taken from Reform and Representation: A New Method Applied to Recent Electoral Changes. This metric has been criticized because someone who has extremist opinions equally divided between right-wing extremism and left-wing extremism would appear in the middle of the chart, despite not being “moderate”, but I think it’s still a good illustration.

Here’s what I wrote to Rob Richie replying to his thoughts the Medium article I wrote. I think I nailed it. What do you think?

If you can get through to Rob Richie, you win the internet. :)

You say:

that IRV has leveled the playing field between Parties by virtually eliminating the spoiler effect

As you can see from this article, I don’t believe it actually does this. It protects identical candidates from vote-splitting, but when there are multiple candidates with non-identical ideologies, it still suffers from vote-splitting and the spoiler effect. (I’ve been meaning to make an animation, similar to these, of the worst-case IRV scenario, where the most-representative candidate is eliminated first, their votes transfer outward to the candidates around them, who are in turn eliminated, and so on until only the two least-representative candidates remain in the final round.)

I agree that your “balanced IRV” would be more likely to elect consensus winners than regular IRV. I’m not sure about the exact properties and how it compares with the many other systems like it. I generally prefer Condorcet-compliant methods if we’re going with ranked ballots.

I think you’re wasting your time talking to FairVote, though. From what I can tell, they are thoroughly married to IRV and won’t even consider variations to it. “Momentum” is the only argument that works for them and they aren’t going to risk hurting that by acknowledging the flaws of IRV.

I think more people should pester Represent.US instead. They exclusively promote IRV in their marketing, but they do get behind other voting reforms if the local chapters prefer them, such as STL Approves. I would like to see their marketing shift towards promotion of “Voting reform” in general rather than exclusively pushing IRV.