Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

What Is Voter Satisfaction Efficiency?

electionscience.github.io

Post reply on HN
Print view View original

21–30 of 128 posts

Re: What Is Voter Satisfaction Efficiency?

#21

Post by Sniffnoy »

dane-pgp wrote: post #16

It may be hard to model, but I would have liked to see how Asset Voting compared to the other voting systems. Asset Voting seems to be almost unheard of, even among voting reform enthusiasts, despite being over 100 years old and one of the simplest systems out there. Basically it works the same way as a FPTP election, except that after the votes are counted, the losing candidates one by one get to reassign their vote…

I've seen some voting systems include a "delegation" feature, where instead of submitting a full ranking of candidates (or whatever), you can just vote for a single candidate and your full ballot will be determined by their full ballot. What you're describing seems to be basically IRV but where you are forced to delegate?

(OK, I guess the runoff isn't truly instant, but still.)

I worry that this might have the same sort of monotonicity problems as IRV normally does, only in terms of the candidates' ballots instead of the voters'.

Re: What Is Voter Satisfaction Efficiency?

#23

Post by jabl »

marcosdumay wrote: post #9
cosmojg wrote: post #2

Looking at the results, it's a wonder that IRV (aka "ranked-choice voting") gets more lip service than STAR and approval voting. These latter two seem obviously and measurably superior. My guess is that it all comes down to a lack of education. Maybe civics ought to be a required course in high school and/or college?

Personally, I fail to see how STAR and approval are superior. If I were to lobby for some electoral change, I would lobby for approval votes, because they are easy to lobby for (just change "you gt ONE vote" into "vote for as many people you like"). But compared to ranked choice, it creates a situation where you can't differentiate candidates between who you clearly have a preference, and this has large odds of chang…

OTOH, with ordinal methods you are forced to distinguish between candidates that you think are equally good. Say if you have three candidates, A, B, and C, where A and B are very similar and you like both, though with a slight preference to A, whereas you dislike C very much. With approval voting you'd approve A and B, with score voting (say the vote is an integer between 1 and 5) you could vote A=5, B=4, C=1 (or A=5, B=5, C=1 if you think A and B are really close). Whereas with an ordinal system you'd vote A-B-C, but you have no way of 'telling the voting math' that the distance between A and B is much shorter than between B and C.

Looking at the results of these simulations described in the article, and similar simulations using 'Bayesian regret', it seems that these issues tend to make cardinal systems better. Also, considering the various gotchas of voting systems, there is the famous Arrow impossibility theorem. However, it applies only to ordinal systems. There is no similar thing for cardinal systems.

Re: What Is Voter Satisfaction Efficiency?

#24

Post by eyelidlessness »

vlovich123 wrote: post #18
infogulch wrote: post #14

Earlier quoted context omitted.

This is one of the great aspects of states (localities in general) mostly governing themselves even in important things like voting. Some state tries something new, and everyone observes and can learn from the results. If all of the new things states are trying were concentrated in just one state the churn would destabilize it too much. By spreading out the chaos we collectively get to try many new things at once, bu…

One major drawback is that efficiencies of scale can’t be taken advantage of and inequities at scale are easier to propagate. So we have a very slow process that ends up disenfranchising a larger amount of people than I’d be comfortable with in a modern developed democracy. This is extremely evident when you start comparing how well governments run in other Western democracies.

I meant to respond to GP earlier and you got to it before me admirably, but if I may be more terse: some of these states are still experimenting with rehabilitating the Confederacy.

Re: What Is Voter Satisfaction Efficiency?

#25

Post by JumpCrisscross »

_delirium wrote: post #11
rayiner wrote: post #7

Not knowing on the same day who won the NYC primary is much more annoying than I thought it would be. I wonder if there is a psychological effect in play here, like lag in a game controller. Makes you feel less like you’re actually in control.

Do you know what the source of the lag is? There's a concept of "rounds" in IRV, but if you have access to computers, a "round" should resolve in less than a second, if election data is digitized. Is the lag digitizing the votes in the first place?

> Do you know what the source of the lag is?

Absentee votes. Also, voters have some time to “cure” mistakes, e.g. forgetting to sign their ballot envelope.

TL;DR Nothing intrinsic to IRV.

Re: What Is Voter Satisfaction Efficiency?

#26

Post by js8 »

I like range voting (of which approval is a special case). It's simple, and it always takes everybody's point of view into account (it doesn't lose any voter information), which is a really nice property IMHO.

Yes, you can argue that the more extremely somebody puts the ranges, they have an advantage. (I would argue we should always normalize the cast vote so that most wanted candidate gets 100 and the least wanted 0, but then we take away people's ability to express only mild interest.) And through that, they can in certain cases manipulate the outcome.

I feel like we cannot really meaningfully say that there is a difference between "real" preferences and the "manipulated" ones. For example, this article (from my cursory reading) has a model how media (or others) convince somebody to vote for B when his "real" preference is A. But how can we claim that he truly didn't change his "real" preference to vote for B, based on the media discussion?

Ultimately, I find it somewhat patronizing to claim that, if the real preferences are (internally in humans) expressed on numerical scale 0-100, and we ask for the number from that numerical scale 0-100, this number can somehow not correspond to real preference. I think there is then a contradiction in the definition of "preference", because then the preference has to be something more complicated, like some condition based on other people's choice, not a simple internal value that is independent of what other people want.

Re: What Is Voter Satisfaction Efficiency?

#27

Post by JumpCrisscross »

troydavis wrote: post #10
cosmojg wrote: post #2

Looking at the results, it's a wonder that IRV (aka "ranked-choice voting") gets more lip service than STAR and approval voting. These latter two seem obviously and measurably superior. My guess is that it all comes down to a lack of education. Maybe civics ought to be a required course in high school and/or college?

In Seattle, many of us reached the same conclusion as you did. We’re now trying to change Seattle’s primary elections to use approval voting: https://seattleapproves.org/ Approval voting is extremely effective at capturing voters’ preferences, simple to implement, and incredibly easy to explain - not just the ballot, but how the ballots are tabulated and how the winner is chosen. The entity you linked to, The Center…

> easy to explain

What is the simple way to pitch its advantage to IRV without using technical terms?

The spoiler effect is simple to explain because we’ve seen it. A hypothetical centre squeeze isn’t.

My belief is IRV opens the door to approval voting. (I am still sceptical of it, though I welcome experiments like the one your propose in Seattle.) When approval voting advocates oppose IRV in jurisdictions with plurality voting, however, it just splits the effort and leaves FPTP in place.

Re: What Is Voter Satisfaction Efficiency?

#28

Post by nicopappl »

vlovich123 wrote: post #17
dane-pgp wrote: post #16

It may be hard to model, but I would have liked to see how Asset Voting compared to the other voting systems. Asset Voting seems to be almost unheard of, even among voting reform enthusiasts, despite being over 100 years old and one of the simplest systems out there. Basically it works the same way as a FPTP election, except that after the votes are counted, the losing candidates one by one get to reassign their vote…

That’s an interesting concept although it does set the likely stage that a candidate pisses off their base by making some kind of personally politically convenient deal. Also the strong opening it leaves for even the appearance of corruption isn’t my favorite property. Are there real works examples of how such systems operate in practice?

Yes, Swiss federal parliamentary elections have a similar system. The name of the other parties your vote goes to if the one you voted for didn’t pass is printed on the voting bulletin.

The system is a bit more complex, since it’s for the parliament and you aren’t really voting for parties, but for lists, and the votes get transferred depending on how many seats the list filled, but the basic idea is here.

This actually influenced my voting choice. I went for a more niche party who had little chance to get a seat. I knew that if it didn’t get enough votes to get a seat, my vote wouldn’t go to waste. It would go to the larger party which already has many seats but doesn’t align with my political views as well as the niche one.

It turned out well since the niche party got a seat despite the polling predictions.

Re: What Is Voter Satisfaction Efficiency?

#29

Post by remolacha »

js8 wrote: post #26

I like range voting (of which approval is a special case). It's simple, and it always takes everybody's point of view into account (it doesn't lose any voter information), which is a really nice property IMHO. Yes, you can argue that the more extremely somebody puts the ranges, they have an advantage. (I would argue we should always normalize the cast vote so that most wanted candidate gets 100 and the least wanted 0…

Are you familiar with quadratic voting? It takes into account much of what you’re taking about, with the benefit of being less gameable.

Re: What Is Voter Satisfaction Efficiency?

#30

Post by ike77 »

It seems it formalizes giving a greater weight to the loud minority.

I'm afraid it incentives anger as an effective political signal.

Then you end up like in France, where the perception is that a protest is required for any meaningful change to be enacted.

Post reply on HN