Live data from Hacker News

Show HN: RankedVote – SurveyMonkey but focused on ranked-choice voting

rankedvote.co

301–310 of 326 posts

Re: Show HN: RankedVote – SurveyMonkey but focused on ranked-choice voting

#301

Earlier quoted context omitted.

> it happens only when you don't specify who your vote should go to if your first pick doesn't have enough supporters And your second pick, and your third pick... right? A big reason for electoral reform is to encourage a wider range of candidates to run, which means voters may be faced with having to come up with a total ordering across a dozen different options. You seem to be saying that voters who aren't informed…

> You seem to be saying that voters who aren't informed enough about their relative preferences for all available candidates, deserve to have their votes thrown away Not at all, if we get to the point where people are having to choose between and rank 50+ different candidates I might agree there's a problem, but a handful of choices simply isn't. In the Alaska election they had just three to rank, since the primary w…

> These days you could pull out your phone while in the voting booth, or have a small slip of paper in your wallet, purse, or pocket.

You're completely right about the possibility of bringing a slip of paper, and I was probably over-estimating the scale of the burden there. I would like to point out, though, that we really don't want to normalize the idea of people bringing out their phones while in the voting booth, as that seems like a way to encourage vote buying or coercion (even though the victim/voter may have ways to circumvent this problem, and even though phones aren't the only way someone could take a camera into a voting booth).

> This might surprise you, but many many people aren't terribly informed about the candidates and still manage to cast a vote under our current system.

In fact that doesn't surprise me, as I was already aware of this fact, but I see it as an existing limitation of democracy that shouldn't be made any worse. The simpler and less ambiguous the question on the ballot paper, the more of an even playing field there is for all voters. I used to think that to maximize the chances of finding the optimal winner, an election should try to demand that voters enter as much information as possible on their ballot paper, but my view has completely changed on this. Even under FPTP, the election algorithm is being fed with thousands of bits of information, and the output only needs to be roughly 1 or 2 bits of information, so there is no harm in making the ballot papers as uninformative as voters are uninformed (as long as the algorithm encourages honest votes, and combines them fairly).

> there's never been an easier time for illiterate people to get informed.

I agree, but my reference to "literacy" tests was meant to convey the fact that supposed literacy tests have in practice been used as pretences to exclude voters who were likely to vote for a party other than the one administering the test. As such, my objection wasn't that illiteracy itself was a major problem, but that any unnecessarily complicated process could be seen as having a disenfranchising effect, which would undermine confidence in elections even if the effect were not significant, or be used as an argument to prevent the adoption of the reform before it was even used, either by people who genuinely care or by concern trolls. Anyway, for completeness, here is an example of a literacy test issued by the state of Louisiana:

https://www.ferris.edu/HTMLS/news/jimcrow/question/2012/pdfs...

Re: Show HN: RankedVote – SurveyMonkey but focused on ranked-choice voting

#302
post #32

Ranking the voting systems: STAR Voting > Approval Voting > Ranked Choice Voting > Plurality ("pick only one") voting. Ranked Choice Voting is marginally better than plurality voting, but it has problems. The chief defect with Ranked Choice Voting is its non-monotonicity, whereby increasing your support for your genuine favorite can actually hurt their odds of winning. This may be what happened in Alaska [1]. STAR Vo…

> Ranking the voting systems: STAR Voting > Approval Voting > Ranked Choice Voting > Plurality ("pick only one") voting. Funny is that you ranked them and not scored them nor just say whether you approve them or not. If you have to vote for these voting systems using approval voting, which ones would you approve?

scores would be more like:

STAR or score or approval = 5 IRV/RCV = 2 plurality = 0

how i'd approve them in a real election would be based on electability, which would be true with any voting method used in an actual contentious election.

Re: Show HN: RankedVote – SurveyMonkey but focused on ranked-choice voting

#303
post #32

Ranking the voting systems: STAR Voting > Approval Voting > Ranked Choice Voting > Plurality ("pick only one") voting. Ranked Choice Voting is marginally better than plurality voting, but it has problems. The chief defect with Ranked Choice Voting is its non-monotonicity, whereby increasing your support for your genuine favorite can actually hurt their odds of winning. This may be what happened in Alaska [1]. STAR Vo…

Ranked voting is supposed to prevent winner take all strategic voting. Voting by score encourages you to score everybody at zero except your favourite(s). This arguably means there will be less information in the vote tallies, although it gives voters more control over how their opinion will be responded to, and that is valuable

this is obviously false. simply consider that green party supporters usually vote democrat. with score or approval voting, they'd ALSO want to vote for the green.

whereas with ranked voting methods, they'd still strategically rank the democrat in 1st place, crushing any hope for the greens.

http://www.youtube.com/watch?v=JtKAScORevQ

Re: Show HN: RankedVote – SurveyMonkey but focused on ranked-choice voting

#304
post #105
post #32

Ranking the voting systems: STAR Voting > Approval Voting > Ranked Choice Voting > Plurality ("pick only one") voting. Ranked Choice Voting is marginally better than plurality voting, but it has problems. The chief defect with Ranked Choice Voting is its non-monotonicity, whereby increasing your support for your genuine favorite can actually hurt their odds of winning. This may be what happened in Alaska [1]. STAR Vo…

Isn’t giving each candidate a score even more gameable than ranking candidates. If I like candidate A and think candidate B is their best opposition I can give candidate A 5.0 and candidate B 0.0 even if my real opinion of candidate B is 3.0. Lowering candidate B’s score artificially definitely helps candidate A.

no. scoring is more resistant to strategy.

consider a green party supporter who normally votes democrat. with score voting or approval voting, he also votes green (of course).

with ranked voting, he still ranks the democrat strategically in 1st place to avoid getting the republican, and that makes it impossible for third parties to grow. ranking is too vulnerable to strategy.

http://www.youtube.com/watch?v=JtKAScORevQ

Re: Show HN: RankedVote – SurveyMonkey but focused on ranked-choice voting

#305
post #105

Earlier quoted context omitted.

Isn’t giving each candidate a score even more gameable than ranking candidates. If I like candidate A and think candidate B is their best opposition I can give candidate A 5.0 and candidate B 0.0 even if my real opinion of candidate B is 3.0. Lowering candidate B’s score artificially definitely helps candidate A.

Absolutely. Naive exaggeration is a fantastic way to make your vote count more than others in score voting. It also encourages more negative campaign tactics, since convincing a voter to rank your opponent 0 rather than 4 is literally 4 times as powerful as convincing them to rank you a 5 rather than a 4.

you have this exactly backwards, as explained by 3 different math phd experts here.

http://www.youtube.com/watch?v=JtKAScORevQ

https://www.rangevoting.org/NESD

https://electionscience.github.io/vse-sim/VSEbasic/

Re: Show HN: RankedVote – SurveyMonkey but focused on ranked-choice voting

#306
post #105

Earlier quoted context omitted.

Isn’t giving each candidate a score even more gameable than ranking candidates. If I like candidate A and think candidate B is their best opposition I can give candidate A 5.0 and candidate B 0.0 even if my real opinion of candidate B is 3.0. Lowering candidate B’s score artificially definitely helps candidate A.

This is a case for approval over score voting. It's not a good case for RCV over approval!

no, it's not a case for approval voting over score voting, because score voting is better than approval voting if even one voter votes sincerely.

if enough people use honest scores, then score voting is better than approval voting even for the "honest suckers".

https://www.rangevoting.org/ShExpRes

Re: Show HN: RankedVote – SurveyMonkey but focused on ranked-choice voting

#307
post #147
post #32

Ranking the voting systems: STAR Voting > Approval Voting > Ranked Choice Voting > Plurality ("pick only one") voting. Ranked Choice Voting is marginally better than plurality voting, but it has problems. The chief defect with Ranked Choice Voting is its non-monotonicity, whereby increasing your support for your genuine favorite can actually hurt their odds of winning. This may be what happened in Alaska [1]. STAR Vo…

STAR voting can be easily gamed- tell your supporters "vote MAX for me and 0 for everyone else". Anyone who gives more than a 0 to their second choice candidate will hurt their first choice.

this is absurd. you're talking about what the CANDIDATE wants, not what voters (who actually do the voting) want.

STAR voting is shown to outperform virtually all other methods with any mixture of honest or strategic voters.

https://electionscience.github.io/vse-sim/VSEbasic/

Re: Show HN: RankedVote – SurveyMonkey but focused on ranked-choice voting

#308

Earlier quoted context omitted.

Huh? Ranked voting isn't vulnerable to bullet voting. If you only rank a single candidate, there's zero benefit, you're just throwing away the rest of your vote. It's not giving that candidate any extra influence, the way it would in approval voting.

> Ranked voting isn't vulnerable to bullet voting. "Vulnerable to bullet voting" is the same thing as "violating the later-no-harm principle," which means that some forms of ranked voting are vulnerable, while others aren't. - Borda Count is vulnerable to bullet voting, for the same reason as Score (giving points to anyone other than your first choice makes it less likely that your first choice wins, because it incre…

later-no-harm is a bug not a feature. it's the whole reason IRV is so vulnerable to tactical voting.

https://clayshentrup.medium.com/later-no-harm-72c44e145510

for instance, alaska just elected peltola even though begich was preferred to peltola by a large 5% margin. palin voters preferred begich to peltola 10-to-1. but IRV ignored their 2nd choices. therefore they would have been better off strategically ranking begich as their favorite.

whereas with approval voting, once they cast a strategic vote for begich, it's safe to approve palin (and anyone else they prefer to their lesser evil). this is, in a nutshell, why score voting and approval voting (rated, not ranked, methods) are so superior. this is social choice theory 101.

https://electionscience.org/commentary-analysis/rcv-fools-pa...

Re: Show HN: RankedVote – SurveyMonkey but focused on ranked-choice voting

#309
post #5

Hi! I actually made a similar site for my own purposes - https://poller.io - and I've been using it with friends for a while. One interesting finding. I was expecting that Instant Runoff would be the method we all preferred. I also included plurality winner, Borda Count and the Condorcet winner (if there was one) in the list of results. After using this with friends for decisions, I realized that a lot of choices tha…

KUDOS to you for respecting the Condorcet Winner. After spending a notable portion of my life diving deeply into voting systems, their side effects, and the various poorly-thought-out arguments that people make, I basically only feel wincing pain when a lot of these projects come out. So I'm glad that the concept of the Condorcet Winner hasn't been forgotten. The IRV zealots have basically won the messaging war, usin…

cordorcet methods are overly complicated and don't optimize the right thing. the goal is maximal social utility in the face of strategic behavior. cardinal methods are simpler and generally better at this.

https://electionscience.github.io/vse-sim/VSEbasic/

Re: Show HN: RankedVote – SurveyMonkey but focused on ranked-choice voting

#310
post #94

Earlier quoted context omitted.

it would be a shame to spend our effort squabbling over which alternative, than giving support to the one that has the most momentum and chance of succeeding now I think the bigger shame would be to spend so much effort promoting RCV, only to have people become jaded and disillusioned with voting method reform in general when they realize RCVs many glaring problems. The biggest problem with RCV IMO is that it looks l…

I don't see a problem in the Alaska election. Perhaps I'm missing something? I posted this in a previous discussion but I'll repost because I'm keen to hear other perspectives. From https://fivethirtyeight.com/features/what-democrats-win-in-a... > 40 percent of voters had chosen Peltola as their first choice, 31 percent had chosen Palin and 29 percent had chosen Republican businessman Nick Begich III. Under the rules…

the big problem is that begich was by far the most preferred candidate, but didn't win. and palin voters would have been better off to rank begich in 1st place. the whole thing is a mess.
Post reply on HN