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

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

rankedvote.co

Post reply on HN
Print view View original

61–70 of 326 posts

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

#61

Post by ajmurmann »

theptip wrote: post #58
kirse wrote: post #54

Earlier quoted context omitted.

Yea, it's been interesting to watch how RCV has filtered into the public consciousness over the past 5-10 years and slowly increased in terms of how many people parrot it as the answer to all our voting ills. There was a site posted to HN years back that mathematically analyzed a ton of these voting schemes, I cannot seem to find it on Google but it was some sort of hybrid scheme that apparently has the best overall…

IRV is a Schelling Point, and I think it's important to consider it as such. The biggest critique of alternatives to FPTP is "they are too complicated". When you look at an individual system like IRV it seems hard to sustain that objection, but then a bunch of voting system nerds enter the room, and now we're comparing Condorcet to Borda, and critiquing a bunch of mathematical properties of each system, and drawing g…

IRV is hard to evaluate though. How long would it take to teach a member of the general public to evaluate IRV results?

You also cannot have meaningful estimates until all results from every district are in. How long it took in New York to get any results might leave people disillusioned and opposed to anything but FPTP.

Using Borda and Condorcet is also clearly a straw man, as AFAIK nobody is actually pushing for those in practice. The real runners seem to be IRV, Approval and STAR. Of those IRV is clearly the most compared to evaluate.

IMI they are all pretty simple and in an ideal world it wouldn't matter that much. However, we need to be highly aware that as soon as a major election gets won with anything but FPTP, we'll see the losing party "flood the zone with shit" to undermine the election results and the method. We've already seen the beginning of this after the election in Alaska. We need to keep it as simple as possible, otherwise the shit will drown everything else out.

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

#62

Post by c7b »

Feature suggestion: the system automatically tells you when you have a https://en.wikipedia.org/wiki/Condorcet_paradox situation and offers to decide the outcome based on a coin flip instead.

Only half joking there, the amount of negative results about ranked-choice voting in social choice theory is quite amazing. I'm not sure whether I'd place it on the efficiency frontier of the 'complex but has desirable properties' versus 'easy to understand but far from perfect' trade-off.

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

#63

Post by bombcar »

godelski wrote: post #57
bombcar wrote: post #46

Earlier quoted context omitted.

This right here is the problem, these things get way too damn complicated for normal people really fast.

You don't think normal people can understand: "Rate this candidate from 0 to 5 stars?" If so, we really need to rethink a lot of how we ask for people's opinions, because this is probably the most common way.

They can do that - but the game theory around "don't rate the candidate you don't want ..." stuff is way too complicated.

And if it is not as simple as "Bob got the most votes, Bob won" it's going to get harder and harder to explain exactly what is going on, which is going to get people annoyed and angry.

As long as the result is the same as would have come from first past the post, nobody will care.

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

#64

Post by yellowapple »

bhauer wrote: 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…

It would be an even greater shame to be stuck with first-past-the-post while we squabble over its replacement.

I vastly prefer score or approval voting over ranked-choice, but I also vastly prefer even ranked-choice over FPTP and will be voting accordingly here in Nevada. We can always adopt score or approval voting later.

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

#65

Post by godelski »

bhauer wrote: 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…

I want to say why the cardinal systems (STAR/Score/Approval) are usually better than the ordinal (Ranked/RP/Schulze). It is all about ,,embedding''. More efficient embedding systems mean you can get a higher information gain. Let's look at two different elections but with the same premise.

Suppose you REALLY like both candidates A and B. Then suppose that you really dislike candidate C.

Ranked:

A > B > C (A/B position determined by flipping a coin)

Ordinal:

A:5 B:5 C:0

What's interesting here is that the first (ranked) system tells us (embeds) that our preference for A over B is the same as our preference of B over C. But in the second case we are able to tell the system that we actually like both A and B equally and that the distance to C is much larger. Ranked does not capture how we don't like C and doesn't capture that we do like B. Because of this the ranked systems frequently fail the favorite betrayer criteria (RCV fails btw). In a realistic scenario this means that in a ranked system Bernie would spoil the election for Biden (or vise verse). Although Gary Johnson wouldn't be a spoiler in either Ordinal nor Cardinal I'd argue that a weak spoiler is less of a worry than a strong spoiler.

I will also argue that Ordinal methods are far simpler to calculate and do statistical verification on. Ranked systems require a while loop where we pop an element from our array each iteration and reallocate that element's votes until there is a singular winner. Cardinal systems on the other hand are simply an sum argmax. This sum argmax can be done in parallel and also makes it far easier to verify elections with random subsets of votes (a fairly common practice btw). The simplicity of the algorithm used to tally votes is extremely important and increases the transparency of the election.

At the end of the day, the two systems are

Ranked: "Put the candidates in order of preference"

Cardinal: "Rate each candidate from 0 to 5 stars"

People are very used to both types of questions. I see several arguments saying the latter is more complicated but this clearly isn't true or we'd have to change the vast majority of surveys.

I also want to state that there are no globally optimal voting systems. Yes, there are plenty of flaws in cardinal systems. But because no global optima exists we have to instead argue that the tradeoffs are worth it. There are some objectively better systems than others though. To play fair like this I will say that I believe that the above advantages are worth the slight reduction of VSE in a 100% honest election that you'd obtain from using either RP or Schultz style elections.

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

#66

Post by clolege »

gamegoblin wrote: post #60

I used to be a hardcore pro-RCV supporter (I liked Yang, and Yang made RCV one of his main issues, outside UBI), but I've become a little more jaded in the last few years. Empirically, it seems like RCV-IRV (RCV from here on) not much better than Plurality. For example, look at all the countries and cities with RCV and without. It's not at all clear to me that those with RCV are much better than those without. Both N…

I recently asked Fairvote volunteers which cities/countries they point to as RCV done “right”. All of the places I’m familiar with have poor reputations for their politics (SF, Australia) or have had problems with it (NY’s democratic primary, Alaska).

All they could come up with is statistics that more women and people of color are elected with RCV, and there is not enough evidence yet that Approval voting will do the same.

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

#67

Post by ajmurmann »

fasthands9 wrote: post #59
bhauer wrote: 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…

One thing that I think gets lost in political debate around voting is that RCV also changes the behavior/selection of the candidates. Its not just that the candidates are identical and you vote with a different mechanism - but candidates have an incentive to moderate. This may mean the outcome is the same (in terms of which candidate wins) but their positions represent the median voter more. I would imagine approval…

This is also where it would be great to skip primaries, since primaries favor more extreme candidates.

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

#68

Post by gamegoblin »

fasthands9 wrote: post #59
bhauer wrote: 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…

One thing that I think gets lost in political debate around voting is that RCV also changes the behavior/selection of the candidates. Its not just that the candidates are identical and you vote with a different mechanism - but candidates have an incentive to moderate. This may mean the outcome is the same (in terms of which candidate wins) but their positions represent the median voter more. I would imagine approval…

This is actually not the case - RCV suffers from the "center squeeze" effect [1], in which moderates are squeezed out in the early rounds. That said, things like Approval have the opposite problem, in which moderates are given an advantage (some approval people will say this is a feature, not a bug).

For an edge case, consider an election with 3 factions: socialists, libertarians, and fascists. There is also a 4th candidate, a boring status quo moderate, running. Imagine the populace is split 1/3 between the first 3 factions as their 1st choice, but everyone selects boring status quo moderates as their 2nd choice, because everyone in the first 3 factions hates the other ideologies.

The boring status quo moderates get eliminated in round 1, and eventually 1 of the 3 extremes will win, resulting in 2/3 of the country being very angry. Not ideal.

Obviously this is just an unrealistic edge case, but it demonstrates the effect.

[1] https://electowiki.org/wiki/Center_squeeze

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

#69

Post by godelski »

fasthands9 wrote: post #59
bhauer wrote: 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…

One thing that I think gets lost in political debate around voting is that RCV also changes the behavior/selection of the candidates. Its not just that the candidates are identical and you vote with a different mechanism - but candidates have an incentive to moderate. This may mean the outcome is the same (in terms of which candidate wins) but their positions represent the median voter more. I would imagine approval…

That's kinda the entire point. One of the (many) metrics for voting systems is called Voter Satisfaction Efficiency (VSE). This is basically a global L2 distance from each voter to each candidate. To maximize VSE we want to minimize that distance. In other words, we want winning candidates to be most representative of a population's opinions and values. You can see a comparison of different systems here[0]. Note that there are other criteria that matter, but there are plenty of methods that have high VSE and the other criteria we want. Also note that the spread in different types of strategies represents the method's vulnerability to manipulation. If a strategy still yields a high VSE then it is not an effective strategy.

[0] https://electionscience.github.io/vse-sim/

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

#70

Post by gamegoblin »

theptip wrote: post #58
kirse wrote: post #54

Earlier quoted context omitted.

Yea, it's been interesting to watch how RCV has filtered into the public consciousness over the past 5-10 years and slowly increased in terms of how many people parrot it as the answer to all our voting ills. There was a site posted to HN years back that mathematically analyzed a ton of these voting schemes, I cannot seem to find it on Google but it was some sort of hybrid scheme that apparently has the best overall…

IRV is a Schelling Point, and I think it's important to consider it as such. The biggest critique of alternatives to FPTP is "they are too complicated". When you look at an individual system like IRV it seems hard to sustain that objection, but then a bunch of voting system nerds enter the room, and now we're comparing Condorcet to Borda, and critiquing a bunch of mathematical properties of each system, and drawing g…

A group in Seattle actually got an Approval voting ballot initiative enough signatures to get on the ballot this year (where RCV has been trying and failing for decades), but the City Council, sensing their re-election chances under such a system in danger, intervened and modified the initiative to add RCV as a 3rd option, because they know they have a better election odds with RCV, and RCV will probably beat Approval on name recognition alone.

A cynical take would be that IRV has gotten traction in the US precisely because it does not seriously threaten the existing power structure. It might nibble away on a few edges, but if you look at all the countries and cities with IRV, they are not obviously better than those still stuck on FPTP.

Post reply on HN