What's great about RCV (not IRV specifically) is that if the ballot data sets are retained, you can write software to recount them using other RCV algorithms. I'm curious when we'll next have a Condorcet Winner that isn't the IRV winner.
Nope! This is a problem with high-cardinality ordinal voting methods. Since factorial is a very fast growing function, you can easily encode a pattern in down-ballot candidates and buy and verify a specific ballot.
Commonly what's done is that we truncate the ballot and only allow people to express, say, 5 preferences.
I don’t like rank choice voting because it can result in someone with fewer first place votes than the first or second candidate to win if the conditions are right.
We’ve seen surprise results where an obscure candidate has won when too many top choices split the first choice votes and someone relegated to the floor vacuums up all the throwaway votes (people think they have to “spend” all their votes)
> an obscure candidate has won when too many top vs fixates split the first choice votes and someone relegated to the floor vacuums up all the throwaway votes
That’s the point. The compromise candidate won. If the top candidates had spent less time mud slinging they may have remained in others’ top two or three.
> approval voting is a simpler system that also tends to reward moderate or "compromise" candidates
I don't have a strong preference between approval and RCV. But I think people closer to politics would. Being able to "be behind" one candidate is a meaningful promoter of civic engagement; forcing that decision is deeply entrenched in our politics.
That makes approval voting a tougher sell for implementation than RCV. Given how much better RCV and approval voting are than FPTP, it seems sensible to go with RCV.
Implementation is actually a lot simpler. With approval voting, there is no change to the physical paper ballot; you just allow the voter to punch multiple holes on a single ballot, to represent the candidates she "approves" of.
That's not the phase of implementation that's difficult with any of these systems. It's getting the rules changed. Approval voting changes civic engagement in a fundamental way. (It also has much less real-world data behind it.)
> That makes approval voting a tougher sell for implementation than RCV.
Approval voting was passed by a 2-1 landslide majority in both Fargo and St Louis. Public opinion research by the Center for Election Science shows random American voters are much more supportive of approval voting than IRV, relative to the status quo. By a huge margin.
As I explain elsewhere, the problem is you can have a a pool of good candidates who SPLIT/dilute the top votes (and none win) and then you can have the poor candidate who does not split the vote and all the secondary votes go to this unlikely candidate and that candidate wins... to the chagrin of most voters (who cast votes for the top contenders).
You're last name doesn't have to be Arrow to know that there's a potential for a non-compromise candidate to win, indeed for the candidate that a majority don't want in office to win.
> there's a potential for a non-compromise candidate to win
Agree IRV isn't the perfect system. But we have the downsides of being broadly explicable (Condorcet method's weakness), continuous with existing civic culture (approval voting's weakness), squeezing the centre (RCV/IRV's weakness) and being better than autocracy (plurality's weakness).
With some irony, the game of reforming voting systems is itself a plurality mechanism. IRV, re-branded as RCV, has taken off. It has precedence. The distance between each of RCV's competitors and it is dwarfed by the distance between each of them and FPTP. Given that, unifying around the winning "candidate" makes sense.
Put another way, bringing up approval voting in an IRV vs. FPTP discussion favors FPTP. If one wants a world in which approval voting reigns, I think it would be easier to get there with IRV as an in-between.
I've never really understood the criticism of Condorcet's method being hard to understand. If one candidate would beat all other candidates head-to-head, that candidate should be the winner. That's really it. People get hung up on the loop-breaking algorithms when they're not really the point.
> I don’t like tank [sic] choice voting because it can result in someone with fewer first place votes than the first or second candidate to win if the conditions are right.
Because the way people understand it is that if candidate #1 doesn't win, then candidate #2 would win, maybe candidate #3, but not candidate #5. And people are often disappointed to learn that the first, second or third, 1st choice vote-getters are not the winners but rather the #1 3rd, 4th, or 5th choice vote getter can end up winning (again because people think they HAVE to cast all votes --which obviously they don't, but that enables the possibility of the throwaway candidate winning)
The rest of the comment seems to explain that the problem is that poorly considered candidates win because people don't give as much care below the top spot in a wide pool of candidates:
> We’ve seen surprise results where an obscure candidate has won when too many top choices split the first choice votes and someone relegated to the floor vacuums up all the throwaway votes (people think they have to “spend” all their votes)
I think its an understandable concern under any single round ranked-ballots method where you arr likely to have a large pool of cnadidates. But it can be simply remedied with a two-round system using the same method iterated to select a manageable pool of candidates (possibly gating in the top couple first-place vote getters even if they otherwise wouldn’t be in the top) for the runoff round, which still uses the same ranked-ballots method with more confidence that all the candidates have been vetted because there is a narrower field.
This keeps the integrity of whatever ranked-ballots method is chosen while avoiding the (small, but not nonexistent) risk of winning by flying under the radar.
> I don’t like tank choice voting because it can result in someone with fewer first place votes than the first or second candidate to win if the conditions are right.
Any system other than plurality allows a candidate with more first place votes to beat one with fewer. Any system other than plurality or majority/runoff allows a candidate with fewer first place votes than the second place.
But, much as I dislike IRV, I’m rather skeptical of the “winner should always be one of the top two by first-place vote count” criteria suggested.
OTOH, using any ranked ballots method, I can see an argument to do a wide open “primary” using the method and then if the winner is not also the majority first-place winner doing a separate “runoff” election using the same system of, say, 5 runoff candidates, determined as:
(1) the candidate that won using the ranked-ballots voting method selected
(2) any of the top two by first place votes not selected by (1)
(3) the first candidate elected by the selected ranked ballots method when the candidates admitted to the runoff by (1) and (2) are disregarded, plus enough candidates elected by the same method disregarding all previously-admitted candidates to complete the pool of 5 candidates.
The idea being to deal with the risk of “candidate was elected due to insufficient exposure and vetting” risk, without fundamentally compromising the desirable behavior of the selected ranked-ballots method.
Would that be meaningful when the method is known to the voters ahead of time? It’s like the team with fewer total yards winning the Super Bowl... wouldn’t be that surprising since both teams know that it’s points that decide the game.
I suppose some meaning could be lost if people are voting tactically from the awareness that they're in an IRV election. But on the other hand, if a voting system's implementation incentivizes voters to vote against their true preferences, then that by itself should be an indication that the voting system is flawed. (It's not true that every voting method incentivizes tactical voting; this is a common misreading of Arrow's Theorem.)
I agree that Arrow's Theorem is often misread (as it doesn't apply to score-based voting systems, or single-winner elections) but it's perhaps more important to mention the Gibbard–Satterthwaite theorem here, which is similarly misrepresented.
That theorem effectively says that all sensible voting systems will in certain conditions allow a voter to gain a better outcome by voting insincerely (i.e. tactically). However, it doesn't say how easy it is for a voter to determine the correct tactical vote to make, so potentially a voter would need perfect information about all the other votes cast in order for their tactic to succeed.
Thank you, I was really confused by this because the NYC borough of Brooklyn is actually synonymous with Kings County, and I assumed that OP had merely forgotten the s.
Without MMP it's not going to change who gets in. The 49.9% of people opposed to Kshama Sawant get no voice (yes, that's city of Seattle, but easier example for non-locals) in IRV(RCV) nor FPTP.
Tacoma did this a few years ago and nothing came of it.
This "pyramid of control" observation is extremely under-appreciated. We are living in a time of extreme polarization and division, not just between parties but within them, as societies try to deal with an increasingly complex world where shifting battlefronts no longer align with traditional ideological boundaries, all while we're being mesmerized by the novel psychological environment that social media creates.
Even if the "pyramid of control" doesn't drive parties towards extremes, it perhaps drives them towards memes; that is, simplistic but clear ideas that stand out among the noise. Unfortunately, while that might be enough to create a coherent identity for a party, it doesn't do anything to guarantee that the party has any sort of broad support from the population, which further alienates people from politics.
A weird take on the U.S. constitution: it actually established 4 "branches" of government. Citizen oversight was one of 4 powers designed to keep government in check. It's hard to look at today's America and feel like citizens have much oversight of government. Our "branch" has been systematically eroded. The current first-past-the-post system has been gamed heavily by two parties to reduce the impact citizens have on the outcome of elections.
RCV helps restore civilian oversight (many forms of voter suppression, i.e. gerrymandering, are less effective under RCV). My current thinking: feeling this will result in more "left" or "right" leaning governing bodies is a reflection of how citizen's want their government run. If the "silent majority" is right leaning, this will result in right leaning politics; left would yield left. But the "leaning" of the politics is less important to me as an outcome. The most important outcome is that citizens are restored to their rightful role in government: acting as a proper "check and balance" to the other 3 branches.
You end up with Candidates who are neither principled nor charismatic but have blessings of Party Boss and fits in a narrow zone of like/dislike factor.
I imagine less. In primaries you attract voters that are more fanatical and extreme so they have a better chance of pushing through a candidate they see as ideologically pure. When the opposition party has no chance of winning it becomes the de facto election and you have a better chance of ending up with polarizing candidate.
NYC isn’t as super progressive as people think but there is certainly a progressive streak and they are louder than their numbers. They organize and they vote in primaries and that has been their strategy the last decade to assume outsized power. Because many moderate liberals won’t likely vote for a Republican in today’s climate it works. Hence BDB as mayor whom is despised by moderate liberals.
So this should redirect power to the Democratic machine in NYC and make it more difficult for far left progressives who really been hijacking the Democratic Party by virtue of the primary system.
* You can vote for the random third party candidate without worrying about throwing away your vote.
* You seem to get a bit more moderation where we are if you also have an open primary / no primary type voting which gets really nice.
California does top two past the pole of any party inf the general which means even republican votes can matter in a dem on dem election (common out here) or dem votes can pull back things in a R vs R general (rarer but does happen).
In RCV your first point is not actually true. Its better then First Pass the Post (like literally anything else). Score based voting actually does have the effect you want.
Ranked Choice Voting is not really a good idea. It has very strange unintuitive behavior.
Doing Score based voting is just much better and arguably simpler and less work.
I specially like the Star-Voting, where they do a runoff at the end to cover some extra cases. They also have a few variants for multi-winner and representational voting.
Its also much more familiar for people from real live, like online review.
P.S: Wow a lot of down-votes. RCV is just the alternative that got the most marketing, but once you start looking into the different systems I really don't understand people can stick with RCV just because its slightly better then the system before. If you are gone invest all this effort in improving the system, why not go to something easier and objectively better.
This video shows the issue very well (one of the people involved in Star voting made it):
Do you have any data around how many people in practice rank candidates between 2 and 4 (assuming a 1-5 scale)?
Given how polarizing politics tends to be, especially in a two party system such as America, I’d wager that most people put all 5’s for candidates of their party and all 1’s for candidates of the other party.
But the thing is that's beside the point. Even if they do that, it still allows them to express approval of multiple candidates, instead of just one in plurality voting. And the fact that there's no weird elimination rounds means that:
- it's easier to teach people ("most votes wins")
- it's easy to double check the results ("most votes wins")
- it's incentive compatible (you never put someone you hope loses above someone you hope wins)
Even if you never use 2 – 4 on a 1 – 5 scale, it's still better because you're making the same decision for multiple candidates.
Only using the max and min scores in score voting / range voting has a name: approval voting. For each candidate, you're answering the question: "Would I be ok with this person being elected?"
I don't have data at hand as I have not researched this topic in a long time, but from everything we know, people tend to not do only min or max. From everything we know, the majority of people don't vote purely strategically, specially when the voting system gives you serious disincentives to do so.
There is a clear intensive not to do that as it seriously hurts your overall preference. And Star Voting makes this even stronger compared to normal score voting because of the run-off.
Even if you assume that everybody is a 100% total party zealot it still only breaks down to approval voting and that is still a pretty good system.
> Given how polarizing politics tends to be,
Politics is polarized because of the voting system, if you look at actual data of preferences then you would see that there is far more agreement then can be expressed.
> I’d wager that most people put all 5’s for candidates of their party and all 1’s for candidates of the other party.
You would lose that wager. Most people are not party activists.
NYC is an interesting example to use here because city and state politics are almost wholly dominated by one party. Instead of general elections being a Republican vs Democrat contest, you (in general) see elections around here being decided in Democratic primaries, along a spectrum that is generally far-left to center-right.
The presumed next mayor of NYC is going to be decided within the democratic party primaries today (or over the coming weeks as rounds of RCV progress), not in the general election later in the year. There's still a major spread of ideologies involved, but they're clustered within one party - which seems to have a lot of people thinking tactically about their votes, instead of just going straight-up party line.
Giving 5s to your preferred candidates and 1s to your disliked candidates is not necessarily a problem, as others have pointed out, that's just approval voting and a good technique in its own right.
If you want to avoid that, the runoff at the end of a STAR (Score Then Automatic Runoff) is explicitly designed to incentivize voters to provide more differentiation between candidates.
Well, maybe. Or disillusionment with IRV might spoil the waters for other attempts at reform, when people start to realize all the claims about ranked-choice eliminating the spoiler effect are actually false.
Certainly that's what happened in Burlington, VT where a spoiler candidate swung the 2009 mayoral election, and IRV was subsequently repealed. They didn't say "oh oops, well this experiment failed, let's try approval voting instead"; they just went back to plurality voting and that was that.
> where they do a runoff at the end to cover some extra cases
One major motivation for the runoff at the end in techniques like STAR (Score Then Automatic Runoff) is to encourage voters to give different scores to candidates (assuming they actually feel differently about them) rather than strategically voting all 5s and 1s and basically making it the same as approval voting.
Very weak evidence offered but the motivation behind the article was near the top.
"They also appeal to more ideological voters—especially on the left—by arguing that they can express their views with more precision in a ranked-choice system."
As I'm guessing you know (but for the benefit of others), although the WSJ news section is rated quite middle of the road, their editorial board is rated as right-leaning as "The American Conservative", "Newsmax," and Fox News online news.
I personally think none of them are good enough points, they can all be dealt with. And 3 just means you can vote for who you like best, not just the ones in power.
Yeah, the idea of "increases existing divides" is extremely bad faith.
It lets similar candidates make a case for their small differences, but allows:
* voters to not waste their vote on one of 10 different similar candidates, hoping they picked the one with the most backing so they don't get a split vote
* more than just two or three candidates to run and have a reasonable chance of winning
The more people running, the better. Sure, there's some level where it gets a little over the top, but there are safeguards in place for that like minimum number of signatures from verified residents of a city/county/state, so there's currently extremely little risk of RCV being enacted and 100 candidates showing up on ballots.
And even if that did happen, so what? Look up a few of the candidates and pick the one that seems to represent you best.
The WSJ editorial board’s argument is very weak, and seems very partisan. They cherry picked a few very strange, rare scenarios that would still not be any different with the current first past the post system.
Ranking things in order of preference is something we do all the time. “They’re out of my favorite fruit, but I’ll get this other one instead.” “You don’t have that chair in this color? Okay, I’ll take that other color.” etc
The WSJ editorial board says, “Major parties could be weakened to the benefit of more extreme candidates.” That already happens with first past the post. The 2016 GOP primary season for President is an example of this: the eventual GOP candidate did not win 50% of the vote in the majority of 2016 primaries. Instead, more GOP voters wanted a more moderate candidate, but the many moderate candidates split the vote. Thankfully the WSJ article points this out, but totally undersells it, and doesn’t make a strong case to at least use it in those situations.
RCV reduces the chances of an extreme point of view getting elected, because they have to be able to get approval from at least 50% of everyone voting, rather than relying on their opponents to split the vote.
Get more people running, so we have more options to vote for, and we can express more accurately who and what we support. RCV, or one of the many other alternatives to FPTP, is the best shot at doing this, and in my opinion one of the few ways to truly improve the US’s democracy without a bunch of massive changes that are even less popular with politicians (limiting donations, lobbying, etc).
If you want to get involved bringing RCV to vote in your state, all 50 states have an RCV movement. You can find and join your states movement here: https://www.rankthevote.us/states
It frankly forced me to research more candidates than I would have. The amount of familiarity I have for every e.g. City Council candidate (which were RCV) versus DA candidate (who were standard voted) is night and day.
IIRC in some applications of ranked choice voting it's possible to rank an arbitrary number of choices. So even voting the old-fashioned way would be fine.
Great to see this on Hacker News! I'm a former YC Founder who's been working with NYC on voter education for this year's elections through RankedVote (https://www.rankedvote.co).
Do you think there is a need to educate voters on how to tactically vote in ranked choice voting? Every voting system has strategies to use it effectively and most voters are not used to the tactics necessary for ranked choice voting.
We see a lot of education in how to use the ballot simply, but very little education on "advanced tactics".
In particular, a lot of NYC voters aren't ranking either Wiley or Adams, which is a huge mistake as those are very likely to be in the final round.
Thanks! The idea is to use software to promote this reform (as well as Final-Five Voting...which is Open Primaries + RCV)
As for advanced tactics, if you're really into it, Rob Richie, the CEO of FairVote (leading national advocacy org) put this in the New York Times over the weekend. It spells out a bunch of scenarios (e.g. "I want Garcia to win and Stringer to lose.") and how to tactically vote for each.
BUT the point I'd want to drive home is that ranked-choice allows you to be FAR more able to "vote for who you believe is best" than in the more common most votes wins approach (where you have to be strategic the moment a 3rd candidate enters).
Yeah, that article is exactly what I was thinking about, but I think it would be better if it mentioned more general tactics (such as the inherent advantage of making sure to rank front runners since they will last longer).
Once this NYC election is over we should look at the voting records and see how many votes were lost due to poor tactics (forgetting to rank front runners).
Huh... yeah, so this pizza ballot (which I'm just going to have to assume is similar to the real ballot, as otherwise I'm not sure why they are doing this ;P) definitely isn't at all what I was expecting, as I was at least expecting to get to rank all the candidates; I can definitely see some weird effects happening with a ballot like this if people have to decide to use up ranking slots on people they dislike just so they can provide comparative rankings between them as maybe they'd end up in the final round. I mean, wasn't the entire point of this that I'm supposed to get to just vote for the people I like in the order I like them, rather than having to second guess stuff like "well, I hate both of these people, but since I hate this one less than that one and I bet a lot of people like both of them I'd better rank at least one of them above a candidate I prefer"? (And yes: I appreciate the trilemma that says that no voting system is perfect, but truncating the rank mechanism is seeming to leave some of the goal sitting on the table.)
> Every voting system has strategies to use it effectively and most voters are not used to the tactics necessary for ranked choice voting.
Just to respond to this one small point, but I think the most common mathematical definition of "fairness" in market designs (including voting mechanisms) is that your utility-maximizing action should be identical to your true ranked preferences (this is called "strategy-proof"). In the case of voting, Gibbard-Satterthwaite says that there's no strictly strategy-proof mechanism (under a few restrictions) but I think the instant-runoff voting which NYC is using is mostly strategy proof (i.e. strategies only exist in rare circumstances).
Fwiw the burden of learning about all these candidates seems high to me, but apparently New Yorkers don't ind.
> I think the instant-runoff voting which NYC is using is mostly strategy proof (i.e. strategies only exist in rare circumstances)
The very comment you are replying to pointed out that "a lot of NYC voters aren't ranking either Wiley or Adams" which is a failure to apply strategy where it manifestly exists. "There are front runners" doesn't seem to be a rare circumstance.
Well, you're right, but we mean different things. These voters are failing to rank their true preferences (you and I believe), which is going to lead to a less optimal outcome than if they had. But if they did rank their true preferences (i.e. they are ok with Wiley or Adams, but just at the bottom of their list), then it would be fine.
This is a valid communication problem with ranked lists, but it's not that there exists a "strategy" per se besides ranking all of your acceptable outcomes.
> These voters are failing to rank their true preferences (you and I believe)
I expect they are being honest about their preference order, but that this has less of an impact on the outcome (in whatever direction they desire) than if they had been dishonest about their preference order.
This does assume that they actually prefer one of the front-runners to the other by a more-than-negligible margin; perhaps that isn't the case.
> This is a valid communication problem with ranked lists, but it's not that there exists a "strategy" per se besides ranking all of your acceptable outcomes.
There were more than six candidates, and voters could only rank their top five. Voters with more than 5 "acceptable outcomes" need to vote strategically when they get near the bottom of that list.
Alternatively, there is also no such thing as "strategy" in FPTP - "you just vote for one of your acceptable outcomes". That's... not a useful definition of the words involved.
Edited to add: It occurs to me that perhaps you weren't aware of the cutoff in the NYC case and were speaking of IRV with a full list? In that case I agree that strategy being necessary is substantially more rare, leaving aside for the moment whether it is sufficiently rare.
Why do you push IRV rather than simpler methods with higher VSE like Approval or STAR? Shouldn't we be highly advocating for a system that does not fail the favorite betrayer criterion? And I think Arizona is the perfect example why transparency and low complexity is essential (higher VSE is an added bonus given these).
Simple...because it's being used in a significant (and growing) number of places. Approval and STAR are great methods as well. But they're not what's in use in Maine. Or NYC. Or will be in use in Alaska.
So, in the spirit of focusing where I can have the most impact, I've chosen to most directly support RCV.
First past the post is the enemy here. Not RCV, STAR, or Approval.
> because it's being used in a significant (and growing) number of places.
Which is because of advocation over the last (mostly) 2-3 years. We're only two states in, that's not significant momentum. FYI, we have approval in Fargo and Lane County/Eugene are close to passing STAR (I'd argue RCV hinders that effort due to confusion).
> First past the post is the enemy here. Not RCV, STAR, or Approval.
I'd disagree actually. The enemy is non-representative voting.
Why I'll actually fight IRV is because history. We've gone down this path before and had states/cities/counties use IRV and then revert back because it didn't solve the problems they set out to solve. So, what is different this time around? There's a "good enough" bar that needs to be met. IRV doesn't meet that.
IRV is only better than FPTP in a limited set of circumstances, and can often be worse at selecting the winner.
Most importantly, IRV will not break us out of the two-party system, which is the root of many problems in American politics.
Are you aware that IRV significantly increases spoiled ballot rates, and disproportionately in low-income areas? It's actively harmful, roughly equivalent to knocking a couple percent off of the Black vote.
Please reconsider your support for it. I am sure you have good intentions; my guess is that most IRV supporters probably aren't aware of the harm they're doing. I also thought IRV was an okay system (not perfect but better than plurality etc.) until I learned this information relatively recently.
> Most importantly, IRV will not break us out of the two-party system, which is the root of many problems in American politics.
I want to expand on why this claim is true, since IRV is frequently propositioned as a way to solve spoilers. Most importantly IRV does not pass the Favorite Betrayer Criterion[0] which states
> voters should have no incentive to vote someone else over their favorite
IRV only handles weak spoilers (e.g. Jo Jorgensen spoiling Trump) and not strong spoilers (e.g. Sanders spoiling Biden).[1] We're not actually concerned about Jorgensen spoiling Trump or Stein spoiling Biden. We're concerned with Sanders spoiling Biden (or vise versa). This is where the favorite betrayer criterion comes in.
IRV isn't even the best ordinal (i.e. ranking candidates) system because it fails the monotonicity criterion[2], which states
> A ranked voting system is monotonic if it is neither possible to prevent the election of a candidate by ranking them higher on some of the ballots, nor possible to elect an otherwise unelected candidate by ranking them lower on some of the ballots (while nothing else is altered on any ballot).
This is an extremely undesirable property and will continue to promote tactical voting. I can go into it more but maybe just check out Clay's works or ask him yourself since he's here in the thread. He's far more informed on this stuff than I am since he's an actual expert.
If anything, this "momentum" argument says you get more value by supporting STAR voting and/or approval voting than a method that already has traction.
I have a strong preference of cardinal over ordinal systems, but I won't say all ordinal systems are flawed. I rather like condorcet methods but I think their complexity is too much of a barrier. I've started to come around to Election Science's similar decision of pushing for Approval over STAR because of this and I will gladly argue how STAR is far less complex than IRV (STAR has a max of 2 rounds and IRV will almost always have significantly more. The only time it has two rounds is if there are strictly 3 candidates).
I joke about a conspiracy theory of this being the reason that IRV is propositioned instead of a more representative and better system like cardinal voting but honestly I believe that the reason is just because it gained more momentum early on. There's been CGP Grey videos on IRV (which proposes the exact misunderstandings I'm discussing in these threads), Hasan Minhaj (ditto), and others. There's been no major celebrity that is advocating for cardinal style voting. And at the end of the day understanding why cardinal is better than ordinal is not trivial to understand, unless you understand information encoding but that's not common knowledge (though it should be here and I'm surprised it isn't).
> And at the end of the day understanding why cardinal is better than ordinal is not trivial to understand, unless you understand information encoding but that's not common knowledge (though it should be here and I'm surprised it isn't).
What would you say to someone like me who feels that IRV is perhaps the worst possible alternative voting system to advocate for? It feels like someone at some stage must be very dishonest, or otherwise dangerously uninformed, to think that IRV is worth advocating for over alternatives like approval voting, range voting, or any Condorcet voting system. I'm very worried that most places will have the political will to improve the voting system only once in a century and we'll have wasted it on a system that's unusually ill-behaved. I'm particularly concerned about IRV's non-monotonicity, whereby it's possible to hurt a candidate by ranking them higher, and likewise it's possible to help a candidate by ranking them lower. How can anyone feel they're voting honestly in an honest election when this is the case?
Honestly IRV is _even worse than plurality_. It doesn't solve the problems it sets out to solve (it entrenches two-party domination [1]), it has ridiculous monotonicity violations [2], all for a lot more complication in counting the votes (you can't distribute the counting well without transmitting the contents of all of the ballots) and possibly wrecking the secret ballot (you can encode and buy specific down-ballot rankings).
Seriously, it's all of the disadvantages and very limited upside.
> you can't distribute the counting well without transmitting the contents of all of the ballots
There are only so many combinations of rankings that are possible. It shouldn't be hard to encode each of those possible orders and transmit them as a whole.
> There are only so many combinations of rankings that are possible.
O(n^2) isn't great though...
The algorithm to count is rough and requires many rounds (since you count, knock out the lowest candidate, recount, and repeat until a >50% favor is achieved by one candidate). Cardinal systems on the other hand just require summing the columns and taking argmax. This is far simpler (in fact we can do most of this in parallel making a far better runtime).
Two reasons: Arrow’s impossibility theorem and the fact that people are not perfect logicians.
I honestly thought that after learning Arrow’s impossibility theorem that Condorcet is not especially important. Since we can’t have a perfect voting system, we have to pick an imperfect one, and among the alternatives, it’s important to capture information from voters. Instant runoff captures a lot of information.
Approval voting I think is much too tactical and, strictly speaking, worse than IRV. No contest.
IRV lets people express their preferences in a fairly understandable way. The strategy I see people talking about is “rank all the front-runners, and use the leftover spots to rank the people you want to win, even if they’re not likely to win.”
So if A B C and D are front-runners, and E and F are the two other candidates you like, you come up with a ranking for those six and put the top five on your ballot.
The idea that people who like this system are “dangerously misinformed” or “dishonest” is needlessly inflammatory.
First off, Arrow's doesn't apply to all systems. You'll need to look into both Gibbard's and Gibbard-Satterthwaite's theorems.
Second off, just because you can't find a global optimization in a highly dimensional space doesn't mean there aren't local optimizations along criteria we care more about. Appealing to Arrow's is a cop-out.
> Approval voting I think is much too tactical and, strictly speaking, worse than IRV. No contest.
You're going to have to back this up with some strong evidence. Approval has higher VSE, is simpler, is more resistant to spoilers and tactical voting.
> IRV lets people express their preferences in a fairly understandable way.
Actually your argument holds true for any ordinal or cardinal system. Cardinal even having more flexibility since you can give two candidates the same score. And in cardinal if you want to rank your candidates, no problem. Better yet, you have better encoding opportunities because you can specify the distance between your ranking instead of the uniform spacing that ordinal systems force upon you. (BTW, given what the person above you wrote, I would assume that they know how ranked voting works and explaining how it is going to come off as you calling them dumb).
I disagree that cardinal voting is understandable. It’s how we rate restaurants and review products on Amazon, and I don’t think it translates to an election with multiple options.
The issue here is not just how logical humans “homo economicus” behaves, but how actual voters behave.
Not really interested in engaging with the rest of this comment right now, but suffice it to say that I don’t think you’re accounting for human behavior in practice, which is messy and illogical. I don’t think it’s reasonable to say that I’m appealing to Arrow’s impossibility theorem as a cop-out—I’m saying that since we don’t have a game theoretic solution the the problem, we should look at the actual behavior of imperfect, irrational humans as the deciding factor.
> I disagree that cardinal voting is understandable. It’s how we rate restaurants and review products on Amazon, and I don’t think it translates to an election with multiple options.
I disagree but also don't see this as a problem. If you rank candidates you still get the majority of the desirable properties. Rank with non-equal distances, even better. Hell, it isn't even bad if you bullet vote (that's just approval voting). Investigating non-optimal ways of voting under any voting system is an extremely important analysis. So for the voter there is no problem. I'm also kinda put off that you give real world examples of humans using cardinal methods and then claiming that we can't understand it (HN is using cardinal voting...)
But we also have to consider the counting of votes side of "understandable." Plurality is pretty damn easy, and this is clearly why we use it. Approval is almost as easy (just just sum multiple columns). Range/Score isn't much harder. Then STAR introduces 2 rounds of counting. Then we look at IRV and we see that we have tons of rounds. This isn't typically so bad in a presidential election where there are realistically about 4 candidates, but that complexity increases real fast elsewhere. Just watch NYC. There's going to be at least 5 rounds (probably more). This is far more complex. We only have to look at Arizona to understand why this part of the "understandable" question is important.
One big problem with approval voting is that it presents voters with a difficult conundrum: what do you do with candidates you don't particularly like but would still strongly prefer over one-or-more other candidates? If people are too lenient with their approval it increases the risk of someone no one really likes getting elected over someone a majority would have preferred. If people are too stringent you start running into the same problems as FTTP.
Approval voting has some nice mathematical properties, but I think in practice trying to pidgeon hole people's preferences into a binary decision would be a major source of voter frustration and lead to tactical voting.
Fine, go STAR or Score/Range. Honestly I prefer those systems (in that order and I'd argue most people that are pro cardinal systems have that same preference[0]). Nice benefit is that people can bullet vote and we collapse to Approval which is a "good enough" system.
> but would still strongly prefer over one-or-more other candidates?
In fact, this is why I argue for STAR or score. It encodes information for better than a ranked (ordinal) system. In any ranked system you encode your preference with equidistant from one another. Where as when you score you can indicate a much stronger preference.
Here's an example. Let's say I REALLY like candidate A, moderately like candidate B, and strongly dislike candidate C.
Ranked:
A > B > C (with my encoding I'm saying that my preference of A over B is the same as my preference of B over C)
Scoring
A: 10, B: 7, C:0 (with this encoding I can indicate that my preference of A over B is not as large as my preference of B over C. Obviously we are capturing more information here)
Let's just encode information better, I agree. But also let's consider other factors like how easy it is to count the votes (which every cardinal system is going to beat ranked systems).
[0] That same preference where the distance between STAR, Score, and Approval is smaller than the distance between preference of Approval over IRV (e.g. STAR: 10, Score: 8, Approval: 7, IRV: 3, Plurality: 0).
Small correction: The term "bullet voting" means voting for one candidate only (001000), rather than voting 0 or 1 on each candidate (101011). It is caused by low engagement from the voters, who only take the time to learn about one candidate, their favorite. It has been a problem for approval voting in practice, and it is unclear to what extent it affects score methods in practice.
I think STAR is slightly worse than Score, is comparable to Approval, is much better than IRV, and is likely better than Condorcet methods. STAR has some odd behavior which can be explored in a 3-person race. Score has less-problematic behavior caused by risk-taking with equilibrium voters (not like voters behave in any way similar to Nash equilibria though, and who knows what the real-world behavior will be).
However, STAR's main benefit may be in overcoming political resistance, if its properties are simpler to convince voters. Majority criterion sounds nice even when it is inefficient. Much like Top Trading Cycles losing to Gale-Shapley in school choice algorithms (excluding Boston).
> However, STAR's main benefit may be in overcoming political resistance, if its properties are simpler to convince voters. Majority criterion sounds nice even when it is inefficient.
Out of respect for you taking the time to respond, I will elaborate my claims more specifically. My assertion regarding bullet voting is isolated to elections where the voting body cannot be bothered to engage carefully, because they barely care about the race. In addition, the drawbacks must be then compared with other voting methods, which may face the same bullet issue in equal measure. The bullet voting comparison I am making is between Score and Approval; I think Score will face less bullet voting than Approval does.
A typical example where bullet voting should occur is a down-ballot election. Examples of low-engagement elections at https://en.wikipedia.org/wiki/Approval_voting#Other_organiza... bear out that these elections degrade in behavior to plurality. However, it can't be determined just from this behavior that IRV or Condorcet would do any better.
1-3 have major engagement from voters and I do not expect approval voting to suffer unduly from bullet voting. 4, an alumni association, should. However, the question then becomes, what about other voting methods? Because if the voter has weak information on the other candidates, he may simply bullet vote under all voting methods. In this context, Score needs testing. Voting values between 0 and 1 would give a much better idea of how voter engagement is affecting their votes, rather than their second choice always collapsing to zero. This is why I claim that it is "unclear to what extent it affects score methods in practice". Granular cardinal scores in even one such election would mean a lot for determining voter behavior.
I do not support FairVote's arguments and consider them dishonest. Notice that in my original post, I carefully describe bullet voting as a consequence of low engagement, which is supported by past real-world low-engagement elections. This is very distinct from FairVote's arguments, which are not sophisticated and demonstrate a willingness to make blind claims. The distortions that FairVote argues for require more justification than they give.
I don't believe much in that simulation. While I agree with your characterization of it as the "best computer modeling" in voting theory, it would still be considered a fatally flawed paper under the standards of most fields, and I faced some heavy obstacles when trying to analyze it. I think this field needs more/better academics. I described some of my objections at https://news.ycombinator.com/item?id=27612876
> In addition, the drawbacks must be then compared with other voting methods, which may face the same bullet issue in equal measure.
A robust cross-system analysis shows that approval voting is more robust to strategy than almost any other method. Most ranked voting methods, for example, fail the favorite betrayal criterion.
1. Extensive game theoretical analysis, and even computer modeling, has shown that approval voting resists tactical behavior better than virtually any other voting method.
There's even an entire book focused on the game theory and tactics of voting methods, which advocates score voting, approval voting being score voting on a 0 to 1 binary scale.
Approval voting elections have been successfully held in 2020 and 2021 in Fargo in St Louis respectively, and there were no indications of voter confusion or anything like that.
For anyone like me who hadn't heard of Gibbard's theorem, it's actually even simpler (and more depressing) than Arrow's theorem. To quote Wikipedia:
For any deterministic process of collective decision, at least one of the following three properties must hold:
1. The process is dictatorial, i.e. there exists a distinguished agent who can impose the outcome;
2. The process limits the possible outcomes to two options only;
3. The process is open to strategic voting: once an agent has identified their preferences, it is possible that they have no action at their disposal that best defends these preferences irrespective of the other agents' actions.
So basically, it's impossible to completely eliminate strategic voting. No matter the method: ranked vs. cardinal vs. anything else you dream up can't help.
>Gibbard's theorem states that a deterministic process of collective decision cannot be straightforward, except possibly in two cases: if there is a distinguished agent who has a dictatorial power, or if the process limits the outcome to two possible options only.
.. it also mentions that gibbard's is specifically about irv.
All that aside, I can't understand the idea that we only get to change things once. My understanding of history is at odds with the concept.
> .. it also mentions that gibbard's is specifically about irv.
No, Gibbard's is not limited to any particular voting method. I think you're misreading the next paragraph (and also confusing IRV, which is a particular method, with ranked choice, which is a whole category of methods). Note the distinction between Gibbard and Gibbard-Satterthwaite:
> A corollary of this theorem is Gibbard–Satterthwaite theorem about voting rules. The main difference between the two is that Gibbard–Satterthwaite theorem is limited to ranked (ordinal) voting rules: a voter's action consists in giving a preference ranking over the available options. Gibbard's theorem is more general and considers processes of collective decision that may not be ordinal: for example, voting systems where voters assign grades to candidates.
> it also mentions that gibbard's is specifically about irv.
This isn't true
> I can't understand the idea that we only get to change things once.
Those of us concerned about IRV and promoting cardionality actually looked at history. Between 1910 and 1920 40 US cities used Bucklin voting (similar to IRV, slightly better even) and all repealed them[0]. So looking at history we see that people recognized the need for a better voting system, implemented something similar to IRV, saw that it didn't make things better, and subsequently said "fuck it, we'll go back to FPTP because it is easier." (I should also mention that in Australia, since 1949, 90% of Lower House elections, which use IRV, are equivalent to using FPTP[1])
So we're looking at history (and modern times) and saying "hey, this didn't work and actually ended up causing us to take a step backwards. Maybe we shouldn't repeat the same mistake."
I hope this clarifies our differing understanding of history.
A little, I think we may have different ideas about what the difficulties were in the early 1900s vs now - I believe that the circumstances are different enough now that broad conclusions about what is and isn't feasible cannot be drawn, but that's just a piece of the puzzle and I think your point deserves merit.
What I would ask, then, is rather than not doing IRV, what should we do, in your opinion?
I'm looking at this as a sort of crisis situation, as our ability to affect our politics is in a state of constant erosion, and the process of capture at work here can only end in systemic collapse - the further power gets concentrated, the more centralized our decision making becomes, and the more vulnerable we become to systemic single point of failure.
I would love to see STAR voting become a thing. I think of IRV in the current context as a proof-of-concept that might show people that we can change the structure of voting. I can't see anyone wanting to go back to FPTP, because I don't really see anyone who thinks it's even remotely working.
> it's impossible to completely eliminate strategic voting. No matter the method: ranked vs. cardinal vs. anything else you dream up can't help.
I think this comment is a defeatist at best and deceitful at worst. Just because there is no global optimization does not mean that all optima are equal. We can in fact have optima that are better than one another (including all optima we know about!). This is a common feature of highly dimensional solution spaces.
The big issue here is that not all criteria are weighted equally, by desire of effectiveness. So we find an optima where we optimize features that have a large weights and care less about optimizing features with small weights. By doing this we can compare systems in their desirability and select the best ones. This is not cause for throwing up your hands and giving up.
As an example: cardinal systems, when compared to ordinal (ranked) systems, are more resilient to strategic voting and simpler (both for the voter and for the people tallying the votes, aka transparency). The cost? Slight decrease in maximal VSE. BUT if we look at the min, mean, median, or modes of VSE given different strategies cardinal system outperform ordinal (aka, desirable). You can see this by comparing with this chart[0]. For example with STAR0-10 we have maximal VSE of .983 and minimal of .912 (actually this makes it strictly better than plurality!). But if we look at our best ordian, RP, we see RP has a maximal VSE of .988 and minimal of .870. So on terms of maximal there's a 0.005 difference but on minimal there's a difference of 0.042! We can easily tell here that STAR is much more resistant to strategic voting than RP (Shulze is even worse!). Doing the same for IRV we see .07/.115 (max/min comparison of STAR0-10 vs IRV on VSE).
So we can compare. We can select better methods. But is there a ''perfect,, solution? No. But don't let the lack of the ability to create a perfect system detract from the ability to compare systems. Not all is lost.
> I think this comment is a defeatist at best and deceitful at worst.
Wow, very HN! I don't actually mind very much, but seriously, consider applying the principle of charity?
I agree with you, and did before you wrote this comment too! Hence my use of the word "completely". I was just surprised that strategic voting couldn't be completely eliminated; before yesterday I expected that you could, in exchange for losing other nice properties. Of course being impossible to eliminate completely doesn't mean it shouldn't be minimized.
I thought Gibbard-Satterthwaite (or maybe just Gibbard's version) applied to cardinal systems as well. This seemed likely to me for some reason, as if an ordinal method could approximate any particular cardinal discretization by just including "ghost" candidates that can be packed (ordinally) between your actual candidates.
> I thought Gibbard-Satterthwaite (or maybe just Gibbard's version) applied to cardinal systems as well.
Correct, but it is also a weaker version of Arrow's (also as Clay points out, Arrow's isn't really about voting[0]...)
> This seemed likely to me for some reason, as if an ordinal method could approximate any particular cardinal discretization by just including "ghost" candidates that can be packed (ordinally) between your actual candidates.
Okay, but this just adds complexity. Cardinal is already simpler than ordinal systems (both for voters _and_ for those counting the votes). There's absolutely nothing wrong ranking candidates in a cardinal system (it's actually pretty unlikely that you'll have the same preference for multiple candidates so this is going to naturally happen). The difference? In cardinal you can better express your preference of one candidate over another (I give an example here[1]). So now we've added "encoding efficiency" to the added benefit of cardinal systems.
Cardinal systems are better than ordinal systems in almost every single way (the only thing I can think of ordinal systems doing better at is that RP and Schulze perform better on maximal VSE, but as I discuss here[2] that is pretty limited as well as unlikely considering strategic voting and the ability to manipulate people exists).
I wouldn't choose to use the word "dishonest" but I sympathize - in nearly all news media explanations of "ranked choice voting", the description solely explains instant runoff, and makes no mention whatsoever that there is a choice of algorithm that comes with the choice to use ranked ballots, such as Condorcet. The words "instant runoff" are almost never mentioned.
Choice of language matters, and "ranked choice voting" (to solely mean "instant runoff voting") is a conceptually misleading term meant to load the debate. It dis-educates rather than educates on how voting systems work.
Advocates of other voting systems have to first start out by explaining this misdirection and unpacking the wrong mental model, then get to the merits of one voting system vs. another - because people have been told that "ranked choice voting" works a certain way and only a certain way.
I’d say that part of the cost of switching voting systems is educating voters how the new system works. Many voters simply don’t care about the finer details and just have a couple desiderata—we want to have our vote count, and we want to be able to vote for our favorite candidate.
Instant runoff is super easy to explain and achieves those desiderata, at least to some extent.
It’s hard to quantify the differences in mathematical terms against soft concerns like “we can educate voters and explain how this system works”. For that reason, I would like to see a non-mathematical argument against IRV, one that accounts for the other half of the voting problem.
I'm always baffled by this claim that simple explanations favor instant runoff.
Here's approval voting: "Upvote the candidates you like. The candidate with the most total votes wins." It's not just mathematically simpler, it's simpler in informal terms, a smaller change from current plurality voting, and it seems less messy in the strategic dynamics insofar as we've tested these different systems for real.
Even score voting is easier than IRV. You can rank if you want to. But you can better specify preference because you're not ranking with the same preferential distance between candidates and people. We use score all over the place (we could call Hacker News score voting with a small range: -1/0/+1).
> Instant runoff is super easy to explain and achieves those desiderata, at least to some extent.
Except IRV doesn't. Not only are cardinal systems easier but IRV doesn't achieve the most desired feature that people are looking for: no spoilers (or alternatively put, allowing votes for 3rd parties without "wasting" your vote). In fact IRV increases spoilage while cardinal systems (which is alternatively proposed) decreases[0]. Meaning we're doing the opposite of what we're trying to accomplish.
As for simpler, I'll refer you to this[1]. The short story is just rate candidates. And if you end up ranking, no worries. It's more difficult to mess up.
> Approval voting I think is much too tactical and, strictly speaking, worse than IRV.
On the contrary, approval voting gets better results than IRV with any measure of strategic or honest voters. See extensive computer simulation results by Harvard stats PhD and voting methods expert Jameson Quinn. Brown (50/50) is probably the most realistic setting.
A simple example of IRV strategy is next year's Alaska senate race. Murkowski would beat either rival head-to-head but is likely to be eliminated based on first-place votes. So Democrats want to strategically rank Murkowski 1st in order to help her survive to beat Tshibaka (Trump Republican) so they at least get their lesser evil.
Here's a good comparison of approval voting vs. IRV by experts. Full disclosure, I was a CES co-founder and have written extensively on this topic for 15 years.
> I honestly thought that after learning Arrow’s impossibility theorem that Condorcet is not especially important.
I've visited Kenneth Arrow at his home and co-founded a non-profit that interviewed him. His theorem only applies to social welfare functions, not voting methods, if properly understood. But if anything, the moral is to AVOID ranked voting methods and instead use rated voting methods.
> On the contrary, approval voting gets better results than IRV with any measure of strategic or honest voters. See extensive computer simulation results by Harvard stats PhD and voting methods expert Jameson Quinn. Brown (50/50) is probably the most realistic setting.
If you look at the graph with the honest strategy you get better results with IRV which is basically his point.
We shouldn't assume 100% honest strategy in reality. It is an idealistic metric. It's not practical since people aren't 100% informed. Nor should we only consider optimal strategy. We need to consider the max, min, and median to determine robustness and better estimate real world results. Both the max and min for Approval is better than the max and min of IRV (in other words, the range is smaller and the expected result is more representative).
> IRV lets people express their preferences in a fairly understandable way. The strategy I see...
That's the concern though. You likely don't want to just "express your preferences." You need to strategically rank candidates. But then what are you being asked to do exactly when you go to vote? I don't think it would fly if the instructions on the ballot gave you guidance on how to "strategize" your rankings. But if it doesn't give that guidance, it's misleading.
And I understand that all voting mechanisms have their own unique pitfalls; so we have to fall back to pragmatic questions. And one major pragmatic question is: What's the status-quo/What is everyone already accustomed to? Asking a large voting body to switch methods is not easy and I think it's being motivated largely by a misleading "grass-is-greener" claim about rank voting.
The non-monotonicity tactical voting possibility is overblown: it only really works in the toy "which is the favourite pizza of these 30 students" examples.
In real world elections, the conditions where it is even theoretically possible arise only rarely (A > B > C with three choices, but B > A and A > C with two choices, and A's lead over B significantly greater than B's lead over C in the three choice scenario, and A's lead over C in the two choice scenario more than twice B's lead over C in the three choice scenario) and more importantly, they're not predictable enough beforehand. Advocating this kind of tactical vote stands at least as much chance of hurting your candidate as helping them, so nobody does it.
When you analyse real-world large scale IRV elections, you find that cases where the IRV winner isn't the Condorcet winner are rare, and this balances against the very real benefit of having a counting method that is easy to explain and understand.
The non-monotonicity is actually pretty important and we have real world examples of this. Clay mentions the Alaska race in this comment[0]. We've seen this issue in Southern states that used this to disadvantage black voters. This is also the reason Bernie would spoil Biden (and vise versa). So I'm not sure what you're getting at.
Let's look at the "Bernie would spoil Biden" example.
The idea is that the Republicans endorse only one candidate, Trump, and the Democrats endorse two: Bernie and Biden. In the non-tactical voting case, the second-last round votes shake out like:
Trump 45
Biden 28
Bernie 27
Bernie is eliminated, the votes in his pile split 24 / 3 between Biden and Trump (a preference flow of 89% to Biden) and the final round of counting ends up:
Biden 52%
Trump 48%
..but tactical voting intervenes! A small number of Trump voters (2% of the total electorate) are organised to tactically switch their votes to 1. Bernie 2. Trump, resulting in a second-last round of:
Trump 43
Bernie 29
Biden 28
Now Biden is eliminated, the votes in his pile split 20 / 8 Bernie/Trump (a weaker preference flow of 71% to Bernie, because of Biden voters too conservative to vote for Bernie) and the final round is now:
Trump 51
Bernie 49
Tactical voting has won the day!
However, this really illustrates the problems here for the prospective tactical voter:
1. They need pretty perfect information to pull this off. Not just on first round votes, but on how the preferences are going to flow as well. If those Biden votes flowed a little weaker to Trump, all they'd have done is elect Bernie instead; if Trump had been a little stronger overall than they expected in the no-tactical-voting, their tactical voting attempt might have backfired completely and turned a fair Trump win into a loss! Opinion polling just isn't this precise.
2. They have a co-ordination problem. If they switch too few votes, the scheme fails and they just elect Biden with a greater margin than before; too many and the scheme fails and they elect Bernie, a candidate they're presumably less happy with actually getting the Presidency than Biden.
3. All of this only works if the balls line up perfectly in the first place, even setting aside the problem already mentioned of how you know the balls are going to line up. If the second-last round votes are instead Trump 45 / Biden 30 / Bernie 25 then the Bernie to Biden preference flow has to fall under 66% for the scheme to be possible.
Rather than trying to engage in this dubious and risky tactical voting attempt, the Trump campaign would be far better served just spending their resources trying to turn out more of their voters. After all, the theoretical possibility only exists when the margins are tight in the first place.
A tangential matter is that even in an IRV election it still probably makes sense for the parties to either endorse only a single candidate each - otherwise their candidates will waste some of their resources attacking their co-party candidate when they could have launched them against their main opposition - or at least mutually agree to distribute campaigning material advising to give a second preference to their co-party candidate (a so-called "preference swap" arrangement).
> You can get a sense for what it's like to vote in a ranked-choice election here:
That’s a neat visualization. The step where mushroom pulls ahead of sausage really shows the flow.
Though I do question the integrity of a pizza topping poll that manages to exclude pineapple-ham, not individually, but as a joint entry. Did it miss the filing deadline?
> And you can see Mayor Bill de Blasio eating a pepperoni pizza as a result of this here:
Worst mayor ever eating one of the worst pizza toppings ever.
I think that it would be interesting to see this pizza example for comparing ranked voting against acceptance voting.
After all, in this example the plurality vote winner is still the clear, uncontested winner in ranked choice, so this example makes it look like ranked voting is just a way of complicating a process that "just works". I know this is not actually true, but this is how some people see alternative voting methods.
On the other hand, with acceptance voting you would probably get a different result as most meat eaters are indifferent towards mushrooms while many vegetarians would be passionately against pepperoni. So, I suspect that with acceptance voting the results could be switched, for better or for worse.
No, you don't want dumb people to have representation, because idiots will spread misinformation (everything from climate change being fake to masks being unnecessary) and adopt corrupt practices to garner their votes. Smart people are less vulnerable to misinformation.
You want people who believe science, have a sense of foresight for several years, understand international politics, and how resources and economics work to decide the future of the country.
Another downside is that while it does prevent spoilers, RCV doesn’t prevent other forms of vote splitting. It keeps an unlikely outlier candidate from tanking the consensus candidate, but it doesn't prevent splits between three or more viable candidates. This is especially problematic because one most important supposed qualities of RCV, according to its advocates, is that it makes more candidates more viable.
An Australian comic author has written an easy to understand comic about exactly this. Let Dennis the election koala explain it, with a worked example: https://www.chickennation.com/voting/
The really unfortunate thing about NYC's ranked choice voting is that you're limited to ranking only 5 candidates. This means you still need to engage in tactical voting.
You can't just rank the candidates you like, in the order you like them, because your genuine top 5 choices might not include the people who are most likely to win. You'll be "throwing your vote away" unless you include the most-likely-to-win candidates - that is, unless you vote tactically.
I've seen lots of discussion in NYC where people are trying to figure out the right tactical voting to do given their views and given the fact that they can only rank 5 candidates. Ranked choice voting is supposed to make tactical voting essentially irrelevant, so this is disappointing.
I don't disagree it's not ideal if you have more than 5 choices you like, but let's not over criticize a small flaw without appreciating the massive progress here. I read "horrible and unfortunate" and can see how people would take that to mean "back to FPTP".
I voted today and put 3 small candidates and two who "have a chance". I'm really excited to see how those 3 candidates shake out in the rounds, and that alone might give them a lot more power next race. That's super exciting to me and I think we can celebrate that and then push to improve the issues after :)
Yes, sorry, "horrible" was a bit strong and I edited that out of my comment immediately after posting it, heh...
I do appreciate the progress, but I worry that people will regard this incarnation of ranked choice voting as "more complicated" when the entire point is that it's less complicated for the individual voter...
Could you elaborate how RCV is less complicated? Just by eliminating tactical voting? In traditional voting you often limit yourself to the incumbents, isn't that a lot fewer choices?
Limiting yourself to the likely candidates requires a level of awareness around polling - that information is practically forced down our throats since voting in America pretty much requires it, but polling information (what other people think of the candidate) isn't an important piece of information for the voter - it's their opinion the system ideally wants them to express.
RCV is, at a maximum, equally as complicated as FPTP style voting since you have the option to elect only a single choice on your ballot (unless they're doing something weird). The form of the ballot may be visually more overwhelming but that's just from unfamiliarity.
Right, but you can argue ranking N items takes more thought thank picking 1. I like the "interoperable" way, just making choices 2 and onward optional.
Given the wording in the article I'm almost certain that underfilling the selection wouldn't result in a spoiled ballot. People can continue to just pick their favorite dude and go one with their lives. From a survey mentioned in the article apparently only a quarter of potential voters choose to only make a single choice[1].
A voting system where you're limited to only voting for the incumbent is indeed very simple (although not very democractic)... I assume you meant "limit yourself to the frontrunners". :)
In an election where there's exactly two candidates, RCV (or most other alternative voting systems) is the same as simple majority. You're right that with exactly two candidates, tactical voting isn't a concern, and simple majority voting is the simplest and best system. But many elections do not have exactly two candidates - primaries being the most salient example.
With RCV/alternative voting systems, you don't need to know who's popular or "electable" or any of that. You can just go with your most naive, direct take on the candidates.
Say you like two candidates A and B, but most of all you want to make sure candidate Z loses. Just rank A and B first, then everyone else equally, then candidate Z. You don't have to think about who is most likely to beat Z - you don't even have to follow the election at all.
How much time must people be putting into politics that they have such nuanced views that make it impossible to express a clear opinion on 4 top candidates and stick a major party candidate in spot 5? I guess New York politics must be vastly different from the West coast, because it's pretty uncommon to see a single race with more than 5 candidates in the first place, and truly unheard of for there to be 5 candidates that have any chance of winning in my experience. Having such strong opinions that this qualifies as "really unfortunate" sounds absolutely exhausting to me.
Even if they didn't limit it to 5, you still need to engage in tactical voting in IRV if you don't want to risk throwing your vote away. The spoiler effect is actually still there, it's just harder to understand. The problem is that your second choice only counts if your first choice is eliminated, so it's very possible that none of your rankings matter and have no influence on the outcome if you don't pick one of the top two candidates as your first choice.
> it's very possible that none of your rankings matter and have no influence on the outcome if you don't pick one of the top two candidates as your first choice.
I'm not sure what you're describing. As long as one of the top two candidates appear on your ballot (at any rank) then you have influenced the last round of voting.
Suppose you have three candidates, A, B, and C, and the votes look like this:
45% of voters rank A 1st, B 2nd, C 3rd
25% of voters rank B 1st, A 2nd, C 3rd
30% of voters rank C 1st, B 2nd, A 3rd
Nobody has the majority, so B is eliminated, and A wins round 2 with 70%. But 30% of voters preferred B over A, and if they had ranked B first, then B would've won in the first round with 55%. By voting honestly, the C voters threw away their votes. None of their later rankings mattered, because C was not eliminated before a victor was declared.
All of the strategic situations with instant runoff seem to require more detail on the outcome than anyone can predict in practice. So I’m not sure this will ever actually be a thing for voters think about.
I'm struggling to find a source, but I did read an article once about how some elections in Australia have fallen back to "lesser of two evils" precisely because of the need to vote strategically in order to make sure your vote isn't spoiled.
But suppose you're right and voters rank candidates honestly because they can't predict how to vote strategically; ultimately, I think that is arguably even worse, because even if voters can't predict how to vote strategically in advance, they can certainly still feel resentful after the fact when the results are tallied and it becomes clear that they got screwed over by voting honestly. At least in plurality voting, it's easy to understand the tradeoffs and choose whether you want to vote for a spoiler candidate or not. Surely it will not lead to greater voter enfranchisement for people who find out after they voted that they accidentally spoiled their own votes, and could've had a say in the outcome if they'd expressed their rankings differently.
If this still all sounds hypothetical, keep in mind that this is exactly what happened in the 2009 Burlington, VT mayoral election. The Republican candidate acted as a spoiler for the Democrat (who had the broadest popular support) and instead the Democrat was eliminated and the third-party Progressive candidate was elected. This led to wide-spread dissatisfaction and IRV was repealed there in 2010.
Also, if my example seemed too abstract, let's imagine a real-world presidential election using IRV in a red state like West Virginia. In 2020, Trump got about 70% of the vote. Now suppose Mitt Romney ran as well in a three-way race against Trump and Biden. Since Romney is also a Republican, then it's likely some of the Trump voters would've ranked him first, but most of the Democrats would've still preferred Biden. But since Romney is more moderate than Trump, the Democratic voters would probably still prefer him over Trump. Now the results might look like this:
45% of voters rank Trump 1st, Romney 2nd, Biden 3rd
25% of voters rank Romney 1st, Trump 2nd, Biden 3rd
30% of voters rank Biden 1st, Romney 2nd, Trump 3rd
This seems like a fairly plausible real-world situation, and it also seems pretty easy to imagine how Democrats could look at the polling numbers and realize that Biden is never going to win in their state, and that ranking Romney 1st is their only chance to stop Trump.
We've had this in Australia since day 1 of Federation.
One important factor the article didn't quite emphasise is how this voting system can break up the two party duopoly without creating 1920s-Reichstag-esque anarchy.
It's completely normal to put a third party as your first preference, while still indicating that you'd prefer Kodos to Kang.
In practical terms, it means the Dems would actually have to do some work in order to win the black vote, and likewise for the Republicans and working class white people.
Fellow Aussie and RCV-enthusiast here, but minor correction: RCV was introduced in 1918 in response to elections where a minority candidate won due to vote splitting.
Would you mind elaborating a bit on that point? From what I've seen RCV does tend to have a really strong moderating pull on elections that something like MMP manages to dodge, but it does mean that big tent parties will often experience strong pull off and, honestly, the African American community is often under served by a lot of Dem economic policies - though I'm really speaking in demographic generalities here.
Worth noting Australia is still generally a 2-party system in the lower house (1 representative each area) but less so in the senate (where each state elects the top 12 senators). Australians still look at envy at the New Zealand and German systems of Mixed Member Proportional representation (MMP)
We here in NZ would also like some kind of ranked choice in conjunction with MMP - as it stands, people are still reluctant to vote for smaller parties that may not reach the 5% vote threshold required.
Edit: I may have confused RCV with STV. Regardless, it would be nice to know that if my first option didn't make it into Parliament that I could prefer a second party, and so on.
As the wiki article for STV helpfully states, "Another name for STV is multi-winner ranked-choice voting". Also, "Instant runoff voting (IRV) is the single-winner analogue of STV".
So ranked choice voting is a super-set with many flavours within.
MMP is a fantastic system, but it can face resistance if it relies on Closed Party Lists[0], which allow a party to put a potentially controversial candidate at the top of the list, almost guaranteeing them a seat and forcing party supporters to accept that as the cost of choosing the party.
Open lists are better, but I believe they still have the problem that they make it unclear where representatives' mandates come from, especially if a representative can resign and be automatically replaced by someone lower on the list without a by-election.
Other problems exist with list-based systems, such as treating independent candidates differently to candidates with established parties, and the possible use of "decoy lists"[1].
The best implementation I've seen of MMP, though, is that used in Baden-Württemberg, which avoids party lists by filling the proportional seats with the candidates who came closest to winning the elections in their local districts.[2]
I don't see a practical party with closed list, as long as the minimum threshold isn't too low. Don't like the people at the top? Vote for a small party which will join the coalition.
Even more fundamentally, politics is a "team sport". The vetting of individuals when power boils down to alliances and other relations with the politicians is stupid waste of voter's time.
The two jurisdictions in Australia that use Hare-Clark (Tasmania and the ACT) no longer have a political duopoly.
MMP would probably work well in South Australia (where overall the electorate is fairly evenly split right-left, but many of the right-leaning voters are locked up in a few geographically large rural electorates and the left-leaning voters are distributed more evenly across more urban and suburban electorates).
It looks like the article got the Australian implementation somewhat wrong.
Under cons, it says that voters have to rank all candidates on the ballot paper.
In reality you have a choice. You can either rank all, or if you are lazier you can just select your favourite candidate and then your first choice candidate's preferences will be used instead.
Also for some larger ballots (usually the senate with nearly 100 options) there is now a requirement to only rank say the top 10 or so for the ballot to count, so you don't need to number all 100.
This is one of the best features of the Australian system. If you want to do the basic effort you can just tick one box, but if you care about the ordering you can also make your preferences count if you so choose.
As an outsider looking in to American politics, I feel changing to preferential voting is the best bang for buck change to move away from extreme politics. Hopefully this catches on elsewhere.
In reality you have a choice. You can either rank all, or if you are lazier you can just select your favourite candidate and then your first choice candidate's preferences will be used instead.
This is not true. In the elections for the Australian House of Representatives (single-member electorates, comparable to the mayoral election under discussion where there is a single winner), you must number every box for your vote to count.
In elections for the Australian Senate (multiple member electorates) you don't have to number every box, but you have a choice: you can either vote "below the line" for individual candidates, in which case you must number at least 6 boxes for your vote to count (the ballot paper advises you to number at least 12); or you can vote "above the line" for groups of candidates (in which case you only have to number one, but the ballot paper advises you to number at least 6). A vote "above the line" for a group is equivalent to numbering the candidates in that group in order from top to bottom, but it doesn't imply a vote for a candidate in any other group.
The voting systems in Australian states and territories vary from the above; some of them do allow only numbering a single candidate in single-member elections ("Optional Preferential Voting" or OPV) and some still have group-ticket voting in multi-member elections where you can assign your vote to the preference ticket submitted by a candidate (though hopefully the last few jurisdictions to still have this will be getting rid of it, because it is being gamed).
In Canada in 1952, the province of British Columbia switched to a ranked ballot and it nearly ended in disaster.
The ranked ballot was an attempt by the traditional Liberal and Conservative parties to keep a rising socialist party from gaining power, as they assumed no Liberal or Conservative voter would choose the socialists as their second choice.
But there was so much animosity between the parties that most voters chose a fringe right-wing party as their second choice, as a protest vote.
The two traditional parties were wiped out and the fringe right-wing party narrowly won, despite having no leader and little governing experience.
Fortunately, saner heads took over the fringe party and quickly steered it to the political center as a sort of big tent party for former Liberal and Conservative voters. But it could have gone very differently.
The ranked ballot idea was quickly dropped, but the event shaped British Columbia politics for the next 60+ years. That fringe party elected by accident became the province's natural governing party.
I suppose the moral of the story is that no voting system can prevent irrational voters from shooting themselves in the foot.
What aspects of parliamentary systems make it difficult to apply lessons from their voting systems to the voting systems used to elect representatives to the US Congress?
Ranked choice does avoid vote-split issues, but is insufficient for breaking a two party system because the races are still single-winner. That we still have primaries makes it worse.
Proportional representation is what is desperately needed to break the two party system.
> One important factor the article didn't quite emphasise is how this voting system can break up the two party duopoly
I'm not convinced this is true. 2 parties control 80%. 2 coalitions control 95%. I'm also not sure how you can look at America and not recognize that our parties are more similar to coalitions (come on, AOC/Bernie are not the same party as Pelosi/Biden).
A better explanation is that Australia uses a parliament which has proportionate representation. My evidence of this is that other parliaments without IRV have similar party structures as Australia. So we have examples where you remove one aspect and things don't change, I don't think that makes for strong evidence that the thing removed is causing the effect (how can it cause an effect somewhere that it doesn't exist!).
So don't compare Australia vs America to prove IRV creates more parties. Compare Australia/France/UK vs America to prove parliament causes the desired outcome.
> We've had this in Australia since day 1 of Federation.
Australia adopted the preferential system in 1918.
> It's completely normal to put a third party as your first preference, while still indicating that you'd prefer Kodos to Kang.
The evidence shows most voters don't, which is why Australia's IRV elections still maintain a duopoly. In spite of the fact that the senate, which uses (proportional) STV, has escaped duopoly.
I've always seen the proportional nature as being key to other countries lack of a two party system (though they have two major coalitions). I'm curious why you think many look to Australia's IRV as the reason for it being a multi-party system when many parliamentary systems that don't use IRV have even more parties. How do you combat this common misconception? Also it seems dishonest to not consider coalitions when US parties aren't as uniform as most other country's parties. If we look at coalitions it does not appear that Australia (or most others) have escaped duopoly.
And as always, good to see you in the comments and thanks for all the information.
It's worth mentioning that the way this has taken away seats from the two major parties is not mostly through the rise of third parties, but the election of independents unaffiliated with any party. This has invariably occurred in electorates which are so homogeneous they were safe for one or the other major party, usually (but not always) rural.
The most spectacular example of this was the election of three rural independents who then proceeded to install Julia Gillard from Labor as Prime Minister, even though they were conservative. Their stated reasons were Labor's promise to build a fibre-to-the-premises broadband network and their assessment of the conservative leader. Other examples include members in wealthy Sydney suburbs who, while otherwise conservative, wanted action on climate change, and members in farming electorates unhappy with the way the Nationals have been captured by mining and housing development interests and failed to represent farmers.
(The Nationals, formerly the Country Party, always join the conservative Liberals in coalition, apparently having forgotten the purpose of coalitions.)
An exception is the Greens, who usually score one seat based on the central business district of Melbourne. This includes very high density apartment towers, and student accommodation for the University of Melbourne, and is demographically odd.
Many observers believe the rural independents may eventually coalesce into a new party which displaces the Nationals. I see no evidence thay the Liberals' several humiliating losses to independents in "safe" seats is causing them to rethink their policies. This may or may not change once they are unable to rely on whatever replaces the Nationals to form government.
I'm Israeli and I was recently shocked to learn the rest of the world trusts electronic voting and vote by mail. We have a very simple and strict voting process here - the key feature which builds the trust in voting is that every 4 poll workers are responsible for the integrity of only up to 800 votes. We don't have fraud allegations because you would have to blame over 150 different people simultaneously (and often much more, on the order of hundreds of people, when you take into consideration the practical constrains) to change a single seat. It's very simple to convince a person of the integrity that this process. We have a high turnout which completely contradicts the claim that mail / electronic voting increases turnout. Electronic voting / mail voting give the power of changing the results to a handful of people who control the voting machines / know the voter rolls. Whether they will or had abused it is anyone's guess - but pretending that controversy over election process is inevitable is plainly wrong. The entire premise of elections is to prove to the losing side that the process was fair, so I find it unbelievable so many countries overlook that and give things like convenience and corona safety a priority.
Most western democracies don’t have a problem with fraud allegations in voting. It’s really only the US. And even there, the allegations are mostly a political ploy to push for policies that disenfranchis voters of other parties. Which again doesn’t exist in other Western democracies.
French had massive mail voting fraud in 1975, after which they banned it.
The entire point of elections is proving your government is trustable, you can't trust your government with it because then it's like a self signed certificate, circular trust problem.
Who rigs elections in Russia? The government. Who is most likely to rig elections? The government.
You can't trust government with elections, and just because the government shows you documents doesn't mean anything. Governments fabricate documents all the time. If all you require from corrupt politicians to remain in control is a bunch of mailed documents, you can be sure they'll get those documents mailed. I'm just completely dumbfounded people living within countries with these elections think they are democracies.
>We have a very simple and strict voting process here - the key feature which builds the trust in voting is that every 4 poll workers are responsible for the integrity of only up to 800 votes
Can you say more about this? So at each polling place 4 workers are assigned 800 votes, and then they rotate out to a different group of 4? I don't see how that inspires more trust than our current system, the government could simply be lying that they rotated the 4 poll workers out. Or, in a large city this would be a complete cluster as you'd need hundreds of poll workers on standby to rotate in & out every 30 minutes or so. I would definitely be interested in hearing more though
I really hate that Instant Runoff Voting, which is almost literally the worst possible voting method with ranked choice ballots, has somehow gotten the name “Ranked Choice Voting” stuck to it.
It’s like if “Democratic Elections” became a specific name for Plurality voting.
Yeah, when you name the spec after one particular implementation, you're gonna cause confusion. Anyway, I handle it by just using RCV to refer to how the ballot is filled out. The vote counting method is IRV.
"Worst possible" might be a bit strong, but an important property that a voting system should have is that it finds the Condorcet winner if one exists. The Condorcet winner is the candidate who beats every other candidate in a head-to-head election. IRV is not guaranteed to find a Condorcet winner.
The problem with IRV is that you can have a candidate who is everyone's second choice, but then loses out on the first ballot because too few people ranked them first.
I've been voting RCV and I've yet to see this actually occur in practice. The candidates who are marketing themselves so well that they are second on almost everyone's ballot (in a 13 person race!) are generally FIRST on a fair number (relative to all 13 folks).
Worst possible ranked choice alternative, not worst possible voting method (i.e., still better than FPTP, but not without its weaknesses). Frustrating as there are more straightforward, and objectively better mechanisms which are ignored in favor of IRV. It's hard not to feel conspiratorial that maybe the Powers That Be want worse voting mechanisms.
This page is a good explainer of IRV's weaknesses and shows some of the alternatives: https://ncase.me/ballot/
Since you mention it, let me offer a candidate for "worst possible voting method (that has actually been used in a democracy)", namely "plurality-At-Large".[0]
It basically takes the worst properties of FPTP and extends them to an entire governing body, allowing a party with less than 50% support to win 100% of the seats (by design).
Fortunately it has been mostly abandoned in the US thanks to the Voting Rights Act and subsequent SCOTUS decisions.
> Fortunately it has been mostly abandoned in the US thanks to the Voting Rights Act and subsequent SCOTUS decisions.
A close variant (vote for up to N, top N win, at large) is used in a lot of local elections still; all multimember district methods are banned in House elections specifically to prevent plurality at large from being used there. (Though the Senate uses temporally-segregated two-seat plurality at large in every state.)
A succinct version (others have linked more detailed, largely outcome-oriented analyses, which are also valid and important) is that, of seriously-considered ranked-ballots methods, it both throws out the most information from the ballots and uses information rather unequally across ballots, attempting to simulate FPTP elections with narrower candidate pools until you get a majority winner.
One of it's big downsides is that you can, in some cases, cause a candidate to lose by ranking them too highly. In voting-theory terminology, it fails the monotonicity criterion. In at least that one respect it's worse than first-past-the-post. (I mean, imagine if people could say "Gore would have won in 2000 if only a few less people had voted for him". That sounds nonsensical, but that's how IRV works sometimes.)
I think first-past-the-post and IRV are both bad voting systems that are unsuitable for real-world use in elections with more than 2 candidates. Approval voting is simpler than IRV and (as far as I know) it isn't worse than FPTP in any way, so I think we should push for that.
> Approval voting is simpler than IRV and (as far as I know) it isn't worse than FPTP in any way
FPTP might be better than approval voting, because approval voting favors weak moderate candidates with no agenda.
With FPTP the people alternate lefts and rights and the system develops. With moderate candidates the system may stop progressing: no reforms, no bugfixes.
If the voting public prefers a weak moderate candidate with no agenda, then that's the candidate that should win. I think weak moderates usually flop, though, if there's no particularly compelling reason for people to want that person to be president.
The way I look at is that it's a kind of relief valve. If both major parties (assuming there are two, maybe there'll be more...) nominate not-so-great candidates, then the voters have an option to elect someone like Tom Hanks. I mean, I would have voted for Tom Hank in 2016 if I had an approval voting ballot and he was one of the options.
> If the voting public prefers a weak moderate candidate with no agenda
No voting public prefers other candidates, and moderate candidate as backup.
I give you an example I posted in another thread:
* 60% want a dem candidate to win
* 40% want a rep candidate to win
* 70% are willing to join strategic voting to avoid allowing other party to win
In this case, the dem candidate looks like a correct winner.
With ranked choice the the dem candidate would win.
But with approval voting strategic "moderate" candidate wins. If they are moderate, that's good. But they can easily be just weak nobody.
It is not exactly what voting public prefers. That choice is a fake choice, like a girl, do you want to go to bed or be thrown into a forest with angry wolves.
> if there's no particularly compelling reason for people to want that person to be president.
The reason is to not let opposing party win. Strategic voting.
> If both major parties (assuming there are two, maybe there'll be more...) nominate not-so-great candidates, then the voters have an option to elect someone like Tom Hanks.
And they will if it is some form of ranked choice, but not just approval.
I don't think 70% are likely to vote for anyone unless they're exceptional in some way an almost universally liked. Candidates like Michael Bloomberg or Lincoln Chafee or John Kasich or Jon Huntsman or anyone else trying to go after the moderate vote just wouldn't get those kinds of margins. But anyways, that's beside the point.
In approval voting, strategic voting for moderate/novelty candidates is like this (assuming you have a favorite and least favorite major party candidate and the novelty candidate is somewhere in the middle): you can vote for the novelty candidate if you want. Doing so might cause your favorite major-party candidate to lose if you happen to be the tie-breaking vote. Or it might cause your least-favorite major party candidate to lose to the novelty candidate. You can decide whether the prospect of having your favorite candidate lose to the novelty candidate is worse than the prospect of having your least favorite major party candidate lose to the novelty candidate and vote accordingly. In all cases, if you vote for your favorite candidate you're always helping them win, and if you vote for the novelty candidate you're always helping them win.
How strategic voting works for RCV is that you can put your favorite major party candidate first, the novelty candidate second, and the least favorite candidate last. You know that putting the novelty candidate in the middle won't harm your favorite candidate's chances of winning, since that's a property of RCV (later no harm). However, what you don't know is whether putting your favorite candidate first will help or harm your favorite candidate, and you don't know whether putting the novelty candidate second will help or harm the novelty candidate. Maybe putting your first choice second or third will cause them to win. Maybe putting the novelty candidate first will cause your favorite candidate to win against your least favorite candidate.
This is complicated and hard to reason about, unless you have detailed information about how other people are going to vote. Which seems to imply that such a system would tend to disempower individual voters and empower whoever has access to detailed information about the intentions of the voting public: campaigns that can afford internal polling, Facebook, Google, etc..
The system allows ppl to vote a third party without „wasting“ their vote. It favors consensus candidates — that’s great when a single representative is needed (like a major).
For parliaments, however, the system favors the two main parties, since a candidate needs those 50% support. This means parliament, which _should_ reflect the pluralism of society, will not actually reflect that pluralism.
In effect, this ranked ballot business is a way to appear to engage in voting reform without actually wanting to change anything about the outdated 2-party system.
In Canada, the liberals promised election reform. Until it turned out that ranked ballot was not going to be recommended by the electoral reform commission. Since that was the system the Liberals wanted in order to improve their chances in future elections, and they had little interest in actually representative voting systems, they scuttled the process and reneged on their promise that this would be the last election using the first past the post system.
Both are miles beyond first past the post, and both scare many people just because it's something new and different.
Hopefully you can support any change locally that gets us off of first past the post. Once we get some voting changes, hopefully that makes future changes even easier to make... though I know that's wishful thinking.
Actually, that's debatable. There was an interesting simulation done awhile back [1] that compared voting systems, and they found that if people voted honestly then IRV is better, but if people vote strategically the result isn't any more democratically optimal than FPTP. Which is what we should probably expect to happen when voters figure out that it's only safe to put your first choice first when they're the overwhelming favorite to win or they have no hope of winning. Otherwise it's complicated: you could cause your favorite candidate to lose by ranking them too high.
Obviously this is one simulation and results would be different with different assumptions. But I think it's worth pushing back on the idea that IRV is way better. It solves the problem of minor candidates being spoilers, but what if people actually start voting for minor candidates and they become major candidates? With 3 or more major candidates IRV starts looking like not such a great system.
The problem with talking about this is that the problems with FPTP are readily apparent and easy to explain. The problems with IRV aren't obvious and are harder to explain.
But minor candidates being spoilers have come up at all levels of elections many times for decades.
Splitting the vote happens frequently, again at all levels of elections (at least nationally in terms of the primaries for US Presidents).
If you vote for your favorite candidate, and your candidate is not the last-ranked candidate, you cannot cause your candidate to lose by putting them first.
But it sounds like you’d agree that FPTP needs to go. Great, me too. That’s the real focus for me: let’s start using different methods that eliminate, even to small degrees, the problems of FPTP.
I do agree that FPTP needs to go. I'm worried though that replacing it with RCV substitutes problems we know and understand with another set of problems that we don't know to be worried about because we haven't experienced them yet.
> If you vote for your favorite candidate, and your candidate is not the last-ranked candidate, you cannot cause your candidate to lose by putting them first.
Wikipedia has an example. (Technically it's two voters difference, since a change of one voter's vote would, I think, change the result at most from a win to a tie.)
Possibly naive question here but in my limited exposure to articles about RCV it seems to be popular in more historically liberal areas. Is that true and why might that be the case?
The progenitor Kenneth Arrow was pretty wonkish even among economists. RCV appeals to technocrats, who like the provability of its assertions. It's also a change from tradition. People with these tendencies tend to skew left.
Because liberal (progressive) areas are full of people who are advocating for change. What that change should actually be varies greatly between individuals, so there is lots of fragmentation and in-fighting. The idea that you could vote for your specific niche but fall back to your preferred monolith is desirable because you don't feel like you're wasting your vote by voting for a minor party who will likely never win a seat. Conservative areas are full of people who want to maintain the status quo. They have varying ideas about how things should be but are unified by the fundamental idea that things shouldn't change. Since there is far less fragmentation, there is far less need to give your voice to a niche so they're happy to have their vote go directly to a monolith.
New York City is the one driving the news cycle right now (and rightfully so as 8 million people have the potential to use RCV for the first time). And people view NYC as pretty liberal.
But a great thing about RCV is that it's not inherently partisan or favoring of left or right. As a result, in addition to being used in places like New York City, San Francisco, and Maine...it recently passed for use in Alaska and nearly two dozen cities in Utah. That's a pretty diverse set of locales across the U.S.
It’s exciting to see another community replace the most pessimal voting method with the second most pessimal one! (In terms of Bayesian regret). I hope the experience is not so frustrating that they end up repealing it as several other municipalities have done after being dissatisfied with IRV/RCV’s weird behaviors.
I feel like approval voting gets overlooked too much in discussions of voting systems. It's so much simpler than instant runoff, and so much less prone to spoilers than first-past-the-post.
Yeah, I'm a big fan of approval voting, the amount of tactical voting is much less compared to most other voting systems (including Plurality and Ranked Choice). In particular a big flaw of Ranked Choice Voting is the center squeeze effect. This site has a good overview of it: https://electionscience.org/library/the-center-squeeze-effec...
475 comments
[ 2.4 ms ] story [ 92.0 ms ] threadCommonly what's done is that we truncate the ballot and only allow people to express, say, 5 preferences.
We’ve seen surprise results where an obscure candidate has won when too many top choices split the first choice votes and someone relegated to the floor vacuums up all the throwaway votes (people think they have to “spend” all their votes)
That’s the point. The compromise candidate won. If the top candidates had spent less time mud slinging they may have remained in others’ top two or three.
I don't have a strong preference between approval and RCV. But I think people closer to politics would. Being able to "be behind" one candidate is a meaningful promoter of civic engagement; forcing that decision is deeply entrenched in our politics.
That makes approval voting a tougher sell for implementation than RCV. Given how much better RCV and approval voting are than FPTP, it seems sensible to go with RCV.
That's not the phase of implementation that's difficult with any of these systems. It's getting the rules changed. Approval voting changes civic engagement in a fundamental way. (It also has much less real-world data behind it.)
Approval voting was passed by a 2-1 landslide majority in both Fargo and St Louis. Public opinion research by the Center for Election Science shows random American voters are much more supportive of approval voting than IRV, relative to the status quo. By a huge margin.
Agree IRV isn't the perfect system. But we have the downsides of being broadly explicable (Condorcet method's weakness), continuous with existing civic culture (approval voting's weakness), squeezing the centre (RCV/IRV's weakness) and being better than autocracy (plurality's weakness).
With some irony, the game of reforming voting systems is itself a plurality mechanism. IRV, re-branded as RCV, has taken off. It has precedence. The distance between each of RCV's competitors and it is dwarfed by the distance between each of them and FPTP. Given that, unifying around the winning "candidate" makes sense.
Put another way, bringing up approval voting in an IRV vs. FPTP discussion favors FPTP. If one wants a world in which approval voting reigns, I think it would be easier to get there with IRV as an in-between.
Why do you consider that the "wrong" outcome?
> We’ve seen surprise results where an obscure candidate has won when too many top choices split the first choice votes and someone relegated to the floor vacuums up all the throwaway votes (people think they have to “spend” all their votes)
I think its an understandable concern under any single round ranked-ballots method where you arr likely to have a large pool of cnadidates. But it can be simply remedied with a two-round system using the same method iterated to select a manageable pool of candidates (possibly gating in the top couple first-place vote getters even if they otherwise wouldn’t be in the top) for the runoff round, which still uses the same ranked-ballots method with more confidence that all the candidates have been vetted because there is a narrower field.
This keeps the integrity of whatever ranked-ballots method is chosen while avoiding the (small, but not nonexistent) risk of winning by flying under the radar.
Any system other than plurality allows a candidate with more first place votes to beat one with fewer. Any system other than plurality or majority/runoff allows a candidate with fewer first place votes than the second place.
But, much as I dislike IRV, I’m rather skeptical of the “winner should always be one of the top two by first-place vote count” criteria suggested.
OTOH, using any ranked ballots method, I can see an argument to do a wide open “primary” using the method and then if the winner is not also the majority first-place winner doing a separate “runoff” election using the same system of, say, 5 runoff candidates, determined as:
(1) the candidate that won using the ranked-ballots voting method selected
(2) any of the top two by first place votes not selected by (1)
(3) the first candidate elected by the selected ranked ballots method when the candidates admitted to the runoff by (1) and (2) are disregarded, plus enough candidates elected by the same method disregarding all previously-admitted candidates to complete the pool of 5 candidates.
The idea being to deal with the risk of “candidate was elected due to insufficient exposure and vetting” risk, without fundamentally compromising the desirable behavior of the selected ranked-ballots method.
That theorem effectively says that all sensible voting systems will in certain conditions allow a voter to gain a better outcome by voting insincerely (i.e. tactically). However, it doesn't say how easy it is for a voter to determine the correct tactical vote to make, so potentially a voter would need perfect information about all the other votes cast in order for their tactic to succeed.
https://www.king5.com/article/news/politics/king-county-to-c...
Tacoma did this a few years ago and nothing came of it.
Really hope it catches on.
Otherwise you get a pyramid of control that drives towards extremes.
Historically, let's say dems = 51% of voters, everyone votes party line.
Dems have a primary, and someone (usually pretty far left) motivates base and gets 30% of primary vote and wins primary with most votes.
Let's assume a 10% of dem's vote in primaries.
So you have 10% * 50% * 30% = 1.5% of voters voting for the primary winner.
They then go onto a matchup in general, but because it's a dem district, they win there as well (republicans usually also pick someone further right).
So 1.5% of more extreme / activist voters end up really deciding, because by the time you get to general there is no more competition.
Even if the "pyramid of control" doesn't drive parties towards extremes, it perhaps drives them towards memes; that is, simplistic but clear ideas that stand out among the noise. Unfortunately, while that might be enough to create a coherent identity for a party, it doesn't do anything to guarantee that the party has any sort of broad support from the population, which further alienates people from politics.
A weird take on the U.S. constitution: it actually established 4 "branches" of government. Citizen oversight was one of 4 powers designed to keep government in check. It's hard to look at today's America and feel like citizens have much oversight of government. Our "branch" has been systematically eroded. The current first-past-the-post system has been gamed heavily by two parties to reduce the impact citizens have on the outcome of elections.
RCV helps restore civilian oversight (many forms of voter suppression, i.e. gerrymandering, are less effective under RCV). My current thinking: feeling this will result in more "left" or "right" leaning governing bodies is a reflection of how citizen's want their government run. If the "silent majority" is right leaning, this will result in right leaning politics; left would yield left. But the "leaning" of the politics is less important to me as an outcome. The most important outcome is that citizens are restored to their rightful role in government: acting as a proper "check and balance" to the other 3 branches.
NYC isn’t as super progressive as people think but there is certainly a progressive streak and they are louder than their numbers. They organize and they vote in primaries and that has been their strategy the last decade to assume outsized power. Because many moderate liberals won’t likely vote for a Republican in today’s climate it works. Hence BDB as mayor whom is despised by moderate liberals.
So this should redirect power to the Democratic machine in NYC and make it more difficult for far left progressives who really been hijacking the Democratic Party by virtue of the primary system.
* You can vote for the random third party candidate without worrying about throwing away your vote.
* You seem to get a bit more moderation where we are if you also have an open primary / no primary type voting which gets really nice.
California does top two past the pole of any party inf the general which means even republican votes can matter in a dem on dem election (common out here) or dem votes can pull back things in a R vs R general (rarer but does happen).
Doing Score based voting is just much better and arguably simpler and less work.
I specially like the Star-Voting, where they do a runoff at the end to cover some extra cases. They also have a few variants for multi-winner and representational voting.
Its also much more familiar for people from real live, like online review.
https://en.wikipedia.org/wiki/STAR_voting
P.S: Wow a lot of down-votes. RCV is just the alternative that got the most marketing, but once you start looking into the different systems I really don't understand people can stick with RCV just because its slightly better then the system before. If you are gone invest all this effort in improving the system, why not go to something easier and objectively better.
This video shows the issue very well (one of the people involved in Star voting made it):
https://youtu.be/-4FXLQoLDBA
Given how polarizing politics tends to be, especially in a two party system such as America, I’d wager that most people put all 5’s for candidates of their party and all 1’s for candidates of the other party.
- it's easier to teach people ("most votes wins")
- it's easy to double check the results ("most votes wins")
- it's incentive compatible (you never put someone you hope loses above someone you hope wins)
Even if you never use 2 – 4 on a 1 – 5 scale, it's still better because you're making the same decision for multiple candidates.
Only using the max and min scores in score voting / range voting has a name: approval voting. For each candidate, you're answering the question: "Would I be ok with this person being elected?"
There is a clear intensive not to do that as it seriously hurts your overall preference. And Star Voting makes this even stronger compared to normal score voting because of the run-off.
Even if you assume that everybody is a 100% total party zealot it still only breaks down to approval voting and that is still a pretty good system.
> Given how polarizing politics tends to be,
Politics is polarized because of the voting system, if you look at actual data of preferences then you would see that there is far more agreement then can be expressed.
> I’d wager that most people put all 5’s for candidates of their party and all 1’s for candidates of the other party.
You would lose that wager. Most people are not party activists.
The presumed next mayor of NYC is going to be decided within the democratic party primaries today (or over the coming weeks as rounds of RCV progress), not in the general election later in the year. There's still a major spread of ideologies involved, but they're clustered within one party - which seems to have a lot of people thinking tactically about their votes, instead of just going straight-up party line.
If you want to avoid that, the runoff at the end of a STAR (Score Then Automatic Runoff) is explicitly designed to incentivize voters to provide more differentiation between candidates.
Certainly that's what happened in Burlington, VT where a spoiler candidate swung the 2009 mayoral election, and IRV was subsequently repealed. They didn't say "oh oops, well this experiment failed, let's try approval voting instead"; they just went back to plurality voting and that was that.
One major motivation for the runoff at the end in techniques like STAR (Score Then Automatic Runoff) is to encourage voters to give different scores to candidates (assuming they actually feel differently about them) rather than strategically voting all 5s and 1s and basically making it the same as approval voting.
It’s frustrating that they claim RCV is second best but (presumably) put what they argue is best below the fold.
Very weak evidence offered but the motivation behind the article was near the top.
"They also appeal to more ideological voters—especially on the left—by arguing that they can express their views with more precision in a ranked-choice system."
https://www.allsides.com/blog/new-allsides-media-bias-chart-...
1. may decrease voter turnout by 5-6%
2. not as obviously fair as traditional voting
3. increases existing divides
I personally think none of them are good enough points, they can all be dealt with. And 3 just means you can vote for who you like best, not just the ones in power.
It lets similar candidates make a case for their small differences, but allows: * voters to not waste their vote on one of 10 different similar candidates, hoping they picked the one with the most backing so they don't get a split vote * more than just two or three candidates to run and have a reasonable chance of winning
The more people running, the better. Sure, there's some level where it gets a little over the top, but there are safeguards in place for that like minimum number of signatures from verified residents of a city/county/state, so there's currently extremely little risk of RCV being enacted and 100 candidates showing up on ballots.
And even if that did happen, so what? Look up a few of the candidates and pick the one that seems to represent you best.
The WSJ editorial board’s argument is very weak, and seems very partisan. They cherry picked a few very strange, rare scenarios that would still not be any different with the current first past the post system.
Ranking things in order of preference is something we do all the time. “They’re out of my favorite fruit, but I’ll get this other one instead.” “You don’t have that chair in this color? Okay, I’ll take that other color.” etc
The WSJ editorial board says, “Major parties could be weakened to the benefit of more extreme candidates.” That already happens with first past the post. The 2016 GOP primary season for President is an example of this: the eventual GOP candidate did not win 50% of the vote in the majority of 2016 primaries. Instead, more GOP voters wanted a more moderate candidate, but the many moderate candidates split the vote. Thankfully the WSJ article points this out, but totally undersells it, and doesn’t make a strong case to at least use it in those situations.
RCV reduces the chances of an extreme point of view getting elected, because they have to be able to get approval from at least 50% of everyone voting, rather than relying on their opponents to split the vote.
Get more people running, so we have more options to vote for, and we can express more accurately who and what we support. RCV, or one of the many other alternatives to FPTP, is the best shot at doing this, and in my opinion one of the few ways to truly improve the US’s democracy without a bunch of massive changes that are even less popular with politicians (limiting donations, lobbying, etc).
In New York City, it was arbitrary and up to five. So yes, perfectly backwards compatible.
You can get a sense for what it's like to vote in a ranked-choice election here: https://app.rankedvote.co/elections/7568/Best-NYC-Pizza-Topp...
And you can see the results visualized here (over 20K people voted in this one!): https://app.rankedvote.co/elections/7568/Best-NYC-Pizza-Topp...
And you can see Mayor Bill de Blasio eating a pepperoni pizza as a result of this here: https://twitter.com/therecount/status/1404809865235767315
Do you think there is a need to educate voters on how to tactically vote in ranked choice voting? Every voting system has strategies to use it effectively and most voters are not used to the tactics necessary for ranked choice voting.
We see a lot of education in how to use the ballot simply, but very little education on "advanced tactics".
In particular, a lot of NYC voters aren't ranking either Wiley or Adams, which is a huge mistake as those are very likely to be in the final round.
As for advanced tactics, if you're really into it, Rob Richie, the CEO of FairVote (leading national advocacy org) put this in the New York Times over the weekend. It spells out a bunch of scenarios (e.g. "I want Garcia to win and Stringer to lose.") and how to tactically vote for each.
https://www.nytimes.com/2021/06/18/opinion/nyc-mayoral-elect...
BUT the point I'd want to drive home is that ranked-choice allows you to be FAR more able to "vote for who you believe is best" than in the more common most votes wins approach (where you have to be strategic the moment a 3rd candidate enters).
Once this NYC election is over we should look at the voting records and see how many votes were lost due to poor tactics (forgetting to rank front runners).
https://twitter.com/Beyond2Parties/status/140537245401753600...
Score voting or approval voting actually solve this and are much simpler.
Just to respond to this one small point, but I think the most common mathematical definition of "fairness" in market designs (including voting mechanisms) is that your utility-maximizing action should be identical to your true ranked preferences (this is called "strategy-proof"). In the case of voting, Gibbard-Satterthwaite says that there's no strictly strategy-proof mechanism (under a few restrictions) but I think the instant-runoff voting which NYC is using is mostly strategy proof (i.e. strategies only exist in rare circumstances).
Fwiw the burden of learning about all these candidates seems high to me, but apparently New Yorkers don't ind.
The very comment you are replying to pointed out that "a lot of NYC voters aren't ranking either Wiley or Adams" which is a failure to apply strategy where it manifestly exists. "There are front runners" doesn't seem to be a rare circumstance.
This is a valid communication problem with ranked lists, but it's not that there exists a "strategy" per se besides ranking all of your acceptable outcomes.
I expect they are being honest about their preference order, but that this has less of an impact on the outcome (in whatever direction they desire) than if they had been dishonest about their preference order.
This does assume that they actually prefer one of the front-runners to the other by a more-than-negligible margin; perhaps that isn't the case.
> This is a valid communication problem with ranked lists, but it's not that there exists a "strategy" per se besides ranking all of your acceptable outcomes.
There were more than six candidates, and voters could only rank their top five. Voters with more than 5 "acceptable outcomes" need to vote strategically when they get near the bottom of that list.
Alternatively, there is also no such thing as "strategy" in FPTP - "you just vote for one of your acceptable outcomes". That's... not a useful definition of the words involved.
Edited to add: It occurs to me that perhaps you weren't aware of the cutoff in the NYC case and were speaking of IRV with a full list? In that case I agree that strategy being necessary is substantially more rare, leaving aside for the moment whether it is sufficiently rare.
So, in the spirit of focusing where I can have the most impact, I've chosen to most directly support RCV.
First past the post is the enemy here. Not RCV, STAR, or Approval.
Which is because of advocation over the last (mostly) 2-3 years. We're only two states in, that's not significant momentum. FYI, we have approval in Fargo and Lane County/Eugene are close to passing STAR (I'd argue RCV hinders that effort due to confusion).
> First past the post is the enemy here. Not RCV, STAR, or Approval.
I'd disagree actually. The enemy is non-representative voting.
Why I'll actually fight IRV is because history. We've gone down this path before and had states/cities/counties use IRV and then revert back because it didn't solve the problems they set out to solve. So, what is different this time around? There's a "good enough" bar that needs to be met. IRV doesn't meet that.
Most importantly, IRV will not break us out of the two-party system, which is the root of many problems in American politics.
Are you aware that IRV significantly increases spoiled ballot rates, and disproportionately in low-income areas? It's actively harmful, roughly equivalent to knocking a couple percent off of the Black vote.
https://rangevoting.org/SPRates.html
https://www.yes2repeal.org/spoiled-ballots
Please reconsider your support for it. I am sure you have good intentions; my guess is that most IRV supporters probably aren't aware of the harm they're doing. I also thought IRV was an okay system (not perfect but better than plurality etc.) until I learned this information relatively recently.
I want to expand on why this claim is true, since IRV is frequently propositioned as a way to solve spoilers. Most importantly IRV does not pass the Favorite Betrayer Criterion[0] which states
> voters should have no incentive to vote someone else over their favorite
IRV only handles weak spoilers (e.g. Jo Jorgensen spoiling Trump) and not strong spoilers (e.g. Sanders spoiling Biden).[1] We're not actually concerned about Jorgensen spoiling Trump or Stein spoiling Biden. We're concerned with Sanders spoiling Biden (or vise versa). This is where the favorite betrayer criterion comes in.
IRV isn't even the best ordinal (i.e. ranking candidates) system because it fails the monotonicity criterion[2], which states
> A ranked voting system is monotonic if it is neither possible to prevent the election of a candidate by ranking them higher on some of the ballots, nor possible to elect an otherwise unelected candidate by ranking them lower on some of the ballots (while nothing else is altered on any ballot).
This is an extremely undesirable property and will continue to promote tactical voting. I can go into it more but maybe just check out Clay's works or ask him yourself since he's here in the thread. He's far more informed on this stuff than I am since he's an actual expert.
[0] https://electowiki.org/wiki/Favorite_betrayal_criterion
[1] https://electionscience.org/library/the-spoiler-effect/
[2] https://en.wikipedia.org/wiki/Monotonicity_criterion
https://clayshentrup.medium.com/momentum-e5fd12ffce2a
That's why it's IRV.
Alright, mind sharing your explanation?
Seriously, it's all of the disadvantages and very limited upside.
[1] https://rangevoting.org/TarrIrvSumm.html [2] http://zesty.ca/voting/sim/
https://rangevoting.org/SPRates.html
https://www.yes2repeal.org/spoiled-ballots
IRV means much higher rates of spoiled ballots, disproportionately in low-income neighbourhoods. It's actively harmful.
There are only so many combinations of rankings that are possible. It shouldn't be hard to encode each of those possible orders and transmit them as a whole.
O(n^2) isn't great though...
The algorithm to count is rough and requires many rounds (since you count, knock out the lowest candidate, recount, and repeat until a >50% favor is achieved by one candidate). Cardinal systems on the other hand just require summing the columns and taking argmax. This is far simpler (in fact we can do most of this in parallel making a far better runtime).
With 20 candidates in an election, that's over 2 billion billion possibilities. It's not practical.
I honestly thought that after learning Arrow’s impossibility theorem that Condorcet is not especially important. Since we can’t have a perfect voting system, we have to pick an imperfect one, and among the alternatives, it’s important to capture information from voters. Instant runoff captures a lot of information.
Approval voting I think is much too tactical and, strictly speaking, worse than IRV. No contest.
IRV lets people express their preferences in a fairly understandable way. The strategy I see people talking about is “rank all the front-runners, and use the leftover spots to rank the people you want to win, even if they’re not likely to win.”
So if A B C and D are front-runners, and E and F are the two other candidates you like, you come up with a ranking for those six and put the top five on your ballot.
The idea that people who like this system are “dangerously misinformed” or “dishonest” is needlessly inflammatory.
First off, Arrow's doesn't apply to all systems. You'll need to look into both Gibbard's and Gibbard-Satterthwaite's theorems.
Second off, just because you can't find a global optimization in a highly dimensional space doesn't mean there aren't local optimizations along criteria we care more about. Appealing to Arrow's is a cop-out.
> Approval voting I think is much too tactical and, strictly speaking, worse than IRV. No contest.
You're going to have to back this up with some strong evidence. Approval has higher VSE, is simpler, is more resistant to spoilers and tactical voting.
> IRV lets people express their preferences in a fairly understandable way.
Actually your argument holds true for any ordinal or cardinal system. Cardinal even having more flexibility since you can give two candidates the same score. And in cardinal if you want to rank your candidates, no problem. Better yet, you have better encoding opportunities because you can specify the distance between your ranking instead of the uniform spacing that ordinal systems force upon you. (BTW, given what the person above you wrote, I would assume that they know how ranked voting works and explaining how it is going to come off as you calling them dumb).
The issue here is not just how logical humans “homo economicus” behaves, but how actual voters behave.
Not really interested in engaging with the rest of this comment right now, but suffice it to say that I don’t think you’re accounting for human behavior in practice, which is messy and illogical. I don’t think it’s reasonable to say that I’m appealing to Arrow’s impossibility theorem as a cop-out—I’m saying that since we don’t have a game theoretic solution the the problem, we should look at the actual behavior of imperfect, irrational humans as the deciding factor.
I disagree but also don't see this as a problem. If you rank candidates you still get the majority of the desirable properties. Rank with non-equal distances, even better. Hell, it isn't even bad if you bullet vote (that's just approval voting). Investigating non-optimal ways of voting under any voting system is an extremely important analysis. So for the voter there is no problem. I'm also kinda put off that you give real world examples of humans using cardinal methods and then claiming that we can't understand it (HN is using cardinal voting...)
But we also have to consider the counting of votes side of "understandable." Plurality is pretty damn easy, and this is clearly why we use it. Approval is almost as easy (just just sum multiple columns). Range/Score isn't much harder. Then STAR introduces 2 rounds of counting. Then we look at IRV and we see that we have tons of rounds. This isn't typically so bad in a presidential election where there are realistically about 4 candidates, but that complexity increases real fast elsewhere. Just watch NYC. There's going to be at least 5 rounds (probably more). This is far more complex. We only have to look at Arizona to understand why this part of the "understandable" question is important.
Approval voting has some nice mathematical properties, but I think in practice trying to pidgeon hole people's preferences into a binary decision would be a major source of voter frustration and lead to tactical voting.
> but would still strongly prefer over one-or-more other candidates?
In fact, this is why I argue for STAR or score. It encodes information for better than a ranked (ordinal) system. In any ranked system you encode your preference with equidistant from one another. Where as when you score you can indicate a much stronger preference.
Here's an example. Let's say I REALLY like candidate A, moderately like candidate B, and strongly dislike candidate C.
Ranked:
A > B > C (with my encoding I'm saying that my preference of A over B is the same as my preference of B over C)
Scoring
A: 10, B: 7, C:0 (with this encoding I can indicate that my preference of A over B is not as large as my preference of B over C. Obviously we are capturing more information here)
Let's just encode information better, I agree. But also let's consider other factors like how easy it is to count the votes (which every cardinal system is going to beat ranked systems).
[0] That same preference where the distance between STAR, Score, and Approval is smaller than the distance between preference of Approval over IRV (e.g. STAR: 10, Score: 8, Approval: 7, IRV: 3, Plurality: 0).
I think STAR is slightly worse than Score, is comparable to Approval, is much better than IRV, and is likely better than Condorcet methods. STAR has some odd behavior which can be explored in a 3-person race. Score has less-problematic behavior caused by risk-taking with equilibrium voters (not like voters behave in any way similar to Nash equilibria though, and who knows what the real-world behavior will be).
However, STAR's main benefit may be in overcoming political resistance, if its properties are simpler to convince voters. Majority criterion sounds nice even when it is inefficient. Much like Top Trading Cycles losing to Gale-Shapley in school choice algorithms (excluding Boston).
https://www.rangevoting.org/BulletBugaboo.html
> I think STAR is slightly worse than Score
I would say they are probably roughly equal, but the best computer modeling we have shows that star tends to perform a little better.
https://electionscience.github.io/vse-sim/VSEbasic/
> However, STAR's main benefit may be in overcoming political resistance, if its properties are simpler to convince voters. Majority criterion sounds nice even when it is inefficient.
You're definitely correct on this.
A typical example where bullet voting should occur is a down-ballot election. Examples of low-engagement elections at https://en.wikipedia.org/wiki/Approval_voting#Other_organiza... bear out that these elections degrade in behavior to plurality. However, it can't be determined just from this behavior that IRV or Condorcet would do any better.
> https://www.rangevoting.org/BulletBugaboo.html
1-3 have major engagement from voters and I do not expect approval voting to suffer unduly from bullet voting. 4, an alumni association, should. However, the question then becomes, what about other voting methods? Because if the voter has weak information on the other candidates, he may simply bullet vote under all voting methods. In this context, Score needs testing. Voting values between 0 and 1 would give a much better idea of how voter engagement is affecting their votes, rather than their second choice always collapsing to zero. This is why I claim that it is "unclear to what extent it affects score methods in practice". Granular cardinal scores in even one such election would mean a lot for determining voter behavior.
I do not support FairVote's arguments and consider them dishonest. Notice that in my original post, I carefully describe bullet voting as a consequence of low engagement, which is supported by past real-world low-engagement elections. This is very distinct from FairVote's arguments, which are not sophisticated and demonstrate a willingness to make blind claims. The distortions that FairVote argues for require more justification than they give.
> https://electionscience.github.io/vse-sim/VSEbasic/
I don't believe much in that simulation. While I agree with your characterization of it as the "best computer modeling" in voting theory, it would still be considered a fatally flawed paper under the standards of most fields, and I faced some heavy obstacles when trying to analyze it. I think this field needs more/better academics. I described some of my objections at https://news.ycombinator.com/item?id=27612876
A robust cross-system analysis shows that approval voting is more robust to strategy than almost any other method. Most ranked voting methods, for example, fail the favorite betrayal criterion.
https://electionscience.org/library/tactical-voting-basics/
https://electionscience.org/library/tactical-voting-basics/
There's even an entire book focused on the game theory and tactics of voting methods, which advocates score voting, approval voting being score voting on a 0 to 1 binary scale.
https://electionscience.org/library/tactical-voting-basics/
Approval voting elections have been successfully held in 2020 and 2021 in Fargo in St Louis respectively, and there were no indications of voter confusion or anything like that.
https://electionscience.org/press-releases/st-louis-voters-u...
For any deterministic process of collective decision, at least one of the following three properties must hold:
1. The process is dictatorial, i.e. there exists a distinguished agent who can impose the outcome;
2. The process limits the possible outcomes to two options only;
3. The process is open to strategic voting: once an agent has identified their preferences, it is possible that they have no action at their disposal that best defends these preferences irrespective of the other agents' actions.
https://en.wikipedia.org/wiki/Gibbard%27s_theorem
----
So basically, it's impossible to completely eliminate strategic voting. No matter the method: ranked vs. cardinal vs. anything else you dream up can't help.
>Gibbard's theorem states that a deterministic process of collective decision cannot be straightforward, except possibly in two cases: if there is a distinguished agent who has a dictatorial power, or if the process limits the outcome to two possible options only.
.. it also mentions that gibbard's is specifically about irv.
All that aside, I can't understand the idea that we only get to change things once. My understanding of history is at odds with the concept.
No, Gibbard's is not limited to any particular voting method. I think you're misreading the next paragraph (and also confusing IRV, which is a particular method, with ranked choice, which is a whole category of methods). Note the distinction between Gibbard and Gibbard-Satterthwaite:
> A corollary of this theorem is Gibbard–Satterthwaite theorem about voting rules. The main difference between the two is that Gibbard–Satterthwaite theorem is limited to ranked (ordinal) voting rules: a voter's action consists in giving a preference ranking over the available options. Gibbard's theorem is more general and considers processes of collective decision that may not be ordinal: for example, voting systems where voters assign grades to candidates.
This isn't true
> I can't understand the idea that we only get to change things once.
Those of us concerned about IRV and promoting cardionality actually looked at history. Between 1910 and 1920 40 US cities used Bucklin voting (similar to IRV, slightly better even) and all repealed them[0]. So looking at history we see that people recognized the need for a better voting system, implemented something similar to IRV, saw that it didn't make things better, and subsequently said "fuck it, we'll go back to FPTP because it is easier." (I should also mention that in Australia, since 1949, 90% of Lower House elections, which use IRV, are equivalent to using FPTP[1])
So we're looking at history (and modern times) and saying "hey, this didn't work and actually ended up causing us to take a step backwards. Maybe we shouldn't repeat the same mistake."
I hope this clarifies our differing understanding of history.
[0] https://clayshentrup.medium.com/momentum-e5fd12ffce2a
[1] https://en.wikipedia.org/wiki/Australian_House_of_Representa...
What I would ask, then, is rather than not doing IRV, what should we do, in your opinion?
I'm looking at this as a sort of crisis situation, as our ability to affect our politics is in a state of constant erosion, and the process of capture at work here can only end in systemic collapse - the further power gets concentrated, the more centralized our decision making becomes, and the more vulnerable we become to systemic single point of failure.
I would love to see STAR voting become a thing. I think of IRV in the current context as a proof-of-concept that might show people that we can change the structure of voting. I can't see anyone wanting to go back to FPTP, because I don't really see anyone who thinks it's even remotely working.
I think this comment is a defeatist at best and deceitful at worst. Just because there is no global optimization does not mean that all optima are equal. We can in fact have optima that are better than one another (including all optima we know about!). This is a common feature of highly dimensional solution spaces.
The big issue here is that not all criteria are weighted equally, by desire of effectiveness. So we find an optima where we optimize features that have a large weights and care less about optimizing features with small weights. By doing this we can compare systems in their desirability and select the best ones. This is not cause for throwing up your hands and giving up.
As an example: cardinal systems, when compared to ordinal (ranked) systems, are more resilient to strategic voting and simpler (both for the voter and for the people tallying the votes, aka transparency). The cost? Slight decrease in maximal VSE. BUT if we look at the min, mean, median, or modes of VSE given different strategies cardinal system outperform ordinal (aka, desirable). You can see this by comparing with this chart[0]. For example with STAR0-10 we have maximal VSE of .983 and minimal of .912 (actually this makes it strictly better than plurality!). But if we look at our best ordian, RP, we see RP has a maximal VSE of .988 and minimal of .870. So on terms of maximal there's a 0.005 difference but on minimal there's a difference of 0.042! We can easily tell here that STAR is much more resistant to strategic voting than RP (Shulze is even worse!). Doing the same for IRV we see .07/.115 (max/min comparison of STAR0-10 vs IRV on VSE).
So we can compare. We can select better methods. But is there a ''perfect,, solution? No. But don't let the lack of the ability to create a perfect system detract from the ability to compare systems. Not all is lost.
[0] https://electionscience.github.io/vse-sim/vse.html
Wow, very HN! I don't actually mind very much, but seriously, consider applying the principle of charity?
I agree with you, and did before you wrote this comment too! Hence my use of the word "completely". I was just surprised that strategic voting couldn't be completely eliminated; before yesterday I expected that you could, in exchange for losing other nice properties. Of course being impossible to eliminate completely doesn't mean it shouldn't be minimized.
I thought Gibbard-Satterthwaite (or maybe just Gibbard's version) applied to cardinal systems as well. This seemed likely to me for some reason, as if an ordinal method could approximate any particular cardinal discretization by just including "ghost" candidates that can be packed (ordinally) between your actual candidates.
Correct, but it is also a weaker version of Arrow's (also as Clay points out, Arrow's isn't really about voting[0]...)
> This seemed likely to me for some reason, as if an ordinal method could approximate any particular cardinal discretization by just including "ghost" candidates that can be packed (ordinally) between your actual candidates.
Okay, but this just adds complexity. Cardinal is already simpler than ordinal systems (both for voters _and_ for those counting the votes). There's absolutely nothing wrong ranking candidates in a cardinal system (it's actually pretty unlikely that you'll have the same preference for multiple candidates so this is going to naturally happen). The difference? In cardinal you can better express your preference of one candidate over another (I give an example here[1]). So now we've added "encoding efficiency" to the added benefit of cardinal systems.
Cardinal systems are better than ordinal systems in almost every single way (the only thing I can think of ordinal systems doing better at is that RP and Schulze perform better on maximal VSE, but as I discuss here[2] that is pretty limited as well as unlikely considering strategic voting and the ability to manipulate people exists).
[0] https://news.ycombinator.com/item?id=27598975
[1] https://news.ycombinator.com/item?id=27599324
[2] https://news.ycombinator.com/item?id=27600248
Choice of language matters, and "ranked choice voting" (to solely mean "instant runoff voting") is a conceptually misleading term meant to load the debate. It dis-educates rather than educates on how voting systems work.
Advocates of other voting systems have to first start out by explaining this misdirection and unpacking the wrong mental model, then get to the merits of one voting system vs. another - because people have been told that "ranked choice voting" works a certain way and only a certain way.
Instant runoff is super easy to explain and achieves those desiderata, at least to some extent.
It’s hard to quantify the differences in mathematical terms against soft concerns like “we can educate voters and explain how this system works”. For that reason, I would like to see a non-mathematical argument against IRV, one that accounts for the other half of the voting problem.
Here's approval voting: "Upvote the candidates you like. The candidate with the most total votes wins." It's not just mathematically simpler, it's simpler in informal terms, a smaller change from current plurality voting, and it seems less messy in the strategic dynamics insofar as we've tested these different systems for real.
Except IRV doesn't. Not only are cardinal systems easier but IRV doesn't achieve the most desired feature that people are looking for: no spoilers (or alternatively put, allowing votes for 3rd parties without "wasting" your vote). In fact IRV increases spoilage while cardinal systems (which is alternatively proposed) decreases[0]. Meaning we're doing the opposite of what we're trying to accomplish.
As for simpler, I'll refer you to this[1]. The short story is just rate candidates. And if you end up ranking, no worries. It's more difficult to mess up.
[0] https://www.rangevoting.org/SPRates.html
[1] https://medium.com/election-science/star-voting-is-simpler-t...
https://www.rangevoting.org/CondorcetExec.html
On the contrary, approval voting gets better results than IRV with any measure of strategic or honest voters. See extensive computer simulation results by Harvard stats PhD and voting methods expert Jameson Quinn. Brown (50/50) is probably the most realistic setting.
https://rpubs.com/Jameson-Quinn/VSE5key
A simple example of IRV strategy is next year's Alaska senate race. Murkowski would beat either rival head-to-head but is likely to be eliminated based on first-place votes. So Democrats want to strategically rank Murkowski 1st in order to help her survive to beat Tshibaka (Trump Republican) so they at least get their lesser evil.
Here's a good comparison of approval voting vs. IRV by experts. Full disclosure, I was a CES co-founder and have written extensively on this topic for 15 years.
https://electionscience.org/library/approval-voting-versus-i...
> I honestly thought that after learning Arrow’s impossibility theorem that Condorcet is not especially important.
I've visited Kenneth Arrow at his home and co-founded a non-profit that interviewed him. His theorem only applies to social welfare functions, not voting methods, if properly understood. But if anything, the moral is to AVOID ranked voting methods and instead use rated voting methods.
https://www.rangevoting.org/ArrowThm
If you look at the graph with the honest strategy you get better results with IRV which is basically his point.
That's the concern though. You likely don't want to just "express your preferences." You need to strategically rank candidates. But then what are you being asked to do exactly when you go to vote? I don't think it would fly if the instructions on the ballot gave you guidance on how to "strategize" your rankings. But if it doesn't give that guidance, it's misleading.
And I understand that all voting mechanisms have their own unique pitfalls; so we have to fall back to pragmatic questions. And one major pragmatic question is: What's the status-quo/What is everyone already accustomed to? Asking a large voting body to switch methods is not easy and I think it's being motivated largely by a misleading "grass-is-greener" claim about rank voting.
In real world elections, the conditions where it is even theoretically possible arise only rarely (A > B > C with three choices, but B > A and A > C with two choices, and A's lead over B significantly greater than B's lead over C in the three choice scenario, and A's lead over C in the two choice scenario more than twice B's lead over C in the three choice scenario) and more importantly, they're not predictable enough beforehand. Advocating this kind of tactical vote stands at least as much chance of hurting your candidate as helping them, so nobody does it.
When you analyse real-world large scale IRV elections, you find that cases where the IRV winner isn't the Condorcet winner are rare, and this balances against the very real benefit of having a counting method that is easy to explain and understand.
[0] https://news.ycombinator.com/item?id=27598975
The idea is that the Republicans endorse only one candidate, Trump, and the Democrats endorse two: Bernie and Biden. In the non-tactical voting case, the second-last round votes shake out like:
Bernie is eliminated, the votes in his pile split 24 / 3 between Biden and Trump (a preference flow of 89% to Biden) and the final round of counting ends up: ..but tactical voting intervenes! A small number of Trump voters (2% of the total electorate) are organised to tactically switch their votes to 1. Bernie 2. Trump, resulting in a second-last round of: Now Biden is eliminated, the votes in his pile split 20 / 8 Bernie/Trump (a weaker preference flow of 71% to Bernie, because of Biden voters too conservative to vote for Bernie) and the final round is now: Tactical voting has won the day!However, this really illustrates the problems here for the prospective tactical voter:
1. They need pretty perfect information to pull this off. Not just on first round votes, but on how the preferences are going to flow as well. If those Biden votes flowed a little weaker to Trump, all they'd have done is elect Bernie instead; if Trump had been a little stronger overall than they expected in the no-tactical-voting, their tactical voting attempt might have backfired completely and turned a fair Trump win into a loss! Opinion polling just isn't this precise.
2. They have a co-ordination problem. If they switch too few votes, the scheme fails and they just elect Biden with a greater margin than before; too many and the scheme fails and they elect Bernie, a candidate they're presumably less happy with actually getting the Presidency than Biden.
3. All of this only works if the balls line up perfectly in the first place, even setting aside the problem already mentioned of how you know the balls are going to line up. If the second-last round votes are instead Trump 45 / Biden 30 / Bernie 25 then the Bernie to Biden preference flow has to fall under 66% for the scheme to be possible.
Rather than trying to engage in this dubious and risky tactical voting attempt, the Trump campaign would be far better served just spending their resources trying to turn out more of their voters. After all, the theoretical possibility only exists when the margins are tight in the first place.
A tangential matter is that even in an IRV election it still probably makes sense for the parties to either endorse only a single candidate each - otherwise their candidates will waste some of their resources attacking their co-party candidate when they could have launched them against their main opposition - or at least mutually agree to distribute campaigning material advising to give a second preference to their co-party candidate (a so-called "preference swap" arrangement).
That’s a neat visualization. The step where mushroom pulls ahead of sausage really shows the flow.
Though I do question the integrity of a pizza topping poll that manages to exclude pineapple-ham, not individually, but as a joint entry. Did it miss the filing deadline?
> And you can see Mayor Bill de Blasio eating a pepperoni pizza as a result of this here:
Worst mayor ever eating one of the worst pizza toppings ever.
(Seriously even Google dedicates multiple pages of results to De Blasio: https://www.google.com/search?q=worst+mayor+ever)
After all, in this example the plurality vote winner is still the clear, uncontested winner in ranked choice, so this example makes it look like ranked voting is just a way of complicating a process that "just works". I know this is not actually true, but this is how some people see alternative voting methods.
On the other hand, with acceptance voting you would probably get a different result as most meat eaters are indifferent towards mushrooms while many vegetarians would be passionately against pepperoni. So, I suspect that with acceptance voting the results could be switched, for better or for worse.
I'm hoping RCV spreads quickly enough to more states to break the political gridlock and intense partisanship.
You want people who believe science, have a sense of foresight for several years, understand international politics, and how resources and economics work to decide the future of the country.
https://electionscience.org/library/the-center-squeeze-effec...
You can't just rank the candidates you like, in the order you like them, because your genuine top 5 choices might not include the people who are most likely to win. You'll be "throwing your vote away" unless you include the most-likely-to-win candidates - that is, unless you vote tactically.
I've seen lots of discussion in NYC where people are trying to figure out the right tactical voting to do given their views and given the fact that they can only rank 5 candidates. Ranked choice voting is supposed to make tactical voting essentially irrelevant, so this is disappointing.
I voted today and put 3 small candidates and two who "have a chance". I'm really excited to see how those 3 candidates shake out in the rounds, and that alone might give them a lot more power next race. That's super exciting to me and I think we can celebrate that and then push to improve the issues after :)
I do appreciate the progress, but I worry that people will regard this incarnation of ranked choice voting as "more complicated" when the entire point is that it's less complicated for the individual voter...
RCV is, at a maximum, equally as complicated as FPTP style voting since you have the option to elect only a single choice on your ballot (unless they're doing something weird). The form of the ballot may be visually more overwhelming but that's just from unfamiliarity.
1. http://maristpoll.marist.edu/wp-content/uploads/2021/06/WNBC...
A voting system where you're limited to only voting for the incumbent is indeed very simple (although not very democractic)... I assume you meant "limit yourself to the frontrunners". :)
In an election where there's exactly two candidates, RCV (or most other alternative voting systems) is the same as simple majority. You're right that with exactly two candidates, tactical voting isn't a concern, and simple majority voting is the simplest and best system. But many elections do not have exactly two candidates - primaries being the most salient example.
With RCV/alternative voting systems, you don't need to know who's popular or "electable" or any of that. You can just go with your most naive, direct take on the candidates.
Say you like two candidates A and B, but most of all you want to make sure candidate Z loses. Just rank A and B first, then everyone else equally, then candidate Z. You don't have to think about who is most likely to beat Z - you don't even have to follow the election at all.
I wasn't aware you could rank candidates equally, that makes it simpler indeed.
I'm not sure what you're describing. As long as one of the top two candidates appear on your ballot (at any rank) then you have influenced the last round of voting.
45% of voters rank A 1st, B 2nd, C 3rd
25% of voters rank B 1st, A 2nd, C 3rd
30% of voters rank C 1st, B 2nd, A 3rd
Nobody has the majority, so B is eliminated, and A wins round 2 with 70%. But 30% of voters preferred B over A, and if they had ranked B first, then B would've won in the first round with 55%. By voting honestly, the C voters threw away their votes. None of their later rankings mattered, because C was not eliminated before a victor was declared.
But suppose you're right and voters rank candidates honestly because they can't predict how to vote strategically; ultimately, I think that is arguably even worse, because even if voters can't predict how to vote strategically in advance, they can certainly still feel resentful after the fact when the results are tallied and it becomes clear that they got screwed over by voting honestly. At least in plurality voting, it's easy to understand the tradeoffs and choose whether you want to vote for a spoiler candidate or not. Surely it will not lead to greater voter enfranchisement for people who find out after they voted that they accidentally spoiled their own votes, and could've had a say in the outcome if they'd expressed their rankings differently.
If this still all sounds hypothetical, keep in mind that this is exactly what happened in the 2009 Burlington, VT mayoral election. The Republican candidate acted as a spoiler for the Democrat (who had the broadest popular support) and instead the Democrat was eliminated and the third-party Progressive candidate was elected. This led to wide-spread dissatisfaction and IRV was repealed there in 2010.
45% of voters rank Trump 1st, Romney 2nd, Biden 3rd
25% of voters rank Romney 1st, Trump 2nd, Biden 3rd
30% of voters rank Biden 1st, Romney 2nd, Trump 3rd
This seems like a fairly plausible real-world situation, and it also seems pretty easy to imagine how Democrats could look at the polling numbers and realize that Biden is never going to win in their state, and that ranking Romney 1st is their only chance to stop Trump.
One important factor the article didn't quite emphasise is how this voting system can break up the two party duopoly without creating 1920s-Reichstag-esque anarchy.
It's completely normal to put a third party as your first preference, while still indicating that you'd prefer Kodos to Kang.
In practical terms, it means the Dems would actually have to do some work in order to win the black vote, and likewise for the Republicans and working class white people.
https://en.wikipedia.org/wiki/Electoral_system_of_Australia#...
-- an Australian/American living in the US.
https://en.m.wikipedia.org/wiki/Mixed-member_proportional_re...
Edit: I may have confused RCV with STV. Regardless, it would be nice to know that if my first option didn't make it into Parliament that I could prefer a second party, and so on.
So ranked choice voting is a super-set with many flavours within.
Open lists are better, but I believe they still have the problem that they make it unclear where representatives' mandates come from, especially if a representative can resign and be automatically replaced by someone lower on the list without a by-election.
Other problems exist with list-based systems, such as treating independent candidates differently to candidates with established parties, and the possible use of "decoy lists"[1].
The best implementation I've seen of MMP, though, is that used in Baden-Württemberg, which avoids party lists by filling the proportional seats with the candidates who came closest to winning the elections in their local districts.[2]
[0] https://en.wikipedia.org/wiki/Closed_list
[1] https://en.wikipedia.org/wiki/Additional_member_system#Decoy...
[2] https://en.wikipedia.org/wiki/Zweitmandat
Even more fundamentally, politics is a "team sport". The vetting of individuals when power boils down to alliances and other relations with the politicians is stupid waste of voter's time.
By voting for a small party you risk wasting your vote.
MMP would probably work well in South Australia (where overall the electorate is fairly evenly split right-left, but many of the right-leaning voters are locked up in a few geographically large rural electorates and the left-leaning voters are distributed more evenly across more urban and suburban electorates).
Under cons, it says that voters have to rank all candidates on the ballot paper.
In reality you have a choice. You can either rank all, or if you are lazier you can just select your favourite candidate and then your first choice candidate's preferences will be used instead.
Also for some larger ballots (usually the senate with nearly 100 options) there is now a requirement to only rank say the top 10 or so for the ballot to count, so you don't need to number all 100.
This is one of the best features of the Australian system. If you want to do the basic effort you can just tick one box, but if you care about the ordering you can also make your preferences count if you so choose.
As an outsider looking in to American politics, I feel changing to preferential voting is the best bang for buck change to move away from extreme politics. Hopefully this catches on elsewhere.
This is not true. In the elections for the Australian House of Representatives (single-member electorates, comparable to the mayoral election under discussion where there is a single winner), you must number every box for your vote to count.
In elections for the Australian Senate (multiple member electorates) you don't have to number every box, but you have a choice: you can either vote "below the line" for individual candidates, in which case you must number at least 6 boxes for your vote to count (the ballot paper advises you to number at least 12); or you can vote "above the line" for groups of candidates (in which case you only have to number one, but the ballot paper advises you to number at least 6). A vote "above the line" for a group is equivalent to numbering the candidates in that group in order from top to bottom, but it doesn't imply a vote for a candidate in any other group.
The voting systems in Australian states and territories vary from the above; some of them do allow only numbering a single candidate in single-member elections ("Optional Preferential Voting" or OPV) and some still have group-ticket voting in multi-member elections where you can assign your vote to the preference ticket submitted by a candidate (though hopefully the last few jurisdictions to still have this will be getting rid of it, because it is being gamed).
In Canada in 1952, the province of British Columbia switched to a ranked ballot and it nearly ended in disaster.
The ranked ballot was an attempt by the traditional Liberal and Conservative parties to keep a rising socialist party from gaining power, as they assumed no Liberal or Conservative voter would choose the socialists as their second choice.
But there was so much animosity between the parties that most voters chose a fringe right-wing party as their second choice, as a protest vote.
The two traditional parties were wiped out and the fringe right-wing party narrowly won, despite having no leader and little governing experience.
Fortunately, saner heads took over the fringe party and quickly steered it to the political center as a sort of big tent party for former Liberal and Conservative voters. But it could have gone very differently.
The ranked ballot idea was quickly dropped, but the event shaped British Columbia politics for the next 60+ years. That fringe party elected by accident became the province's natural governing party.
I suppose the moral of the story is that no voting system can prevent irrational voters from shooting themselves in the foot.
The two systems are so dramatically different that no cautionary tale would ever be meaningful.
Proportional representation is what is desperately needed to break the two party system.
I'm not convinced this is true. 2 parties control 80%. 2 coalitions control 95%. I'm also not sure how you can look at America and not recognize that our parties are more similar to coalitions (come on, AOC/Bernie are not the same party as Pelosi/Biden).
A better explanation is that Australia uses a parliament which has proportionate representation. My evidence of this is that other parliaments without IRV have similar party structures as Australia. So we have examples where you remove one aspect and things don't change, I don't think that makes for strong evidence that the thing removed is causing the effect (how can it cause an effect somewhere that it doesn't exist!).
So don't compare Australia vs America to prove IRV creates more parties. Compare Australia/France/UK vs America to prove parliament causes the desired outcome.
Australia adopted the preferential system in 1918.
> It's completely normal to put a third party as your first preference, while still indicating that you'd prefer Kodos to Kang.
The evidence shows most voters don't, which is why Australia's IRV elections still maintain a duopoly. In spite of the fact that the senate, which uses (proportional) STV, has escaped duopoly.
http://scorevoting.net/AustralianPol
I've always seen the proportional nature as being key to other countries lack of a two party system (though they have two major coalitions). I'm curious why you think many look to Australia's IRV as the reason for it being a multi-party system when many parliamentary systems that don't use IRV have even more parties. How do you combat this common misconception? Also it seems dishonest to not consider coalitions when US parties aren't as uniform as most other country's parties. If we look at coalitions it does not appear that Australia (or most others) have escaped duopoly.
And as always, good to see you in the comments and thanks for all the information.
The most spectacular example of this was the election of three rural independents who then proceeded to install Julia Gillard from Labor as Prime Minister, even though they were conservative. Their stated reasons were Labor's promise to build a fibre-to-the-premises broadband network and their assessment of the conservative leader. Other examples include members in wealthy Sydney suburbs who, while otherwise conservative, wanted action on climate change, and members in farming electorates unhappy with the way the Nationals have been captured by mining and housing development interests and failed to represent farmers.
(The Nationals, formerly the Country Party, always join the conservative Liberals in coalition, apparently having forgotten the purpose of coalitions.)
An exception is the Greens, who usually score one seat based on the central business district of Melbourne. This includes very high density apartment towers, and student accommodation for the University of Melbourne, and is demographically odd.
Many observers believe the rural independents may eventually coalesce into a new party which displaces the Nationals. I see no evidence thay the Liberals' several humiliating losses to independents in "safe" seats is causing them to rethink their policies. This may or may not change once they are unable to rely on whatever replaces the Nationals to form government.
The entire point of elections is proving your government is trustable, you can't trust your government with it because then it's like a self signed certificate, circular trust problem.
Who rigs elections in Russia? The government. Who is most likely to rig elections? The government.
You can't trust government with elections, and just because the government shows you documents doesn't mean anything. Governments fabricate documents all the time. If all you require from corrupt politicians to remain in control is a bunch of mailed documents, you can be sure they'll get those documents mailed. I'm just completely dumbfounded people living within countries with these elections think they are democracies.
Can you say more about this? So at each polling place 4 workers are assigned 800 votes, and then they rotate out to a different group of 4? I don't see how that inspires more trust than our current system, the government could simply be lying that they rotated the 4 poll workers out. Or, in a large city this would be a complete cluster as you'd need hundreds of poll workers on standby to rotate in & out every 30 minutes or so. I would definitely be interested in hearing more though
It’s like if “Democratic Elections” became a specific name for Plurality voting.
The problem with IRV is that you can have a candidate who is everyone's second choice, but then loses out on the first ballot because too few people ranked them first.
Would screening for a condorcet winner (beats all winner) FIRST then proceeding to RCV address this concern sufficiently?
Ballot completion would remain unchanged.
This page is a good explainer of IRV's weaknesses and shows some of the alternatives: https://ncase.me/ballot/
Since you mention it, let me offer a candidate for "worst possible voting method (that has actually been used in a democracy)", namely "plurality-At-Large".[0]
It basically takes the worst properties of FPTP and extends them to an entire governing body, allowing a party with less than 50% support to win 100% of the seats (by design).
Fortunately it has been mostly abandoned in the US thanks to the Voting Rights Act and subsequent SCOTUS decisions.
[0] https://www.nonprofitvote.org/the-bias-of-at-large-elections...
A close variant (vote for up to N, top N win, at large) is used in a lot of local elections still; all multimember district methods are banned in House elections specifically to prevent plurality at large from being used there. (Though the Senate uses temporally-segregated two-seat plurality at large in every state.)
Correction: it uses either that or similarly segregated majority/runoff, varying by state.
Majority/runoff and plurality are similar FPTP variants, but majority/runoff is slightly less bad.
Wikipedia has an example:
https://en.wikipedia.org/wiki/Monotonicity_criterion#Instant...
I think first-past-the-post and IRV are both bad voting systems that are unsuitable for real-world use in elections with more than 2 candidates. Approval voting is simpler than IRV and (as far as I know) it isn't worse than FPTP in any way, so I think we should push for that.
FPTP might be better than approval voting, because approval voting favors weak moderate candidates with no agenda.
With FPTP the people alternate lefts and rights and the system develops. With moderate candidates the system may stop progressing: no reforms, no bugfixes.
The way I look at is that it's a kind of relief valve. If both major parties (assuming there are two, maybe there'll be more...) nominate not-so-great candidates, then the voters have an option to elect someone like Tom Hanks. I mean, I would have voted for Tom Hank in 2016 if I had an approval voting ballot and he was one of the options.
No voting public prefers other candidates, and moderate candidate as backup.
I give you an example I posted in another thread:
* 60% want a dem candidate to win
* 40% want a rep candidate to win
* 70% are willing to join strategic voting to avoid allowing other party to win
In this case, the dem candidate looks like a correct winner.
With ranked choice the the dem candidate would win.
But with approval voting strategic "moderate" candidate wins. If they are moderate, that's good. But they can easily be just weak nobody.
It is not exactly what voting public prefers. That choice is a fake choice, like a girl, do you want to go to bed or be thrown into a forest with angry wolves.
> if there's no particularly compelling reason for people to want that person to be president.
The reason is to not let opposing party win. Strategic voting.
> If both major parties (assuming there are two, maybe there'll be more...) nominate not-so-great candidates, then the voters have an option to elect someone like Tom Hanks.
And they will if it is some form of ranked choice, but not just approval.
In approval voting, strategic voting for moderate/novelty candidates is like this (assuming you have a favorite and least favorite major party candidate and the novelty candidate is somewhere in the middle): you can vote for the novelty candidate if you want. Doing so might cause your favorite major-party candidate to lose if you happen to be the tie-breaking vote. Or it might cause your least-favorite major party candidate to lose to the novelty candidate. You can decide whether the prospect of having your favorite candidate lose to the novelty candidate is worse than the prospect of having your least favorite major party candidate lose to the novelty candidate and vote accordingly. In all cases, if you vote for your favorite candidate you're always helping them win, and if you vote for the novelty candidate you're always helping them win.
How strategic voting works for RCV is that you can put your favorite major party candidate first, the novelty candidate second, and the least favorite candidate last. You know that putting the novelty candidate in the middle won't harm your favorite candidate's chances of winning, since that's a property of RCV (later no harm). However, what you don't know is whether putting your favorite candidate first will help or harm your favorite candidate, and you don't know whether putting the novelty candidate second will help or harm the novelty candidate. Maybe putting your first choice second or third will cause them to win. Maybe putting the novelty candidate first will cause your favorite candidate to win against your least favorite candidate.
This is complicated and hard to reason about, unless you have detailed information about how other people are going to vote. Which seems to imply that such a system would tend to disempower individual voters and empower whoever has access to detailed information about the intentions of the voting public: campaigns that can afford internal polling, Facebook, Google, etc..
For parliaments, however, the system favors the two main parties, since a candidate needs those 50% support. This means parliament, which _should_ reflect the pluralism of society, will not actually reflect that pluralism.
In effect, this ranked ballot business is a way to appear to engage in voting reform without actually wanting to change anything about the outdated 2-party system.
In Canada, the liberals promised election reform. Until it turned out that ranked ballot was not going to be recommended by the electoral reform commission. Since that was the system the Liberals wanted in order to improve their chances in future elections, and they had little interest in actually representative voting systems, they scuttled the process and reneged on their promise that this would be the last election using the first past the post system.
Hopefully you can support any change locally that gets us off of first past the post. Once we get some voting changes, hopefully that makes future changes even easier to make... though I know that's wishful thinking.
Actually, that's debatable. There was an interesting simulation done awhile back [1] that compared voting systems, and they found that if people voted honestly then IRV is better, but if people vote strategically the result isn't any more democratically optimal than FPTP. Which is what we should probably expect to happen when voters figure out that it's only safe to put your first choice first when they're the overwhelming favorite to win or they have no hope of winning. Otherwise it's complicated: you could cause your favorite candidate to lose by ranking them too high.
Obviously this is one simulation and results would be different with different assumptions. But I think it's worth pushing back on the idea that IRV is way better. It solves the problem of minor candidates being spoilers, but what if people actually start voting for minor candidates and they become major candidates? With 3 or more major candidates IRV starts looking like not such a great system.
The problem with talking about this is that the problems with FPTP are readily apparent and easy to explain. The problems with IRV aren't obvious and are harder to explain.
[1] https://electionscience.org/library/tactical-voting-basics/
But minor candidates being spoilers have come up at all levels of elections many times for decades.
Splitting the vote happens frequently, again at all levels of elections (at least nationally in terms of the primaries for US Presidents).
If you vote for your favorite candidate, and your candidate is not the last-ranked candidate, you cannot cause your candidate to lose by putting them first.
But it sounds like you’d agree that FPTP needs to go. Great, me too. That’s the real focus for me: let’s start using different methods that eliminate, even to small degrees, the problems of FPTP.
> If you vote for your favorite candidate, and your candidate is not the last-ranked candidate, you cannot cause your candidate to lose by putting them first.
Wikipedia has an example. (Technically it's two voters difference, since a change of one voter's vote would, I think, change the result at most from a win to a tie.)
https://en.wikipedia.org/wiki/Monotonicity_criterion#Instant...
But a great thing about RCV is that it's not inherently partisan or favoring of left or right. As a result, in addition to being used in places like New York City, San Francisco, and Maine...it recently passed for use in Alaska and nearly two dozen cities in Utah. That's a pretty diverse set of locales across the U.S.
Article on Alaska: https://www.vox.com/2020/11/19/21537126/alaska-measure-2-ran...
Article on Utah: https://www.sltrib.com/news/politics/2021/05/11/many-utahns-...