3 ms·

You're saying that "later no harm means ... your vote for your second can't hurt your third", which seems strange, given that (by definition) it means your vote

You're saying that "later no harm means ... your vote for your second can't hurt your third", which seems strange, given that (by definition) it means your vote for your third can't hurt your second. If you're right, then your second and third preferences can't harm the chances of each other, and it seems like this result must generalize such that your Nth can't hurt your Mth, for all N and M. That would mean that every possible vote has no effect on the outcome of the election, which seems like a contradiction. Obviously I'm missing something, but I'm happy to put that down to the complexity and unintuitive outcomes of RCV/IRV, since Approval Voting is a much simpler system to reason about (and implement). I still disagree that Approval's lack of later-no-harm is a feature, though. In practice, what it means is that voters have to think strategically about how well they think the various candidates will do, in order to decide whether they have to "risk" giving an approval to a candidate who they genuinely don't like, but who might be the only chance of preventing a detested candidate from winning.

ClayShentrup 4y ago
i didn't say your 2nd can't hurt your 3rd, but it can't *IF* your 2nd is eliminated before your 1st. which often happens. for instance, a simplified version of what just happened in alaska: 35% peltola begich 33% palin begich 18% begich peltola 14% begich palin -- https://electionscience.org/commentary-analysis/rcv-fools-palin-voters-into-electing-a-progressive-democrat/ https://electionscience.org/commentary-analysis/rcv-fools-pa... begich is eliminated first, even though he's preferred to each of his rivals by a sizable majority. but more notably here, palin is a spoiler. her supporters get peltola—their least favorite. BUT if they insincerely/strategically rank begich in 1st instead of palin, then begich wins. i.e. they get their 2nd choice instead of their 3rd. so they have to LIE ABOUT THEIR FIRST CHOICE in order to prevent the spoiler effect. whereas this is mathematically proven never to happen with approval voting. > I still disagree that Approval's lack of later-no-harm is a feature, though. In practice, what it means is that voters have to think strategically about how well they think the various candidates will do, in order to decide whether they have to "risk" giving an approval to a candidate who they genuinely don't like, but who might be the only chance of preventing a detested candidate from winning. you can't argue about strategic voting because: 1. as i just showed you, IRV/RCV has an even more severe vulnerability to this—you have to strategize about whether to support your FAVORITE, not whether to support your 2nd favorite. 2. approval voting gets better results with strategic voters than IRV does with HONEST voters, so even in the worst case scenario, approval voting is still better. see these VSE calculations by harvard stats PhD jameson quinn. https://electionscience.github.io/vse-sim/VSEbasic/ https://electionscience.github.io/vse-sim/VSEbasic/ my link already discussed all of this quite clearly.
- dane-pgp 4y ago
  Thank you for clarifying what you meant. Your comment is much easier to read than the article you linked to. > if they insincerely/strategically rank begich in 1st instead of palin, then begich wins. i.e. they get their 2nd choice instead of their 3rd. So it sounds like the real complaint here is that IRV fails the "favorite betrayal criterion"[0] which requires that "voters should have no incentive to vote someone else over their favorite". You're right, that's a genuine flaw of IRV, and one that Approval Voting doesn't suffer from. Unfortunately, the strict definition is slightly unhelpful here, because in Approval Voting the only way to rate a candidate "over" your first preference is to not approve your first preference, which of course will harm them. So it's almost impossible for Approval to fail this criterion just by the lack of options it gives to voters. Instead we should be considering if voters have an incentive to vote for another candidate over, or as equal to, their first preference. This criterion doesn't have a name, but it's a criterion that Approval doesn't satisfy, for the same reason as your Alaska example. Supporters of Palin have an incentive to approve Begich, i.e. give him equal support on their ballot paper to their actual first preference. You might say that this is fine, because it allows them to get their 2nd preference as the winner instead of their 3rd, but if other people had ended up voting different on the day, then giving this extra support could have denied the Palin supporter the Palin win they were hoping for. > you can't argue about strategic voting You're probably right that IRV has more scope for strategy than Approval, and I might even grant that Approval gets better results in practice, regardless of the amount of strategic voters, but I still think that IRV has a better UX and requires less cognitive load (as long as voters don't need to number every candidate, although that makes IRV even less fair). My point is, as a voter going to fill out your ballot, you know what it means to like one candidate more than another, and you can ignore the candidates you don't know about, and you can probably pick an order randomly if there are two candidates who you like equally well (and randomized ballot ordering helps here, although it potentially hinders illiterate people who need to see the ballot paper long in advance to memorize how to fill it in). By contrast, being asked which candidates you "approve" of basically can't be answered without having to think strategically, or at least it will mean different things to different people. For some people, "approve" means "I'd still be happy if this person wins", but for others it means "I'd grudgingly accept this person winning as at least they're not the worst candidate on the list". The decision for how to interpret the question will be subject not just to personal psychological biases, but candidate campaigning, and misinformation, and will lead some voters into thinking that they need to pick this "threshold for approval" based on a deep understanding of the polling data (and therefore without that, it might be better not to vote at all). In conclusion, I don't think this is a reason not to support Approval Voting, especially where the alternative is IRV, but I do think that as electorates get a chance to use it in practice (and campaigns adjust to the new realities it brings), people might start to be annoyed at its flaws. Fortunately those flaws are smaller than those of IRV, and minuscule in comparison to FPTP, but I still think that voting reform advocates can do even better, in terms of UX and simplicity. [0] https://electowiki.org/wiki/Favorite_betrayal_criterion https://electowiki.org/wiki/Favorite_betrayal_criterion

ClayShentrup 4y ago

i didn't say your 2nd can't hurt your 3rd, but it can't *IF* your 2nd is eliminated before your 1st. which often happens. for instance, a simplified version of what just happened in alaska: 35% peltola begich 33% palin begich 18% begich peltola 14% begich palin -- https://electionscience.org/commentary-analysis/rcv-fools-palin-voters-into-electing-a-progressive-democrat/ https://electionscience.org/commentary-analysis/rcv-fools-pa... begich is eliminated first, even though he's preferred to each of his rivals by a sizable majority. but more notably here, palin is a spoiler. her supporters get peltola—their least favorite. BUT if they insincerely/strategically rank begich in 1st instead of palin, then begich wins. i.e. they get their 2nd choice instead of their 3rd. so they have to LIE ABOUT THEIR FIRST CHOICE in order to prevent the spoiler effect. whereas this is mathematically proven never to happen with approval voting. > I still disagree that Approval's lack of later-no-harm is a feature, though. In practice, what it means is that voters have to think strategically about how well they think the various candidates will do, in order to decide whether they have to "risk" giving an approval to a candidate who they genuinely don't like, but who might be the only chance of preventing a detested candidate from winning. you can't argue about strategic voting because: 1. as i just showed you, IRV/RCV has an even more severe vulnerability to this—you have to strategize about whether to support your FAVORITE, not whether to support your 2nd favorite. 2. approval voting gets better results with strategic voters than IRV does with HONEST voters, so even in the worst case scenario, approval voting is still better. see these VSE calculations by harvard stats PhD jameson quinn. https://electionscience.github.io/vse-sim/VSEbasic/ https://electionscience.github.io/vse-sim/VSEbasic/ my link already discussed all of this quite clearly.

dane-pgp 4y ago
Thank you for clarifying what you meant. Your comment is much easier to read than the article you linked to. > if they insincerely/strategically rank begich in 1st instead of palin, then begich wins. i.e. they get their 2nd choice instead of their 3rd. So it sounds like the real complaint here is that IRV fails the "favorite betrayal criterion"[0] which requires that "voters should have no incentive to vote someone else over their favorite". You're right, that's a genuine flaw of IRV, and one that Approval Voting doesn't suffer from. Unfortunately, the strict definition is slightly unhelpful here, because in Approval Voting the only way to rate a candidate "over" your first preference is to not approve your first preference, which of course will harm them. So it's almost impossible for Approval to fail this criterion just by the lack of options it gives to voters. Instead we should be considering if voters have an incentive to vote for another candidate over, or as equal to, their first preference. This criterion doesn't have a name, but it's a criterion that Approval doesn't satisfy, for the same reason as your Alaska example. Supporters of Palin have an incentive to approve Begich, i.e. give him equal support on their ballot paper to their actual first preference. You might say that this is fine, because it allows them to get their 2nd preference as the winner instead of their 3rd, but if other people had ended up voting different on the day, then giving this extra support could have denied the Palin supporter the Palin win they were hoping for. > you can't argue about strategic voting You're probably right that IRV has more scope for strategy than Approval, and I might even grant that Approval gets better results in practice, regardless of the amount of strategic voters, but I still think that IRV has a better UX and requires less cognitive load (as long as voters don't need to number every candidate, although that makes IRV even less fair). My point is, as a voter going to fill out your ballot, you know what it means to like one candidate more than another, and you can ignore the candidates you don't know about, and you can probably pick an order randomly if there are two candidates who you like equally well (and randomized ballot ordering helps here, although it potentially hinders illiterate people who need to see the ballot paper long in advance to memorize how to fill it in). By contrast, being asked which candidates you "approve" of basically can't be answered without having to think strategically, or at least it will mean different things to different people. For some people, "approve" means "I'd still be happy if this person wins", but for others it means "I'd grudgingly accept this person winning as at least they're not the worst candidate on the list". The decision for how to interpret the question will be subject not just to personal psychological biases, but candidate campaigning, and misinformation, and will lead some voters into thinking that they need to pick this "threshold for approval" based on a deep understanding of the polling data (and therefore without that, it might be better not to vote at all). In conclusion, I don't think this is a reason not to support Approval Voting, especially where the alternative is IRV, but I do think that as electorates get a chance to use it in practice (and campaigns adjust to the new realities it brings), people might start to be annoyed at its flaws. Fortunately those flaws are smaller than those of IRV, and minuscule in comparison to FPTP, but I still think that voting reform advocates can do even better, in terms of UX and simplicity. [0] https://electowiki.org/wiki/Favorite_betrayal_criterion https://electowiki.org/wiki/Favorite_betrayal_criterion