9 Matching Annotations
  1. Feb 2019
    1. Any election system that favors extremists would be considered unreasonable; the same rationale must be applied to moderates.

      Utter nonsense. To paraphrase:

      Any election system that favors unrepresentative candidates [like IRV] would be considered unreasonable; the same rationale must be applied to one that favors representative candidates.

      Uh, no. That doesn't follow.

      FairVote starts from the conclusion that IRV is the best voting method, and then works backwards to try to justify it, in this case arguing that a flaw of IRV is actually a feature, by making a false equivalent between a voting system that favors unrepresentative candidates and one that favors representative candidates.

      The whole point of an election is to find the most-representative candidate.

    2. Agreeing that the Condorcet criterion is desirable is equivalent to saying that moderate candidates should always win.

      Yes, candidates who are moderate relative to the voters should always win.

      The goal of an election is to find the candidate who best represents the electorate. If the electorate is left-wing on average, the winner should be too. If the electorate is "strong on both personal freedoms and economic freedoms", then the winner should be too.

      Anything else is undemocratic.

    3. Condorcet winners are centrist by nature, regardless of the preferences of the electorate.

      This isn't true. It's possible for a Condorcet candidate to be extremist relative to the other candidates or the electorate, since weak preferences are given equal weight to strong preferences. Simple example here: Condorcet winner is not utilitarian winner

    4. not necessarily liked more than other candidates

      This is true, but IRV doesn't choose the candidate who is most-liked (the "utilitarian winner"), either.

  2. Nov 2018
    1. Under range/score, the best strategy to promote the election of a preferred candidate is always to give that candidate the maximum score and then give every other competitor the minimum score.

      Yeah, this is false.

      If you have perfect knowledge of how everyone else is voting (and you usually don't), then the best strategy is to give a maximum score to the frontrunner that you prefer, and also to everyone you like more than them, and to likewise give a minimum score to the other frontrunner, and to everyone you dislike more.

      This is not bullet voting; it's equivalent to Approval Voting, and leads to more moderate winners who are good representatives of the electorate.

      Real-world Score elections don't show this behavior, anyway, because polls are imprecise and the consequences of voting honestly under Score aren't as dire as they are under FPTP or IRV.

    2. Rebuttal to [the original version of] this page at https://www.equal.vote/fv

    3. unlike RCV, it would be subject to tactical voting

      This is nonsense. All voting systems are subject to tactical voting.

    1. On the theoretical front, approval and score voting fail a critical test that voting theorists call the majority criterion.

      "Failing the majority criterion" is actually a good thing, because these systems find the candidate who is most acceptable to the entire population, rather than the candidate who is most acceptable to half of the population.

      Majoritarianism that ignores the desires of half of the population leads to adversarial politics, polarization, and eventually even to violence and civil war.

    2. and works well in practice

      ...which is why it subsequently gets repealed?