About the author

Admin

I hope you enjoy the content we have on this site! If you want to reproduce any articles we write then we appreciate if you give credit to defenseofreason.com and provide a link to our page. However, the ideas themselves are more important than this site, so feel free to adapt the portions you agree with to use in your own writings or videos (as long as you do not copyright portions that include our work).

Related Articles

8 Comments

    1. 1.1

      Aaron

      I definitely prefer the concept of Score Voting. But I think it has the practical problems I mentioned in the article, and would only be ideal if it were possible to forcibly extract honest votes from people. On your blog post, there were a few parts where I think you either made errors or I misunderstood:

      //One can make the argument that any voter so naive as to be unaware of strategy would actually be better off casting a sincere Score Voting ballot than an Approval Voting ballot. This is based on analysis of the effectiveness of these respective behaviors. I.e. observe that the values in column A tend to be lower than those in column D in the first table here.//

      I don’t understand this part of the article, because the chart to which you refer shows that A (Score Voting) is *less effective* at achieving desired results than D (Approval Voting). However it’s interesting that the Borda system represented by C fares better than I would have expected (even better than Approval if there are just 3 candidates).

      //It turns out that if there’s enough sincere voting, then even the “naively honest fools” who vote sincerely do better with Score Voting than with Approval Voting. See table two from this page on the “Shentrup-Smith Experiment”.//

      The chart notes that victimization occurs to honest voters if the number of strategic voters is more than about 45% in a 3-way race (and the required % of strategic voters is even more reduced when there are more candidates to choose from). This presents a problem for Score voting, because as you noted earlier:

      //in the 2000 USA presidential election, polls show that about 10% of the people who preferred the Green Party voted for their nominee, Ralph Nader.//

      That implies that only 10% of Green supporters voted sincerely, which is far below the required ~55-65% needed to overcome the effect of strategic voting as indicated by the simulations you cited.

  1. 2

    Clay Shentrup

    > But I think it has the practical problems I mentioned in the article, and would only be ideal if it were possible to forcibly extract honest votes from people.

    It’s very simple:

    – Score Voting is better for expressive voters (that is, voters who care more about expressing themselves—e.g. the voters who voted for Nader, even though they knew he had no chance)

    – Score Voting is better for tactical voters (who will vote all max and min scores), because the honest voters voluntarily cede voting power to them, making them statistically more satisfied.

    – Score Voting is better for the entire group as a whole, because the expressive voters donate more utility than they lose (that is, voting methods produce more net utility the more honesty there is).

    – Score Voting is better for the (presumably extremely tiny) percentage of voters who would be “accidentally” honest. That is, voters who WOULD be tactical if they were savvy enough, but simply don’t realize that honesty is not the best strategy. Why? Because the ideal Approval Voting threshold strategy is somewhat complex (www.electology.org/threshold), and they could easily screw it up. But an honest Score Voting ballot has something like 90% of the optimal power, statistically speaking (http://scorevoting.net/RVstrat3.html)—and is incredibly easy. So voters that unsophisticated would almost certainly be better off voting honestly than voting strategically.

    Reply
    1. 2.1

      Aaron

      //Score Voting is better for tactical voters (who will vote all max and min scores), because the honest voters voluntarily cede voting power to them, making them statistically more satisfied.//

      That argument only holds true if the majority of those sincere voters predominantly draw support away from the opposition. In other words, if most of the sincere voters are right-wingers rather than left-wingers for example.

      //- Score Voting is better for the entire group as a whole, because the expressive voters donate more utility than they lose (that is, voting methods produce more net utility the more honesty there is).//

      I still don’t understand what you mean by utility. If voters are accidentally aiding candidates they like least, then that is undermining the point of elections which is supposed to be representative of who the voters want in office.

  2. 3

    Clay Shentrup

    > I don’t understand this part of the article, because the chart to which you refer shows that A (Score Voting) is *less effective* at achieving desired results than D (Approval Voting).

    Good catch. I meant to say that column *C* is *greater* than column D. I.e. normalized sincere scores beat sincere Approval. I’ll correct that blog post.

    > The chart notes that victimization occurs to honest voters if the number of strategic voters is more than about 45% in a 3-way race

    But:

    A) It’s not really victimization, because expressive voters are *intentionally* voting sincerely. They *want* to express themselves, and in fact DO NOT want you to force them to use a voting system that restricts them to voting in a binary way “for their own good”. They get more pleasure out of speaking their minds than they get from casting the absolute strongest ballot possible. (And even the handful of “accidentally sincere” voters who may exist will plausibly be better of with a sincere normalized Score Voting ballot than with an Approval Voting ballot.)

    B) Even if they are victimized, it’s not a problem, because what matters is *net* welfare. Sincere voters donate more utility than they forfeit, leading to a net increase in utility. (And if you try to argue against this utilitarian model, you’re in trouble with logic => http://scorevoting.net/UtilFoundns.html)

    Reply
    1. 3.1

      Aaron

      //I meant to say that column *C* is *greater* than column D. I.e. normalized sincere scores beat sincere Approval. I’ll correct that blog post.//

      Agreed, although only for 3 candidates. And that refers to a Borda ranking system, not score voting.

      // It’s not really victimization, because expressive voters are *intentionally* voting sincerely. They *want* to express themselves//

      But I think the question is, do they want the race to have an outcome contrary to what they marked? That would seem to make no sense and be a contradiction.

      However, if expressing their precise preference is that important to enough people then I would support a Score voting system, since I think it will effectively function like Approval voting for the most part and at least be better than what we have now.

      //They get more pleasure out of speaking their minds than they get from casting the absolute strongest ballot possible.//

      That makes no sense to me personally, since it seems to defeat the point of voting. Voting is about the outcome. Each indiviudual vote is not made public so I don’t understand that. Also voting isn’t about pleasure, it’s about electing the voter’s desired candidates. But like I said above, I would be fine with a Score voting system anyway, since it is close enough to Approval and is far superior to Plurality.

      //(And even the handful of “accidentally sincere” voters who may exist will plausibly be better of with a sincere normalized Score Voting ballot than with an Approval Voting ballot.)//

      RIght, but that would be a different system. And is only likely to occur with no more than 3 candidates.

      // Even if they are victimized, it’s not a problem, because what matters is *net* welfare. Sincere voters donate more utility than they forfeit, leading to a net increase in utility.//

      I don’t understand your claim here. utility means the ability to achieve a desired end; of something to be useful or beneficial. But a victimized voter is inadvertently supporting the opposite of what they desire which would mean they are “donating” less utility. I think I just don’t understand how you are using “utility” in this context.

      To sum up, I am not opposed to a Score system. I just prefer an Approval one.

  3. 4

    Bob Wilson

    There was a potential ballot initiative in Oregon that would allow for a single non-partisan primary with with approval voting and the top two would move on to the general election. Unfortunately, it didn’t get enough signatures in time so it won’t be on the ballot. I think that would be the most pragmatic system for now.

    I’d prefer Range voting over plurality, but I’m a little conflicted about it in general. I just don’t know how voter’s are going to handle it. Could one voting group be more inclined towards strategic voting and can that provide a slant in favor or against that group? Also, are voters going to be likely to give candidates they don’t know a middle score? Could this cause a dark horse candidate to win?

    Reply
    1. 4.1

      Aaron

      Your questions are similar to my concerns as well. That’s why I prefer an Approval system. The structure of it would really reduce the problems I see in even Range/Score voting. Although as I said to another commentor, I would still gladly take range/score voting over plurality.

Leave a Comment (Log in with Facebook, Twitter, or WordPress)

2014 Powered By Wordpress, Goodnews Theme By Momizat Team