At 09:08 AM 7/26/2007, Steve Eppley wrote: >It proposes several better methods, including a >simple but probably very effective patch for IRV: letting candidates >withdraw after the votes are cast.
Nice. This could eliminate the center squeeze effect, making IRV much better. Of course, this is kind of a variation on Asset Voting. There are some technical problems, of course. To really make sense, the candidates would have to know the effect of withdrawal, they would have to know that they were going to be eliminated in the first round, or in some later critical round. Commonly, with IRV, the entire ballot isn't even counted.... why count a lot of moot votes, at public expense? However, public ballot imaging, which I think is a fantastic idea, allowing the public to count the vote, would also allow candidates to determine the vote in intimate detail. As long as they have enough time before the truly official count takes place. For a first reform, Count All the Votes (i.e, don't discard overvotes) makes a whole lot more sense, it's a no-brainer better than Plurality. People will argue about whether IRV or Approval are better, we know, but IRV has a lot of transitional cost, some possible harmful effects, such as Center Squeeze that your fix might also address in a way, and so it makes sense to me to go first for cheap and *arguably* as effective or more effective, and this not only does not prejudice further reform, I predict it will power it. Once voters have the ability to vote for more than one, they will want the ability to indicate preference. And then we go toward Range or Ranked methods. The next Range method, Range 2 (CR-3) is quite interesting. And cheap, one more vote position only. Easy to count, easy to understand. ---- Election-Methods mailing list - see http://electorama.com/em for list info
