OK. I seem to have misinterpreted a method based on the trailing end of a private mail thread. Can you please explain it again, using each letter to refer to only one thing? From your latest statement, I can't see why it isn't plurality (reluctance cutoff will always be the minimum, 0) and if it's not, I can't see why it isn't Abd's Bucklin/Range (same reluctance cutoff on all ballots). On the other hand, I think the method that I thought it was ("minimum reluctance for a majority win, using custom thresholds for each ballot") is interesting if someone can give a polynomial-time algorithm for finding the answer. Though of course, that method is just as vulnerable to strategic exaggeration as Range. It's something like using median instead of average rating for Range.<div>
<br></div><div>JQ<br><div><br><div class="gmail_quote">2010/5/24 <span dir="ltr"><<a href="mailto:fsimmons@pcc.edu">fsimmons@pcc.edu</a>></span><br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div>You are over looking the fact that X is the approval cutoff on all of the ballots, including the ones on which X is not approved.<div class="im"><br><br>----- Original Message -----<br>From: Jameson Quinn <br>Date: Monday, May 24, 2010 11:27 am<br>
Subject: Re: [EM] minimizing reluctance by DSV<br>To: <a href="mailto:fsimmons@pcc.edu" target="_blank">fsimmons@pcc.edu</a><br>Cc: <a href="mailto:election-methods@lists.electorama.com" target="_blank">election-methods@lists.electorama.com</a><br>
<br>> 2010/5/24 <br>> <br>> ><br>> > Hi Forest,<br>> ><br></div>> > --- En date de : Sam 22.5.10, fsimmons at <a href="http://pcc.edu" target="_blank">pcc.edu</a> > <a href="http://pcc.edu" target="_blank">pcc.edu</a>> a<div>
<div></div><div class="h5"><br>> > écrit :<br>> > > The alternative X is used as the approval cutoff. On<br>> > > some of the ballots the<br>> > > cutoff is considered excluded (not including X as approved)<br>
> > > but on just enough<br>> > > ballots to make X the approval winner, the cutoff is<br>> > > considered inclusive (so on<br>> > > these ballots X is approved).<br>> > ><br>
> > > The "reluctance" of ballot B in approving X is the<br>
> > > difference between the<br>> > > maxrange value and the rating given to alternative X by<br>> > > ballot B.<br>> > ><br>> > > Elect the alternative X with the least possible reluctance<br>
> > > total.<br>> ><br>> > I don't understand how you determine which ballots approve X. <br>> Is it<br>> > random, or irrelevant? Or do you have to find the selection <br>> which will<br>
> > minimize reluctance to electing X?<br>> ><br>> > Kevin<br>> ><br>> > Forest replies:<br>> ><br>> > Yes, the selection that minimizes reluctance. That’s why I <br>> put the word<br>
> > “possible” in the<br>> > phrase, “minimum possible reluctance.”<br>> ><br>> > Start by approving X on all of the ballots with zero <br>> reluctance for X, then<br>> > move on to the ballots that have<br>
> > a reluctance of one, etc. until there is enough approval for X <br>> to overcome<br>> > the approval for the candidates<br>> > that are preferred over X .<br>> ><br>> ><br>> Um, wouldn't that just be plurality? That is, you'd start with zero<br>
> reluctance, and the plurality leader would win. That would <br></div></div>> automaticallybeat any win with nonzero reluctance.<div class="im"><br>> <br>> I think you mean, then, to "Elect the alternative X with the <br>
> least possible<br>> reluctance total for a majority win". That system is some kind <br>> of hybrid<br>> between Abd's Range/Bucklin proposal and a Condorcet method. I <br>> like it - but<br>> it is basically a non-starter until you have an explicit <br>
> algorithm for<br>> finding the correct winner.<br>> <br>> JQ<br>> </div></div>
</blockquote></div><br></div></div>