2011/10/31 David L Wetzell <[email protected]> > > > On Mon, Oct 31, 2011 at 4:05 PM, Jameson Quinn <[email protected]>wrote: > >> >> strategic incentives, >>>> >>> >>> 2. Whether one ranks a second or third choice could only back fire if >>> there was a close race between your first choice and one of your other >>> choices for being the third place candidate in the first stage. This isn't >>> likely and even then, it's safe to say that the third/fourth place >>> candidate in the first stage has a less than 1/3rd chance to be the winner >>> in the second stage. >>> >> >> You are arguing that the costs of adding preferences are probably nil. >> But essentially the same argument applies to the benefits. >> > > dlw: I'm arguing that the odds of ranking two or three candidates having > unintended consequences against one's first preference are probably nil. > The benefit of increasing the chances that one of one's top two or three > preferences are among the three finalists and the winner is similarly > small, but no where near nil. The odds are always long in most elections, > but the benefit-side is at least 3 times as great as the cost-side. > >> >> >> >>> and LNH. >>>> >>> 3. It's no longer 100%, but it's up there. Similarly, there'd still be >>> a chance of non-monotonicity, but it'd be relatively low and thereby >>> unlikely enough not to affect voters strategies dysfunctionally. >>> >> >> LNH matters most in a Burlington-like scenario, where it arguably kept >> the Republican from winning --- the plurality winner and Condorcet loser >> among the three frontrunners there. That is, by reassuring the Democrats >> that it couldn't hurt them to second-rank the Progressive, it prevented a >> chicken dilemma. This is exactly the situation when the almost-LNH of your >> system breaks down. >> > > dlw: I beg to differ. My approach uses the first stage to reduce the > number of candidates to 3. In Burlington, those three would have been the > Dems, Progs and Pubs so the LNH would still be in place in the second > stage. >
Hmmm.... I could certainly counter that the Dems could theoretically third-rank a Dem clone or a turkey candidate in order to push the Prog out of the top three. The turkey is pretty implausible, but I could imagine it becoming the norm to run two clones, as in early presidential elections when VP was not a separate election. But anyway, you're right, the problem is not as bad as I'd thought. So I guess I'll accept your proposal in the category of systems like IRV - systems I support as better than plurality but don't actively promote because there are better options. > >> Admirable response. I'd have to agree... but of course, I'd guess that >> >2/3 of that will be *me* convincing *you*. :) >> > I revise that estimate to >1/2 :) Jameson
---- Election-Methods mailing list - see http://electorama.com/em for list info
