[EM] FBC comparison; AWP

[Kevin Venzke]

Adam,

--- Adam Tarr <ahtarr at gmail.com> a écrit :
> > In DMC, neither A=B nor B>A is effective. The only ability you have to 
> > weaken
> > the A>B win (assuming you can't reverse it) is to disapprove A!
> 
> This seems to imply the possibility of favorite betrayal in a pretty common 
> case. Say your sincere ranking was A>B>cutoff>C. Furthermore, assume that 
> your approval of B is important to maintaining the strength of the B>C 
> defeat. If it was necessary to reduce the strength of the A>B win, wouldn't 
> the necessary strategic vote then become B>cutoff>A>C, i.e. favorite 
> betrayal?

I believe you are right.

I was thinking about AWP: I think it may be the best (according to FBC) of
any of these Condorcet methods (that is, not including MMPO).

Nothing is different in the case that you need to reverse a defeat.

But what can you do to weaken the A>B defeat? Just approve both A and B
if you hadn't, or disapprove A and B if you hadn't. The latter is probably
favorite betrayal, but I can't see that it would be more effective than the
former. This means you're more free to let A>B remain strictly ranked than
you are under WV.

> (My FBC simulation doesn't envision favorite betrayal of that kind, so I
> > would have to rethink it in order to find statistics for DMC relative to
> > WV and margins.)
> 
> I think that would be worth looking into. My guess (and I stress that it is 
> only a guess) is that DMC would have favorite betrayal incentive less often 
> than margins, but significantly more than wv.

That would be my guess, too. It is tricky to imagine how often the various
cases occur under each method, though.

Kevin Venzke



	

	
		
___________________________________________________________________________ 
Appel audio GRATUIT partout dans le monde avec le nouveau Yahoo! Messenger 
Téléchargez cette version sur http://fr.messenger.yahoo.com