[EM] About random election methods, CDTT,RB

[Kevin Venzke]

Hello,

As I wrote recently, I want to elect from the CDTT set by Random
Ballot. (The CDTT set, again, contains every candidate who has a
majority-strength beatpath to everyone who has such a beatpath back
to them.)

I think the randomness can be justified here by noting that the CDTT
effectively doesn't regard pairwise contests that don't have a 
majority behind either side, so it can be explained that the voters
didn't return a clear decision.

This explanation is easier to support if the voters don't have much
strategic incentive to truncate. That wouldn't be so much the case
if the Smith set were used, since adding a preference for A could
reverse a simple loss that turns A into the CW, even if the voter
prefers the previous winner to A.

With the CDTT, if A has a loss then it can't be reversed by adding
preferences for A, since it's necessarily a majority-strength loss.
So usually the voter doesn't have to worry that adding an A preference
could turn A into the unique winner. It's much more likely to assist
higher-ranked candidates, by increasing the likelihood that a lower
preference will be ousted from the CDTT.

You could make the Smith set and the CDTT equivalent if you disallow 
equal ranking and truncation.

Kevin Venzke



	

	
		
Découvrez nos promotions exclusives "destination de la Tunisie, du Maroc, des Baléares et la Rép. Dominicaine sur Yahoo! Voyages :
http://fr.travel.yahoo.com/promotions/mar14.html