[Election-Methods] Is the Condorcet winner always best?

[fsimmons at pcc.edu]

My last message got mangled by the server, so here goes again!The scenario46 A>>B>C5  B>>A>C5  B>>C>A44 C>>B>Asuggests
that ninety percent of the voters would rationally prefer a Random
Ballot drawing over the sure election of Condorcet Candidate B.As
for Approval; since the B voters are so evenly divided in their slight
preferences relative to A and C, the A and C supporters might
reasonably consider that each has a roughly fifty-fifty chance of
winning if they leave the decision up to B supporters. So there would
be little if any perceived advantage for approving B on very many
ballots.However, these probability perceptions can be easily
thrown off by deceptive polls and other disinformation disseminated by
big money campaigns.  This is a problem for Approval, but not for
Random Ballot. In the category of deterministic methods, it is less of
a problem for DYN (Delegable Yes/No) than for Approval, since DYN
voters can afford to take reported probabilities with a grain of salt.Forest 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20071213/586377be/attachment-0002.htm>