[EM] WV, MDDA, DMC--Order-reversal & truncation

MIKE OSSIPOFF nkklrp at hotmail.com
Thu Oct 20 15:17:33 PDT 2005


I was correct when I said that MDDA doesn't have an offensive order-reversal 
problem.

That's because MDDA meets the important criteria that wv meets, and behaves 
like wv, and, as I've been arguing for a long time, wv Condorcet doesn't 
have an offensive order-reversal problem.

But I was incorrect when I implied or said that MDDA is significantly better 
than wv when it comes to offensive order-reversal, or that unopposed 
offensive order-reversal can't succeed in MDDA. It can succeed in wv, MDDA, 
and DMC.

With 3 candidates, wv, MDDA, and DMC are very similar. With all 3 methods, 
offensive order-reversal against the middle candidate, by a candidate 
against whom Middle has a majority pairwise victory, when not opposed, can 
succeed, depending on requrements about the relative sizes of the factions. 
Those requirements are, in general, different for the different methods, but 
that's only a detail, and not a significant difference.

In all 3 methods (again with 3 candidates) defensive truncation always 
thwarts and punishes offensive order-reversal by supporters of a candidate 
who is majority-beaten by the intended victim.

With wv and MDDA, defensive equal-ranking also thwarts offensive 
order-reversal, but that isn't true with DMC.

Because of compliance with SFC and SDSC, MDDA of course matches very well 
what wv offers when there are more than 3 candidates. Though wv has more 
majority-enforcing criterion compliances, MDDA has the most important and 
effective of those: SFC.

I can't guarantee that DMC has the benefits that I describe here when there 
are more than 3 candidates.

Mike Ossipoff

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today - it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/




More information about the Election-Methods mailing list