"votes against" in pair-defeats? in all pairings?

Steve Eppley seppley at alumni.caltech.edu
Wed Jun 26 12:04:11 PDT 1996


There's a difference in the Condorcet definitions proposed by Mike O
and Bruce A which I'd like to explore.  

Mike's definition would limit the search for each candidate's
largest "votes against" to those pairings in which the candidate was
defeated.  Bruce's definition, if I understand it, looks at *all* the
candidate's pairings--even those pairings the candidate won.

Is the difference significant?  In what circumstances will the
outcome be different?  What's the philosophical basis for limiting
the search to the candidates' pair-defeats? 

I presume Mike has a good reason to use only the pair-defeats.  
Too bad, since the wording of the Condorcet definition could be
considerably simplified if we could remove the sentence about "the
candidate who beats all the others pairwise, if there is one" as
well as the phrase about pair-defeats. 

---Steve     (Steve Eppley    seppley at alumni.caltech.edu)





More information about the Election-Methods mailing list