[EM] wv & Cardinal Pairwise fail Participation and Consistency.

[MIKE OSSIPOFF]

You said:

Could you please show proof that WV (RP/Beatpath/River) passes
Consistency and CWP does not?

I reply:

Sorry, no I can't prove that because it isn't true. WV and Cardinal Pairwise 
both fail Participation and Consistency. In fact few if any rank methods 
other than Borda meet those criteria. In my posting I repeatedly spoke of 
that as a problem of the rank methods, except for Borda. But Borda has its 
own criterion-failure that Plurality doesn't have: Majority Favorite.

My criticism was of all rank methods.

Sure I like wv best, and I consider SFC & GSFC to be valuable. But do you 
think that we should be proposing methods that can do worse than Plurality 
in some ways? How will it look for us, and for single-winner reform, if they 
fail in a way that Plurality can't fail, and if it's a failure that really 
looks bad?

There are other criteria that Plurality & CR pass, but the rank methods 
fail, such as the simple votes-only version of Independence from Irrelevant 
Alternatives Criterion (IIAC):

Deleting a losing candidate from the ballots, and then recounting those 
ballots, should never change who wins.

[end of simple votes-only IIAC definition]

Though my favorite criteria aren't votes-only, I've never opposed all the 
votes-only criteria. I've often mentioned votes-only IIAC, and pointed out 
that Approval meets it. And that, if we accept that as IIAC, then Approval 
meets all of Arrow's results criteria, at least as I've seen them written.

And I've never disparaged Majority Favorite or Participation, which are 
votes-only criteria.

Can you promise that there won't be a big embarrassment when a rank method 
fails that IIAC, or Participation or Consistency, in a big way?

Mike Ossipoff

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