[EM] Schulze's method fails Condorcet's Criterion, right?

[MIKE OSSIPOFF]

Dear Markus--

You said:

>in your example, candidate B is the Condorcet winner of the
>reported opinions. The Schulze method chooses candidate B.
>Therefore your example doesn't demonstrate a violation of
>the Condorcet criterion.

In my example, 60% of the voters, in when they vote their ratings,
vote C over both of the other candidates. That means that C is the
candidate who beats each one of the others, and the candidate who
must win in order for Condorcet's Criterion or Beatpath GMC to
be complied with.


Take another look at the example. 60 voters voted the following
ratings:

A: 0  B: .7  C: 1

The point of that example is that if you use 0-info strategy to
"get from the voted ratings the information that the voting system
needs, and which it would ordinarily get from its own particular ballots, 
then Schulze's method fails Condorcet's Criterion and
Beatpath GMC, when your system of applying criteria is used.

Of course you've never told us how you & your scientists "get from
the voted ballots the information that the voting system needs",
though I've asked. More specifically I've asked many times how you'd
deal with Approval and single-winner Cumulative in that system.
More generally, I ask what is the _uniform_ way that you &  your
scientists get from the voted ratings the information that the voting
system needs, the information that, in an actual election, it would
get from its own method-specific ballots.

Until you come up with an answer to that, Beatpath GMC remains
undefined. So does your version of Condorcet's Criterion, etc.

Mike Ossipoff

_____________________________________________________________________________________
Get more from the Web.  FREE MSN Explorer download : http://explorer.msn.com