[EM] Question about Condorcet methods

[Dave Ketchum]

I doubt anyone has what could be described as a comprehensive list of methods.

Your words about 99th and 100th makes me wonder about your understanding. 
  You should pretty much start with an array, one line and one column for 
each candidate.  You must provide for expanding the array if write-ins 
need remembering as additional candidates.

Looking at each ballot you count for each pair if 99th>100th or 
100th>99th.  Do not count if neither 99th nor 100th is ranked
      For a method difference:  I would count, somehow, when voter ranks 
99th=100th; I get outvoted on this detail in EM debates.

DO NOT DO any switching such as you describe below.  Even if it is far 
down in a voter's ranking, it is what this voter said about this pair.  If 
this pair is far down in the list, there are many candidates this voter 
has ranked as better.
      More to the point, after a voter has ranked the 2 or 3 candidates 
really cared about, this voter probably has no reason to rank any more as 
better than last.

Having counted all the votes into the array, look for cycles:
      No cycles - normal - the winning candidate is liked better when 
compared with each other candidate.  Having elected a winner, we care 
little about losers, but could study about seconds, etc.
      Simple cycle such as A>B AND B>C AND C>A - possible, but expected to 
be rare - it says the collection of voters is not homogeneous.  Now we 
need rules as to where to break the cycle.  Note that it takes at least 3 
strongly backed candidates to get here.
           While the rules for breaking can inspire hot debate, the 
implementing of such should be trivial.
      Complex cycles - demonstrable in EM debates so better have rules, 
but should be REALLY rare in real life.

DWK

On Sat, 14 Oct 2006 14:28:32 -0700 (PDT) mrouse1 at mrouse.com wrote:
> I was just wondering if anyone had a link to a comprehensive list of
> Condorcet methods. I've been toying with one where you start flipping the
> lowest-ranked choices (under the theory that a person cares less about
> switching 99th and 100th place than they do switching 1st and 2nd), but I
> realize that someone has probably already figured it out (complete with
> paradoxical behavior) and I want to save myself some effort. :) I did some
> Google searches, but having to wade through a few hundred links takes more
> time than I'd like to spend.
> 
> Anyway, if anyone knows of a fairly complete list of Condorcet methods on
> the web, along with how they handle circular ties, I'd appreciate it.
> Thanks!
> 
> Michael Rouse
> mrouse1 at mrouse.com
-- 
  davek at clarityconnect.com    people.clarityconnect.com/webpages3/davek
  Dave Ketchum   108 Halstead Ave, Owego, NY  13827-1708   607-687-5026
            Do to no one what you would not want done to you.
                  If you want peace, work for justice.