[EM] Condorcet, Same Ranks, Ties

[MIKE OSSIPOFF]

>I am waiting to see a system with similar properties to
>Condorcet/Schultz or Ranked Pairs, but which allows voters to rank
>multiple candidates at the same preference.

Most advocates of those methods, and other Condorcet versions, are in favor 
of allowing voters to rank more than 1 alternative or candidate at the same 
rank position.

If I rank Gore equal to McCain, that merely means that I'm not casting
a pairwise vote between those 2. I'm voting both of them over everyone
I rank lower, and both of them below everyone I rank higher.

The IRV promoters have refused to make equal ranking an inseparable
part of their proposal. They don't include it when they make an
IRV "enabling proposal".

>Condorcet criterion when there are no ties.
>
>I am also waiting, (perhaps paradoxically) to see a system that
>minimizes arbitrary/random tie breaking, as I feel arbitrary/random
>is very hard to keep from being corrupt, and thus very hard to argue
>for, especially in the US. Even using plurality to tie break, or a
>Borda Count, would likely be more acceptable to arbitrary/random.

With public elections, ties are so unlikely that I always suggest that
if the existing electoral law says to draw lots in the event of a
tie ("Random Candidate"), then we might as well leave that in.
But Plurality would be a natural thing to suggest too, for solving ties
in public elections. Again, ties in public elections are so rare that
any solution of them is perfectly ok.

In small committee elections, there's certainly a case for trying
other methods like Borda, Plurality, etc., before resortinig to
randomness. Decisiveness is gained. I used to advocate the use of
other methods as tiebreakers for that reason.

But, when I was claiming that SSD is better than Schulze (BeatpathWinner)
and Tideman because of situations in which it's more decisive, people
pointed out to me that that decisiveness comes at the expense of
a stratgegy problem. In that method's case, in small elections, it
loses the guarantee that a faction doesn't advantage or disadvantage
itself by running a big set of clone alternatives, when voting is
sincere.

It was argued to me that that "clone independence" is more important
than decisiveness. Better to sometimes choose randomly than cause
factions to have to strategize about running clones, or avoiding
running alternatives that are very similar.

That's why I now agree in advocating Random Ballot instead of
other count rules for solving ties.

For an e-mail poll, like our demonstration poll, Random Ballot could
be difficult, since we can't all watch someone draw a number from a
paper bag. There's a good randomizing method based on the length of
the (say) 10th non-title word in the 1st article on a certain page
of the New York Times or London Times on a particular day. If we're
perfectionists, then, we can still use Random Ballot. Or people might
prefer to just use other count rules, to avoid having to buy a
New York Times.


>
>Finally, how does one resolve the problem of ballot access.
>If we had 1000 candidates for a national president, would people
>really want to go through a ballot ranking all 1000 candidates?

There should still be some kind of requirements for getting on the
ballot, like being chosen by a ballot-qualified party, or getting a
certain large number of signatures.

I don't know if write-in candidates now need signatures to register.
Maybe they should need at least some signatures.

But voters shouldn't have to vote for all the candidates. One good
thing about Condorcet/Schulze is that truncation causes no problems,
as it does with other pairwise count methods, such as Tideman(margins).






>Or would we need to spend money on voting machines that force people,
>for each choice, to choose from a list of 1000 candidates, then
>output the ballot? I suspect we would get many abbreviated ballots.
>
>Our organization has currently decided to use Approval Voting for
>single winner, because it is better than plurality and cardinal
>rankings, extremely easy to implement, easy to show majority support,
>and decreases the likelihood of bad ballots (i.e. overvotes, same
>rankings in single rank systems, mistaken write-in names). However, I
>tend to prefer something that allows rankings and at least picks the
>Condorcet Winner when there are no ties.
>
>Right now IRV is increasingly supported as the only alternative to
>plurality. IRV advocates are very loud and very unified, giving them
>a cascading unfair advantage in the more progressive press
>organizations, especially through their mouth pieces in Center for
>Voting and Democracy, and in the Green Party. (I.e. Nader talks about
>IRV, but no other possibility). If those who believe that IRV is one
>of the worst alternatives, but also believe Plurality is worse than
>IRV, wish to make any headway, we need to unify behind some
>alternative that is easily implemented and understood. It seems to me
>that Ranked Pairs and Plain Condorcet/Schultz would be at the top of
>the lists for a unifying method. Ranked Pairs does seem, to me,
>easier to explain than PC/S. Also all single rank methods (IRV, RP,
>PC/S) suffer from the problem of making bad ballots more likely, and
>RP and PC/S suffer from the problem of an overuse of randomness to
>resolve ties.
>
>If I am to reverse the advocation of IRV in the Green Party without
>appearing to support plurality, I need to advocate an alterative.
>
>Perhaps I should advocating a referendum to Green Party members
>regarding the alternatives, with the results published and ranked
>using Voter's Choice, simply for our own education on these matters.
>Unfortunately, I find that discussions of these systems appear to
>technical for most of our members.
>
>I would find any responses most useful,
>- Moe, Tampa Green Party.
>

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com