[EM] Is Condorcet The Turkey?

[Dave Ketchum]

On Tue, 10 Jun 2003 18:50:54 EDT Dgamble997 at aol.com wrote in part:
 > Kevin Venzke Wrote
 >
 > Also stated
 >
 > Now I will offer reasons why Condorcet is better than IRV.
 > 1. You can generally vote sincerely. In IRV, you have to keep in mind
 > that you cannot support a candidate after he's been eliminated
 >
 > Consider the following example:
 >
 > 40 voters sincerely support candidate A and sincerely dislike candidates
 > B and C. However they dislike C considerably more than B.
 >
 > 39 Voters sincerely support C and sincerely dislike candidates A and B.
 > However they dislike A considerably more than B.
 >
 > The 21 B supporters vote 11B>A>C,  10 B>C>A.
 >
 > If A and C supporters vote sincerely the votes look like this:
 >
 > 40 A     11 B>A>C   10 B>C>A      39 C
 >
 > Under Condorcet A wins.

As does A under IRV.  Condorcet and IRV EACH see that B has little
support, leaving a 2-way race which even Plurality could finish (back at a
3-way race, Plurality could see that B is a loser, but would have no idea
what the B voters think of A and C).
 >
 > If C's Supporters insincerely express a preference for B they can at
 > least defeat the candidate they dislike most (A).

Disagreed.  They can (and do - see above) sincerely prefer B over A, and
therefore express that preference.  By ranking C first they did all they
could to elect C; here they are doing all they can to make sure A does not
win even if C loses.

For the same reason, A voters would rank B as backup to protect themselves
from C.

Back to IRV - if there were any believable chance that their candidate
might come in third on first rank votes, A and C had best rank B second to
protect themselves from their worst enemy.
 >
 > Motives and situations that encourage insincere voting can be found
 > under virtually any electoral system Condorcet being no exception.
 >
Perhaps we do not share a definition for sincerity, which I translate as
following true desire.  Agreed that there are methods in which you vote
for an unwanted candidate to improve odds of your most wanted winning.
This can backfire if you guess wrong as to how others will vote but - how
is it even worth trying with Condorcet?
 >
 >
 > Dave Ketchum wrote
 >
 > For the American system either IRV or Condorcet will be more tolerant of
 > third party votes than Plurality is, so we can expect more of them.
 >
 > Yes I agree. Skimming through the 2002 congressional results I noticed
 > two significant organised groupings other than Democrat/Republican -
 > Libertarian and Green. Since ( correct me if I'm wrong) the Libertarians
 > are generally perceived as being to the right of the Republicans and the
 > Greens to the left of the Democrats, Republicans are unlikely to
 > preference Greens to defeat Democrats nor Democrats likely to preference
 > Libertarians to defeat Republicans.

I am inclined to rank Libertarians in the middle - wanting government to
do less of what the Republicans promote, AND less of what the Democrats
promote.  Let's, for the moment, substitute the imaginary Rightists who
fit your description of Libertarians in this post.
 >
 > Both Condorcet and IRV tend to assist centrist candidates and  work
 > against those perceived as extreme. IRV electing centrist candidates who
 > run ahead of either Democrats or Republicans and Condorcet electing
 > additionally centrists who run behind both Democrats and Republicans.
 >
 > Neither IRV or Condorcet will get Greens or Libertarians into congress
 > they will just prevent a Gore/Bush/Nader situation (because a Green
 > stood a Republican defeats a Democrat) and possibly elect a few  or
 > maybe even a lot of centrist independents.

You seem to assume more purity of candidates obeying party platforms than
exists here.  Anyone can be a member of whatever party they choose, get to
primary election via petition (or whatever the law says in their state),
and win the primary if they convince enough voters.  Not always essential
to be a member of a party to do the above and become the party's candidate.

Look at Ron Paul, a Libertarian in a Republican cloak.  Shouldn't fool
anyone.  When he gets to Washington the Republicans have to honor the
cloak.  Back in Texas they TRY to get him defeated at election time (seems
Republican voters in Texas like this person and do not object to the cloak).
 >
 > As regards your comment on the Kent 1997 situation. Yes a Liberal
 > Democrat victory in the situation
 >
 > C 42%
 > LD 20%
 > LA 38%
 >
 > would probably please most voters most at the level of the individual
 > constituency contest. However the fact that parties obtaining 77.7% of
 > first preference vote only obtain 23.5% of the seats and that a party
 > with 17% support obtains a large majority would most probably cause
 > great dissatisfaction on an across Kent level. The dissatisfaction at
 > the result at the across Kent level  would probably outweigh
 > considerably the satisfaction with the individual constituency results.
 >
Seems like some education should be worthwhile.  If the Cs and LAs would
REALLY rather elect more of their supposed worst enemies, all they would
have to do is rank them ahead of the LDs when they vote.

 > One question comes to mind what might the outcome of Bush, Clinton,
 > Perot (1992) have been under Condorcet?

Note that our Constitution directs how the Electoral College shall
operate, once elected/appointed.  However, each state legislature could
specify Condorcet as the method for electing their state's EC members.

Since the major race was Bush vs Clinton, MANY voters had to take part in
that race to protect themselves against the worst of the two.  Condorcet
would let MANY voters place Perot first or second and place their worst
enemy third.  I see this producing some Perot backers in the EC - hard to
predict how many.

Actually, I do NOT like Condorcet for this election.  Each state is
electing a group of EC members - let us do Proportional Representation
such that if 50% of a state's voters like Clinton, let Clinton have 50% of
the state's electors.
 >
 > David Gamble
-- 
   davek at clarityconnect.com    http://www.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.