Mixed Condorcet-Plurality

[Tom Ruen]

Thanks for your comments, Buddha. :)

Your case is certainly a good one to show IRV's failings.

AC=49, BC=48, CB=3 is an equally interesting election to challenge the
Condorcet winner - do we really want a majority support so badly that we are
willing to elect a candidate with such a small first rank support? I
reluctantly accept Condorcet in this case, and to me it shows voters need to
be very careful in lower rankings in Condorcet.

Many voters might vote a centrist second in a purely reactionary way without
seriously considering if this candidate is superior to the candidate on the
other side of the political spectrum. I've argued about this issue with
friends and see that many people don't think very hard between choices they
overall don't like. I'd recommend people stop ranking in Condorcet unless
they have a clear preference among disliked candidates. This helps prevents
a weak centrist from winning merely on less thoughtful rankings.

Myself, I'm not sure about primaries, not clearly against them. Primaries
are somewhat strange because a party will endorse a candidate before the
primary and then anyone can run in the primary, but in practice only
candidates with lots of money and/or general name-recognition/fame can
afford to go against the endorsed candidate.

Perhaps with methods like Condorcet we can widen the field and reduce the
need for parties, although money remains a factor. Parties have power
because they attract followers who are willing to give time and money. Who
do I give money to if there are 3 candidates I like running? I'd rather
there be an early process to select between them and then focus everyone's
support on that one candidate.

Getting a party to back a single candidate is a way to generate public
interest. Candidates have a reason to defend certain issues to attract
followers and this is what makes elections valuable - because the top two
tend to represent opposing sides of the center and we can see which side has
the most overall support. It is really a tough issue for parties too - do we
endorse a candidate that is strong on our issues, or one that will more
likely appeal to the independents also? A strong party will want to do the
first, while a weaker one may lean to the center to get a better chance of
being elected.

About the plurality process for cycles, I put the question out in hopes
someone else has already made an evaluation of it. I'm interested to know if
it can be monotonic for instance, although I expect it isn't. I'm not
entirely worried about it's failings since we're talking about very strange
elections (with cycles) anyway.

I still like the idea of looking at defending Condorcet in two stages. First
we need to defend it over runoffs when a Condorcet winner exists. Second we
can ask what method we want to use to handle cycles when discovered. The
answer to the second question much less clear to me. I would also like to
see real election examples when cycles exist. I really wonder how often
they'd occur in real elections.

Tom

----- Original Message -----
From: "Buddha Buck" <bmbuck at 14850.com>
To: <election-methods-list at eskimo.com>
Sent: Tuesday, April 10, 2001 11:28 PM
Subject: Re: Mixed Condorcet-Plurality


> "Tom Ruen" <tomruen at itascacg.com> writes:
>
> > This is an example that Condorcet handles reasonably because there are
no
> > cycles. B and C have a united coalition - one of them is guaranteed to
win
> > by any good method.
> >
> > The question here in judging between using Runoffs or Condorcet is
whether
> > you want A supporter's second rank preference to decide a winner between
B
> > and C. Since B and C have a mutually united coalition we might imagine
they
> > were in the same party and a primary could have chosen between them
which
> > should run against A. Well, I mean this shows why IRV can replace
primaries
> > while Condorcet can not.
>
> Actually, this is a simplification of a standard example I use to show
> one of the problems with IRV.  It is not clear that B and C have a
> mutually united coalition.  Try this varient instead:
>
> 41 ABC
> 39 CBA
> 12 BCA
> 08 BAC
>
> Again, you have Plurality selects A, IRV selects C, Condorcet selects
> B.
>
> Personally, I don't care about primaries.  Primaries are tools
> political parties use to directly decide who is going to be on the
> ballot.  I have a philosophical objection to parties having a direct
> say in who is going to be on a ballot.
>
> > My original email said use Condorcet if there is a candidate that beats
all
> > others in pairwise elections and otherwise use plurality among the top
set
> > of mutually pair-defeatable candidates.
>
> That may be a reasonable method, but I'd like to see it compared to
> other Condorcet-based methods.
>