Condorcet is Cute

[Blake Cretney]

Donald E Davison wrote:

> Greetings,
> 
> Blake Cretney wrote: "If a majority of those expressing a preference rank A
> over B, then I say that based on the ballots, a majority prefers A to B."
> Don writes: This statement is true with all methods. Condorcet has no
> special claim to it. If this statement is to be your God, you have no right
> to claim that only through Condorcet can we reach God. The other methods
> have the same God.

I view this statement as a simple fact, or matter of definition.  I
don't see anything about the statement which implies an objective,
which is what I assume you mean by "God".

> 
> Blake: "A Condorcet winner is defined as the case where a single candidate
is
> preferred by a majority to every other candidate in the ballots provided."
> Don: The same can be said for Plurality. If we pair up the candidates of a
> Plurality election, the Condorcet winner will be the same as the Plurality

I don't understand you.  Are you really saying that plurality and
Condorcet always come to the same conclusion?  Is your point that you
don't like pair-wise majority decisions because the plurality winner
is the majority winner if there are only two candidates?

> winner - so what?? We are trying to get away from Plurality. Condorcet is a
> form of Plurality, that uses lower choices. Plurality will elect the
> candidate that has a majority. Condorcet will elect the candidate the has a
> majority of the pairings. Plurality will elect the lead candidate if there
> is no candidate with a majority. Concorcet will elect the lead candidate if
> there is no candidate with a majority of the pairings - two peas in a pod.

Am I missing something.  What do you mean by "lead candidate"?  If
you mean the plurality winner, then your statement about Condorcet is
false.  If you just mean the winner by the particular method, then it
would be true of any method.

> 
> Blake: "The point is that this is a circular tie.  It doesn't matter
whether
> you use a Condorcet completion method, IRV, one of your new methods, or
> don't bother to find a winner at all.  There is still a circular tie."
> Don: There is still a circular tie only if we are using Condorcet as a
> method. The other methods do not have circular ties. If Condorcet stumbles
> and falls into a tie, it is of no concern to the other methods. The other

If you decided to only choose the Condorcet winner, and set up no
plan for what to do when there was one, this would be a problem. 
However, Condorcet fans advocate Condorcet completion methods which
will come to a conclusion with the same kind of frequency as IRV.  So,
having a circular tie is no concern of theirs either.  I don't know if
you're aware of any Condorcet completion methods.

> methods can still do their thing as if Condorcet never existed. I fail to
> understand this adoration for Condorcet. Is this a cult thing??? Condorceet
> is cute, but cute is not a reason to pick a method - cute will not do.
> Condorcet is merely another method, and not a very good method at that.

They are all other methods.  Their quality is what is under debate.

> 
> Blake: "You can't really reduce the number of circular ties by reducing the
> value of lower choices..."
> Don: Oh, but I can and I did with your example.

I think what we have here is a difference of definition.  I've
already explained my definition of a circular tie.  You'll have to
tell me what your definition is.

You may be confused by the term "circular tie" into thinking that
circular preferences represent a tie for all Condorcet-type methods. 
This is not the case.

> 
> Blake: "If you mean that a voter prefers his higher ranked candidates to
> the lower ones, this is certainly true, but irrelevant."
> Don: It is not irrelevant. The voter wants the full weight of his vote to
> be on his most preferred choice as long as his first choice is a contender.

First, let me point out that the two statements are not the same, and
you gave no justification for moving from one to the other.

Consider this example:

40 A B C
13 B A C
12 B C A
35 C B A

IRV eliminates B,
A 53
C 47
A declared winner.

Now, my question is, did the C-1st voters actually want to keep their
second preference for B hidden until B was eliminated and it couldn't
do them any good?  I don't see how you can say that IRV elected their
least favourite candidate on their behalf.

> Condorcet allows the lower choices to work against a voter's first choice.
Or for them.
> It is long past time for pairwise guys to start having some respect for the
> wishes of the voters.
>      The wishes of the voters are not irrelevant.

It is irrelevant to whether a first over second preference is held
more strongly than a third over forth, for example.  That was the
question at hand.

> 
> Blake: "The people who come to the most bizarre conclusions usually seem to
> hold them most strongly."
> Don: Yes, I agree with you. There seems to be a lot of that going around.

I should point out that I made this statement as a justification for
not using strength of preference in voting methods.  It was clear from
context that this was not meant as a cheap shot at anybody on the
list.

> That is the way the pairwise guys are. This is not a perfect world. People
> don't always act the way we would like them to act. But, it is nothing that
> we cannot put up with. I have been putting up with them for the years that
> I have been on this list - life goes on.
> 
> Donald Davison
> 
> 

I find it unfortunate that you didn't answer any of the questions
posed by my previous post.  I think they were all fair, and many
weren't even hostile.  The only way we can hope to come to an
agreement is to honestly attempt to answer each others questions and
explain ourselves in as clear a way as possible.

---
Blake Cretney
See the EM Resource:  http://www.fortunecity.com/meltingpot/harrow/124