[Election-Methods] Simple two candidate election

[rob brown]

On Dec 20, 2007 8:32 PM, Paul Kislanko <jpkislanko at bellsouth.net> wrote:
> Well, what point were you getting at other than if there's only one issue
> you can use median to decide the one issue?
>
> My point is that the example isn't useful. In real elections there are more
> alternatives than "set the value for this one item" examples can solve.

Yes.  I see simplified examples on this list used to explain
theoretical concepts all the time.  I thought that was ok, and often
necessary.  I was trying to help resolve a theoretical point.

(however I would argue that an election in a club where they decide
what their dues are is every bit as "useful" and "real" an election as
one where we vote for president.  Does it have to be government to be
"real"?  For that matter,  I see no reason why a government couldn't
allow the citizens to vote for its budget.  But all of this is
irrelevant to the theoretical concept I was trying to explore.)

> Yes, I got your point, but you didn't answer my question.

Well, ok, but I was trying to get an answer on what I saw as a fairly
simple straightforward question, rather than go the "free association"
route off onto another subject.

I'll back up and explain.

What  I was trying to do is establish a baseline for evaluating
election methods (including ones that would be used for "real"
elections with multiple discrete candidates).  So I wanted to start by
discussing a type of election that does not have potential for cycles
and all those complexities that Arrow stressed out about.

I suggested that a "theoretically perfect" election can be held in the
case where we are looking for a simple numerical value (of course, I
am making the assumption that the closer the result is to a voters
ideal, the happier the voter will be).  My understanding is that Arrow
would have no problems with such an election.

If, however, this is not agreed on (and I have reason to think it
isn't, by the Range voting people specifically), it seems pointless to
talk about condorcet vs. range vs. approval.  Since we can't even
agree on what makes makes a theoretically "ideal"  election in simpler
cases where we have no possibility of cycles and all that ugly stuff.

That's why I was asking the question.  To help establish a baseline
for evaluating election methods.  Not to solve a real world problem
with numerical voting.

> You may think it
> is useful to vote a numerical value, but construct an election where the
> club members decide how to spend their collective dues along your lines.
> With more than one alternative, how do you reduce the election to a single
> rating?

Yeah, I don't know.  Not trying to be a jerk about it, but I'm not
going to answer it, because it has nothing to do with what I was
getting at.  Maybe someone else will.

-rob