[Election-Methods] rcv ala tournament

[CLAY SHENTRUP]

On Dec 30, 2007 2:05 PM, Dave Ketchum <davek at clarityconnect.com> wrote:
> My point is that, whatever Approval can do, the other methods are also
> capable of doing.

what kind of totally meaningless statement is that?  any voting method
is "capable" of electing anyone.  but it will react to the way the
voters actually vote.

> > if you and i both prefer X>Y>Z, but i like Y more
> > than the average of X and Z, and you like Y less than the average of X
> > and Z, then i'll vote for both X and Y, and you'll just vote for X.
> > it's what is called "revealed preference" in ecenomics - we are forced
> > to say something about the relative intensity of our support for Y,
> > even though we have the same ordered preferences.
> That might reveal a bit, though even that not clearly.

well of course it's clear.  one additional vote for Y is as clear as day.

> BUT, assuming I want to indicate preference of X and Y over Z, Approval
> can show my less liking for Z, but cannot, in the same vote, indicate my
> relative preference as to X vs Y.

approval allows you to make fewer statements about your _order_ of
preference, but causes you to say more about your _intensity_ of
preference.  so if you look at its social utility efficiency - it's
_real_ expressiveness, with strategy, revealed preference, etc. taken
into account - it is better than most of the other commonly discussed
methods.

> I am getting dizzy.  With 4 candidates, how do you come up with more than
> 6 pairs to even consider comparing?

ah, you're correct.  i should have said it is "n choose 2", which is
actually n!/2*(n-2)! = n(n-1)/2

> "RCV (with 3 candidates allowed)"?  Is this from some definition I have
> not seen, or something that seemed to make sense in the above statement?

ranked-choice voting is just a name for irv as it has been implemented
in places like san francisco, where you rank just 3 candidates.  it's
a stupid name.

CLAY