Approval Voting fish (3)

[MIKE OSSIPOFF]

Craig Carey says:


>(This message is believed by me to be not worth reading by most.)

So don't say you weren't told :-)


>
>
>At 16:30 03.03.00 , MIKE OSSIPOFF wrote:
> >
>I have deleted paragraphs containing the word "compromise"
>  (an undefined noun).
>Typos apologised for.
>...

I didn't write that.


>
>
> >If the allowed weights must be positive, and you limit how
> >many points you may give, then congratulations, now you have
> >a method that's strategically equivalent to FPTP. A distinct
> >step down from Approval.
> >
>
>The mistake is that you seemed to miss that that can be regarded
>  as an optimisation problem. [One that can't be defined.]

I'm talking about voting strategy as an optimization problem.

>
>
> >Allowing negative votes, and specifying a limit to the sum of
>...
> >But constraining the sum of the votes, or their absolute value,
> >is an unnecessary constraint on voter freedom. With flexible
> >points, limited only by a maximum & minimum that one can give
> >to a particular candidate, the strategically informed voter would
> >want to give the maximum to some and the minimum to the rest. You
> >wouldn't allow that, would require voters to make difficult choices
> >that they don't have to make in Approval.
>
>That is what the method should do: approximately allocate a minimum
>  to candidates just to cahnge them from losers to winners.

Here's the strategy for your "Limited Total Magnitude" method:

(I use "magnitude" to mean "absolute value").

Give your entire points magnitude supply to the candidate whose
strategic value (defined in my previous reply to you) has the
greatest magnitude. If his strategic value is positive, give
him positive votes. If it's negative, give him negative votes.

So, depending on the circumstances, it will be as if you're
voting in FPTP or in negative FPTP. We know how undesirabale
FPTP is. People who've studied negative FPTP say that it's
considerably worse than FPTP.




>People do not want to maximize the utility expectation.

You don't. Fine.


>
>Expectation is a word from statistics (yes?).

Statistics or probability I guess.

>
>Are there some probabilistic ideas that are a basis for your defence of,
>  or derivation of, the Approval Voting method?. Use of probability would
>  lead to a sub-optimal rule failing method.

No, I haven't used probability to defend Approval, only to
calculate strategies.


>
>What is a "utility expectation" ?. This is a simple question able to
>  be answered with a precise definition (just like my request for SARC
>  to be defined).

SARC was defined in my first letter to you. Sorry I didn't
define expectation. The utility expectation for one outcome
is the probability of that outcome multiplied by the utility
of that outcome. If you sum that over all the possible outcomes,
that's the utility expectation of the event for you. In this
case the event is the election, and an outcome is a candidate's
election.


That's all I can write today.

Mike Ossipoff


______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com