[Election-Methods] RE : Re: RE : Re: rcv ala tournament

[rob brown]

On Dec 29, 2007 6:30 PM, Kevin Venzke <stepjak at yahoo.fr> wrote:

> Rob,
>
> --- rob brown <rob at karmatics.com> a écrit:
> > > I can't think of any (serious) method with more straight-forward
> > strategy
> > > than Approval.
> >
> > Please elaborate.  Seems to me that the optimum strategy of Approval
> > involves guessing how others will vote.  And guessing how others will
> > vote
> > involves a) guessing what their best strategy is, which of course is
> > cyclical and becomes a great big hall of mirrors, and b) guessing how
> how
> > their actual voting differs from optimum strategy, which becomes an
> > exercise
> > in psychology.
>
> Well, at least once you have finished evaluating what you think may happen
> (with whatever probabilities) it becomes straight-forward, as you simply
> approve everyone who is better than your expectation (the average of
> outcomes' utilities weighted by their odds of occurring).
>

Sure, once you have evaluated what you think may happen.  But that is a very
big "if".

It is quite possible to have the tabulation system itself do that.....i.e.
ask you what your true preferences are, look at the votes of everyone else,
then cast the absolute most strategic approval ballot on your behalf.  (of
course, it is iterative and prone to cycles, but then, so is plain old
approval voting...but in approval the iteration and cycles just take place
in the much more messy world outside the tabulation method)

Anyway, as I'm sure you know, that system (rank/rate your candidates
honestly, let the system generate the most strategic approval ballot) is
simply DSV, and it works out to be the same as Condorcet.

It doesn't make a lot of sense to me to say that you can easily and
effectively game such a system, if the tabulation system itself already
guarantees you it will cast the most strategic approval vote for you based
on your honest preferences.

Something there doesn't add up.  Why would you have an incentive to tell
anything but the truth to something that is able to act in your best
interest based on the information you tell it?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20071230/14a63756/attachment-0003.htm>