[EM] STAR loses in Eugene, 33%-67%

[Closed Limelike Curves]

>
> So I don't think it's necessarily better than IRV, and STAR still
> inherently suffers from the burden of tactical for voters whenever there
> are 3 or more candidates.  As with any Cardinal method s, including
> Approval.

And as with every ranked voting method, by Gibbard's theorem. The only
difference is every pure cardinal method (not including STAR) doesn't
suffer from the burden of dishonest (order-reversal) voting in the
3-candidate case, because they pass FBC and later-no-help.

Meanwhile, every defeat-dropping Condorcet method based on wv or margins
has a turkey-election as a potential strategic equilibrium.

On Wed, May 22, 2024 at 1:42 PM robert bristow-johnson <
rbj at audioimagination.com> wrote:

>
>
>
> > This is really sad.  Ironically, they were attacked by both the pro-RCV
> camp (who view it as competition) and the anti-RCV camp (who view it as a
> variation of RCV),
>
> I don't view it as a variation of Hare, except it *does* have two rounds,
> the latter with the same rules as IRV.
>
> It's that a close 3-way race can also cause STAR to fail to elect the
> Condorcet winner and then all of the bad IIA and Favorite Betrayal and
> Monotonity stuff follows.
>
> So I don't think it's necessarily better than IRV, and STAR still
> inherently suffers from the burden of tactical for voters whenever there
> are 3 or more candidates.  As with any Cardinal method s, including
> Approval.
>
> It should be ranked ballots and it should elect the Condorcet winner
> whenever such exists.
>
> >and the Eugene Weekly recommended voting No despite promoting STAR
> themselves several years prior. 😩
>
> <shrug>
> ----
> Election-Methods mailing list - see https://electorama.com/em for list
> info
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20240522/dc70ba74/attachment-0001.htm>