A few comments on strategy matters
How can a method be called “monotonic” if adding a ballot that votes X over
y can change the winner from X to y ???
Another reason to prefer Approval & Score.
What’s the difference between changing a ballot & adding one? Well, one is
legal & one isn’t, & happens whenever another person enters the
polling-place.
——-
I define “Foolproof” as “ meeting FBC”.
For voting purposes, I define a “sucker” as someone who’d vote a
“lesser-evil”, someone he despises, over his favorite.
With a foolproof method, the worst that a sucker can do is cancel himself
out, with respect to his despised & not-despised candidates. Then he won’t
come into full-flower as the sucker that he is.
———
A method that isn’t foolproof is “sucker-vulnerable”.
———-
Condorcet & Hare are sucker-vulnerable.
But there’s still hope that they’ll work, because they’d be enacted by
Progressives, who aren’t doing so because they want to bury their favorite
under someone that they despise…but rather because they don’t want to.
——
James Green-Armatage described some Smith-&-Hare hybrids, & said that they
seem particularly strategy-resistant.
I haven’t examined them for the strategic-properties that I value, but it
might well be that they do very well.
When you're adding a ballot it's the participation criterion that's being tested. Philosophically I'd say participation and monotonicity are fairly similar, and neither one seems obviously more important than the other. But mathematically they are different, and it just happens to be much easier for a method to pass monotonicity than participation. The restrictiveness of participation means it it often largely ignored as a criterion. However, I also think it is a big plus for approval and score. Score might not be a realistic aim, but approval voting has a lot going for it and has support behind it.
Toby
On Wednesday, 7 February 2024 at 17:46:03 GMT, Michael Ossipoff email9648742@gmail.com wrote:
How can a method be called “monotonic” if adding a ballot that votes X over y can change the winner from X to y ???
Another reason to prefer Approval & Score.
What’s the difference between changing a ballot & adding one? Well, one is legal & one isn’t, & happens whenever another person enters the polling-place.——-
Yes, Approval, but not Score, has the enormous advantage of being the
absolute minimal way of allowing expression & counting of preference or
relative acceptance among more than 2 candidates.
…& therefore the unique completely unarbitrary method.
On Wed, Feb 7, 2024 at 10:34 Toby Pereira tdp201b@yahoo.co.uk wrote:
When you're adding a ballot it's the participation criterion that's being
tested. Philosophically I'd say participation and monotonicity are fairly
similar, and neither one seems obviously more important than the other. But
mathematically they are different, and it just happens to be much easier
for a method to pass monotonicity than participation. The restrictiveness
of participation means it it often largely ignored as a criterion. However,
I also think it is a big plus for approval and score. Score might not be a
realistic aim, but approval voting has a lot going for it and has support
behind it.
Toby
On Wednesday, 7 February 2024 at 17:46:03 GMT, Michael Ossipoff <
email9648742@gmail.com> wrote:
How can a method be called “monotonic” if adding a ballot that votes X
over y can change the winner from X to y ???
Another reason to prefer Approval & Score.
What’s the difference between changing a ballot & adding one? Well, one is
legal & one isn’t, & happens whenever another person enters the
polling-place.
——-