Has anybody submitted Benham's GLE (Gross Loser Elimination)? Another name
could be Min Fuss Elimination:

Until some candidate is undefeated among those still in play ... eliminate
the one showing the fewest losing votes in its worst remaining pairwise
contest.

On Sat, Apr 6, 2024, 12:33 PM Ted Stern dodecatheon@gmail.com wrote:

If not, then let’s add it to the nominations list.

On Sun, Apr 7, 2024 at 23:57 Forest Simmons forest.simmons21@gmail.com
wrote:

On 2024-04-08 07:26, Closed Limelike Curves wrote:

But if you're right about the results of its clone dependence, isn't
STAR just Range with cowbell? That is, it would reduce to Range itself
due to the existence of rated (rated-equal) clones.

-km

On 2024-04-08 08:57, Forest Simmons wrote:

Nope, so I'm guessing that you're nominating it :-) The new list is:

Approval
RP(wv)
Schulze
IRV
Plurality
Majority Judgement (as a category; includes usual judgement etc.)
Approval with manual runoff
Copeland//Borda (also called Ranked Robin)
MinMax(wv)
Black
STAR
Smith//Score
Baldwin
Benham
Woodall
Schwartz-Woodall
Smith//Approval (implicit - of all ranked)
Smith//Approval (explicit - specified approval cutoff)
Margins-Sorted Approval
Smith//DAC
Margins-Sorted Minimum Losing Votes (equal-rated whole)
RCIPE
Gross Loser Elimination.

If anyone would like to provide short descriptions of one or more of the
methods, that would be much appreciated. I'll see if I can come up with
some later, myself.

-km

Yes, thanks for posting that list-update. My count is 23 also. With so
many, it’s easy to miss one.

Yes, more should be said about the methods.

In fact, doesn’t ranking the methods required a statement, for each one, of
its advantage(s) over other methods? What guarantees or other desirable
properties distinguish each method from the others?

There’s nothing wrong with “electioneering” before an election. In fact,
isn’t it necessary?

I’ve told of some unique advantages of Approval, &  why it’s what I propose
for public political elections.

One thing that I might not have emphasized enough is that, when there are
unacceptable candidates, as I claim there are, the absolutely minimal
Approval has the obvious simple optimal-strategy of approving all of the
Acceptables & none of the Unacceptables.

Compare that to the favorite-burial defensive- strategy of IRV & most
non-wv Condorcet, where (to fully, maximally protect Acceptables against
burial) one must try to guess which acceptable compromise(s) to insincerely
rank over Favorite, & in what order… i.e. where to successfully combine
compromise-support with people who agree with you about some of the
Acceptables.

An intractable strategic morass.

By its ultimate minimalness, Approval is unmatched for easy proposal,
implementation (can be zero-cost), administration, & audit against error &
count-fraud.

I emphasize the count-fraud, because it’s easy to forget that it doesn’t
matter how good a method ideally is, if its complicated,
computation-intensive, un-transparent count facilitates count-fraud.

For polls, & for jurisdictions where rankings are insisted-on, I’ve told
why & how wv Condorcet fully achieves the Condorcet promise—&, in fact,
achieves nothing less than the the ideal for voting-systems: Absence of
need for defensive strategy…allowing uninhibited free sincere expression of
preference-ordering.

On Mon, Apr 8, 2024 at 12:49 Kristofer Munsterhjelm km_elmet@t-online.de
wrote:

Could you elaborate on this a bit more? To me it sounds like:

while (no Condorcet winner) { eliminate the Minimax winner (!) }

elect the Condorcet winner

Or should that have been "most" instead of "fewest" losing votes?

On 4/8/24 02:57, Forest Simmons wrote:

Yesish—although, with STAR, I think of "clones" as rated-next-to-equal.
Clones in the sense of rated-equal-on-all-ballots don't really exist, so
any "clones" would probably be copartisans. (Which is why the runoff
probably makes elections a bit more Condorcet-efficient). In practice,
that's a huge improvement on IRV and FPP for Condorcet advocates (since
Score has a high Condorcet efficiency).

To be honest, every well-designed voting rule—anything that doesn't make
strategy too complicated or require order-reversal—will probably converge
to a rough approximation of a maximal lottery (since that's the
group-strategyproof Nash equilibrium).

On Mon, Apr 8, 2024 at 8:32 AM Kristofer Munsterhjelm km_elmet@t-online.de
wrote: