[EM] Condorcet meeting

[C.Benham]

Kristofer,

> Then I construct a method that's intended to be cloneproof, and then you
> say that clone independence is a bad thing, because more extreme clones
> can support more moderate ones and make the latter win.
I don't recognise that as a quote or paraphrase of what I said.  I 
referred to extreme and
moderate candidates on the same wing.

"Clones" are candidates that all the voters rank together. Although all 
the extreme wing voters
might give their second preference (if they don't bullet vote) to the 
moderate candidate on the same wing,
most of the moderate wing supporters typically prefer a centre candidate 
and/or maybe even a moderate
candidate on the other wing to the extreme candidate on their wing.  So 
they are usually not clones.

As to it being a "good thing", more voters having their favourite on the 
ballot encourages greater turnout
and the lower preferences of sure losers can obviously effect the result 
and that is just as likely to help any
given major candidate as harm it.

> Should this kind of method pass clone independence in
> the sense that cloning some group's candidates should never change the
> number picked from that group to be finalists -- or not?
I don't  see any particular point in it doing that.  We should try to 
allow every significant "group" at least
one finalist candidate.  All good election methods that we might use for 
the final can easily handle a few nobodies
on the ballot, so subject to logistic practicability we should err on 
the side of doing that.

> First I suggest using Approval, to which you say this will lead to the
> selection of a bunch of clones.
If it is just "select the top N most approved candidates" then yes that 
could just lead to all the finalists
being clones of the Approval winner, and if approval ballots are used 
there will be pressure from some
"in the real world" to use that most simple and seemingly obvious system.

But say despite all that we use approval ballots in the primary to 
choose N finalists.

I suggest first the most approved candidate qualifies, then the 
candidate with the most approval
opposition to the approval winner qualifies, and then the candidate with 
the most approval opposition
to those two qualifies, and so on, each time selecting the candidate 
that is most approved on ballots that
don't approve any of the already qualified candidates until we have the 
desired N number of finalists.

That would be much much better than just the FPP or Approval top N 
candidates, but still a lot worse than
my IRV last N suggestion.

Chris B


> *Kristofer Munsterhjelm*km_elmet at t-online.de 
> <mailto:election-methods%40lists.electorama.com?Subject=Re%3A%20%5BEM%5D%20Condorcet%20meeting&In-Reply-To=%3Cec28b577-92c6-73bf-8e2d-dcf02d0bf778%40t-online.de%3E>
> /Thu Aug 31 05:26:41 PDT 2023/
>
>  *
>
>
> ------------------------------------------------------------------------
> On 8/30/23 09:17, Chris Benham wrote:
> >//>/Kristofer, />//>/IMHO this is all far too complicated and completely wrong-headed. />//>/For one thing, it seems to assume that major candidates have some />/interest in, or in some way benefit from, displacing minor candidates />/off the final ballot. />//>/In fact, with voluntary voting, they are just as likely to be harmed />/by them doing that as helped. Major moderate wing candidates can />/benefit by minor (sure loser) more extreme candidates on the same />/wing being on the ballot, because they inspire more voters to turn />/out who will help that moderate wing candidate to defeat the other />/major candidate/s. /
> Then I'm sorry, but I don't know what the desiderata are.
>
> First I suggest using Approval, to which you say this will lead to the
> selection of a bunch of clones.
>
> Then I construct a method that's intended to be cloneproof, and then you
> say that clone independence is a bad thing, because more extreme clones
> can support more moderate ones and make the latter win.
>
> So which is it? Should this kind of method pass clone independence in
> the sense that cloning some group's candidates should never change the
> number picked from that group to be finalists -- or not?
>
> Maybe this shows that we need to nail down what the method is supposed
> to accomplish before designing it.
>
> -km
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20230902/a5748f13/attachment.htm>