[EM] Ranking unacceptable candidates.

[Juho Laatu]

Some shot comments below. I'll also forward the missing message to the EM list.

On 20.10.2013, at 14.40, Michael Ossipoff wrote:

> I hadn't intended to post my previous posting yet. I sent a copy to
> myself, to work on it later, and EM's address wasn't in the "To:"
> field. But it got sent to EM anyway.
> 
> So let my reply to the rest of the message:
> 
> Juho--
> 
> On Sat, Oct 19, 2013 at 1:29 AM, Juho Laatu <juho4880 at yahoo.co.uk> wrote:
>> On 19.10.2013, at 3.31, Michael Ossipoff wrote:
> 
> 
> A number of party platforms offer IRV. That makes it's merits a very
> relevant matter.
> 
> The only way to really judge what the voting systems and their
> properties would be like, is to actually vote a ballot, for those
> voting systems.
> 
> So, just for your own experience (you needn't post it if you don't
> want to), write your ranking of the political parties (of the U.S.,
> &/or your country of residence)--not necessarily a sincere ranking,
> but the ranking that you'd actually vote if it were a geuine official
> public political election, by whatever method you're trying-out, such
> as IRV.
> 
> I guarantee that your IRV ranking will be sincere, if you expect to be
> in a mutual majority. When you try it out, IRV looks pretty good.
> 
> ...And if you aren't in a mutual majority? Well, if your
> policy-preferences are good, and if other people's judgment is as good
> as yours, then you _will_ be in a mutual majorty. A mutual majority is
> the important kind of majority. A cohesive majority of voters who all
> support that majority's favorites over all others, because they all
> agree on the most important matters of policy.
> 
> But try writing a ranking, because that's the only way you'll really
> know how you feel about the methods and their properties.
> 
> That's why I like polls.
> 
> Anyway, I feel that IRV's advantages outweigh its disadvantages.

Depends of course also on the environment and what the other alternative is.

> But
> IRV's disadvantages are gotten-rid-of by Benham and Woodall, which
> retain IRV's advantages.
> 
> ...almost all of IRV's advantages. They lose IRV's LNHa & LNHe
> combination. But the Condorcet Criterion (CC) compliance that they
> bring is worth that.
> 
> You wrote:
> 
> But generally I lean towards Condorcet style methods (in typical
> political elections where the target is to elect good compromise
> candidates).
> [/quote]
> 
> Yes, and that means avoiding dis-satisfied majorities, by electing CWs
> or (when there isn't a CW) Smith-set members.

All Condorcet methods are not Smith compatible. I can live with both approaches.

Juho


> Additionally, electing
> the CW avoids the worst of IRV's strategy-problems for voters who
> aren't in a mutual majority.
> 
> But you can have both. You can have IRV's MMC and no-chicken-dilemma,
> and also have the Condorcet Criterion, and the Smith Criterion--with
> Benham or Woodall.
> 
> Michael Ossipoff
> ----
> Election-Methods mailing list - see http://electorama.com/em for list info