[EM] it's been pretty quiet around here...

[Juho]

Here are some late comments. I first thought that I'd upload some sw  
too to emphasize my viewpoints, but since I couldn't agree with myself  
on how to handle some ties I took a timeout on that front :-).

On Aug 15, 2010, at 3:00 PM, Peter Zbornik wrote:

> Hello all,
>
> Haven't got much news lately, been busy with school, so election  
> methods have taken summer holidays.
> I'm sending a rant below about how to sell proportional elections,  
> most of which is old news I guess.
>
> After trying to promote proportional elections, proportional ranking  
> elections and Condorcet single-winner to my party, the greatest  
> hurdle was to explain how the methods work.
> Especially the top-down ranking multi-winner case.
>
> Simplicity is THE most important factor when trying convince people  
> without a computer/maths degree, especially as I want to use  
> proportional top-down ranking methods for party lists and possible  
> council elections.
>
> Explaining beatpath methods is not easy, and it does not become  
> easier when you go to proportional ranking STV.
> Ranked-pairs seems to be easier to explain and code than Schulze at  
> a first glance.
> I don't know which method would be simpler to explain than Schulze- 
> STV (which also has some nice properties, which makes it easy to  
> explain).
> On the other hand, Schulze-STV handles incomplete ballots completely  
> differently (proportional completion) than standard Schulze methods  
> (winning-votes), which is rather annoying.
> I am not sure how well the multiwinner extention CPO-STV handles  
> large number of votes, seats and candidates although Juho was kind  
> enough to program a web-app.

CPO-STV and many other ranked proportional methods are a computational  
challenge if the number of candidates and votes is large. It is not  
too difficult to write a program that with good probability finds the  
best winner quickly, but such uncertainly may be difficult to market.

>
> I will propose standard STV and proportional top-down rankong STV to  
> my party as an alternative multi-winner method as a safeguard.
> I will promote, not that the current voting system will be replaced,  
> but that each organisation within the party (local, regional etc)  
> can decide on their own on what methods to use, from a set of  
> officially approved methods.
> I mean some people in our party advocate that the elections amount  
> to random sampling of seats from a set of candidates.

Not bad. Maybe you will act as a good testing ground for new methods  
in the future :-).

>
> It would be great, if you could aggree on a method to promote.
> Why not try to vote? :o)

That might be a big fight. I once proposed to the Range proponents to  
use Range voting to decide which voting method is best but I did not  
get any support to this idea (maybe better so for the Range  
promoters) :-). Maybe approval would be one working method, not to  
pick the winner but to provide data on what methods different expert  
consider acceptable for some particular use case. I'm actually  
somewhat surprised on how difficult it is for research oriented people  
to even find approximate consensus on which methods are good for the  
most usual needs. I guess many people are more "promoters" of their  
own favourite methods than "scientists" when they have to decide  
between these two approaches.

> Or why not promote both Schulze and Ranked pairs, but with one  
> preferred of these two options.
> If you start voting in this forum, you might also want to consider  
> introducing a "blocking vote", meaning that the person is so  
> strongly dissatisfied with the vote, that he/she plans to leave the  
> forum etc. if the majority alternative will win. If a significant  
> number of blocking votes is cast (say one vote or 10% of the votes),  
> then there will be re-elections after a new round of discussion.

As discussed above, maybe one could collect such opinions without  
trying to decide which method is the absolute winner. (One problem is  
that list members and voters probably are not a representative set of  
the whole scientific community.)

>
> I guess what most organisations need, is what I wrote down, when  
> hunting for a good election method for the Czech green party.
> 1. a simple method - I think I wrote this before, this is the main  
> criterion
> 2. proportional ranking multi-winner elections for party lists and  
> board/council elections.

There are also other alternatives than proportional ranking based  
approaches.

> 3. draft text to use in statutes
> 4. an open-source freeware program

Availability of such solutions could really help various organizations  
to improve their decision making. (If I had more time I'd do more work  
on the web. Maybe others will do the work faster.)

>
> The points above are maybe not so cool mathematically, but they will  
> most certainly help promoting Condorcet methods.

Condorcet people are quite poor promoters (maybe more "absent minded  
scientists" from this point of view), with the exception of Markus  
Schulze.

> The problem now is not the lack of methods but "voting packs" that  
> organisations can adopt with little work from their side.

Ok, I'll do something in case I'll find time :-).

>
> Otherwise - about the voting methods:
> I strongly consider a second  top-two runoff election between the  
> Condorcet winner and the candidate with the most first preference  
> votes as a safeguard against dark horses and against criticisms from  
> the unconvinced.
> Do you think it is a good idea?

Probably the combined method is not any better. But if that is  
necessary to make the reform acceptable, then why not. In theory the  
Condorcet winner (if one exists) should win also the second round. The  
main impact is maybe the added second round of discussions that may  
change also sincere opinions.

>
> For multiple-winner proportional ranking - STV elections will be one  
> alternative, as it is relatively simple to explain, haven't yet  
> found an "as simple as STV" condorcet multiwinner method. If you  
> know one let me know.
>
> I still haven't got to the point where I start writing down draft  
> text into the statutes and different "customizations".
>
> I have text in statutes for STV (the american greens) and for  
> Schulze single-winner.
> Writing down the Condorcet- proportional ranking STV (like Schulze  
> STV) will prove to be a challenge, which I am not sure I will be up  
> to, some help would be great here, but I will come back to you on  
> this issue.

I guess there is some material also for other STV methods that are in  
use today. For the new ones, maybe the promoter of each method is  
happy to do some work. (Traditionally the descriptions are accurate  
enough to allow hand counting by non-experts. For some of the more  
complex methods that approach will not work but should be relaxed.)

Juho


>
> Best regards
> Peter Zborník
>
> On Sun, Aug 15, 2010 at 12:41 PM, Kristofer Munsterhjelm <km-elmet at broadpark.no 
> > wrote:
> robert bristow-johnson wrote:
>
> On Aug 14, 2010, at 5:37 AM, Kristofer Munsterhjelm wrote:
>
> Since FV thinks IRV is so nice, it's to their benefit to link  
> preferential voting, the concept, to IRV, the method, so that others  
> thing "oh, either IRV or Plurality". Since IRV appears better than  
> Plurality (at least until the summability issues are encountered),  
> this makes it relatively easy to slip in IRV, and the theory then  
> goes, to go from IRV to STV, which is much better.
>
> It doesn't appear that we can change FV's minds from IRV to  
> something better (like Condorcet). When you dig really far down, the  
> issue boils down to "weak centrist! Condorcet winner! weak centrist!  
> Condorcet winner!" and there you go -- and then they sprinkle LNHarm  
> and *perhaps* burial resistance on top.
>
> my experience with Rob Ritchie is that IRV is the only method with  
> an ice cube's chance in hell of being adopted in a governmental  
> election.  the claim is that IRV can be directly related to the  
> traditional delayed runoff and that it is no different, except for  
> no delay (which has the measurable difference in that many more  
> voters participate in the instant runoffs than in the delayed  
> runoff).  but, for that to be true, it should have no more than 2  
> rounds with the top two of the first round going into the second and  
> final round.  of course, that doesn't fix the problems demonstrated  
> in the 2009 Burlington mayoral election (because the "true majority"  
> winner would not have made it to the runoff in either case).
>
> IRV is an emulation of an exhaustive runoff, not of top-two. The  
> emulation of top-two, the Contingent vote, is worse - but I see your  
> point, since IRV is at its surface similar enough to top-two runoff  
> to seem a reasonable way of automating the latter. One might show  
> that IRV denies the people the ability to change their votes between  
> the rounds, but the problem is really that IRV, as a voting method,  
> doesn't give good outcomes.
>
> If we discuss voting in a mechanical manner, as something that has  
> to be done a certain way, then IRV will have an advantage because  
> its mechanics are similar to that of ordinary runoff, which is  
> trusted. I don't think that is the right approach, but I can see how  
> it would appear as such to the voters.
>
> If that is what makes non-IRV methods unlikely to succeed, then it'd  
> seem we have three ways of making the Condorcet method in question  
> pass. The first would be to let organizations use it, like Schulze  
> is being used by technical ones right now, so that the method itself  
> (however complex) becomes trusted. The second would be to make the  
> mechanics similar to something that *is*, as in the focus on  
> championships, tournaments, round robins...
> In an indirect manner, you might also say that Ranked Pairs is  
> similar to majority rule since it goes down affirming majorities  
> until the winner is clear. It's simple to explain because of that,  
> but is it similar enough? I don't know.
> The third would be to somehow manage to inform the people that  
> looking at the outcomes is the right way of considering voting  
> methods. It is intuitive, so it could work as long as the method  
> isn't *too* opaque, but it would have to be pulled off right.
>
> Markus has a good point about Condorcet supporters splitting their  
> own vote by not being sure which method one should support.
>
>
> Cardinal ratings technically pass both because it can pass IIA since  
> it doesn't care about universal domain. However, I think that CR  
> (Range, Score, etc) will be hard to get passed, since it doesn't  
> even pass Majority. Even if Warren is right and social utility  
> comparisons are better than majority rule, most people associate  
> democratic fairness with that if some candidate is preferred by a  
> majority, he should win. There are also the tactical issues: CR  
> reduces to Approval (as Youtube raters found out)
>
> and Approval can reduce to Plurality bringing along the same  
> strategy problems of Plurality.
>
> How so? If you vote Plurality-style, it never harms you to vote for  
> those you prefer to the one you'd vote for in Plurality. You might  
> get a Plurality outcome, but you might also get more (which is  
> better than what Plurality has to offer).
>
>
> and pretty soon voters who want their vote to count must haul around  
> concepts like "maybe frontrunner, plus" (LeGrand's Approval strategy  
> A), something which really should be inside the method rather than  
> outside.
>
> Thus we can't follow FV; and while we could advocate cardinal  
> ratings, I don't think that would be very successful (and in any  
> event, should be DSV instead). That leaves Condorcet, and so I think  
> there should be an organization or group or at least some sort of  
> coherent support for Condorcet.
>
> well, there used to be a condorcet.org or condercet.com (neither  
> have an active web page, although the .com has a dumb page put up by  
> the registrar, just like my audioimagination.com does).
>
> Yes, that's a good idea. Condorcet.org is owned by Blake Cretney,  
> while condorcet.com seems to have been snarfed up by NAmedia  
> (namedia.com), which is in the business of domain parking.
>
> If we were to call it something else, or focus it on a particular  
> method, another domain might be better, but it could still link to  
> the Condorcet site for detailed information.
>
>
> What should such a group do? First, it should state that the concept  
> of ranked voting is different from what method may be used as its  
> back-end. Second, it should have a clear and easily understandable  
> name for Condorcet, or for the Condorcet method it settles upon. The  
> former could be done more simply: "round robin voting", "maximum  
> majority voting", "championship" or "tournament" voting (but beware  
> of equating it with an elimination tournament), etc.
>
> Warren has used the term "beats-all winner" for the Condorcet winner.
>
> I agree with the others - I think a better term would have to be  
> found.
>
>
> as much as i like the Schulze method, since it is so much more  
> difficult to explain and for a lay person to comprehend, there will  
> always be some suspicion around it in the minds of people who want  
> to "Keep Voting Simple" (the signs from the IRV opponents in  
> Burlington).  they won't like any Condorcet, because most  
> fundamentally do not like the ranked ballot, but since Ranked Pairs  
> (which is simpler to understand) and Schulze pick the same winner if  
> there is a CW (the most common case, i believe) and if the Smith Set  
> is no larger than 3 (which, i believe will take up the other 1% of  
> cases), then i cannot imagine how it could be more efficacious to  
> promote Schulze Beatpath over Tideman Ranked Pairs.  but, also for  
> simplicity, maybe the best method to sell is simple Condorcet and  
> then, if no CW exists, pick the plurality (of first choices)  
> winner.  i'm not saying it's best, just that it's simple to  
> understand and that the likelihood that no CW exists will be small  
> (or is, at least, believed to be small).
>
> Condorcet,Plurality is not a very good method. If simplicity is the
> be-all end-all, I would pick something like the "offense least  
> reversal"
> method. I don't know if it has a proper name, but basically just sum  
> up
> the number of voters who stand on the winning side of the matchup and
> the candidate with the highest score wins. This is Condorcet because  
> a CW would get a score equal to the turnout, something that cannot  
> be beaten by any other candidate.
>
> Another option is plain old Copeland, which would be well known from
> sports. Two points for a victory, one for a tie, none for a defeat.  
> The
> advantage of Copeland would be that it sidesteps the WV/margins issue,
> and it's also Smith. The disadvantage is that it ties a lot, but  
> since it does, championships or tournaments that use it probably has  
> a tiebreaker and so one could use that tiebreaker as well. It's also  
> relatively coarse-grained (since it only considers direction of  
> wins) and so may not distinguish weak from strong centrists.
>
> Neither of these methods are cloneproof, but neither is Plurality.
>
>
> the other method, BTR-IRV (which i had never thought of before  
> before Jameson mentioned it and Kristofer first explained to me last  
> May), is a Condorcet-compliant IRV method.  i wonder how well or  
> poorly it would work if no CW exists.  i am intrigued by this method  
> since it could still be sold to the IRV crowd (as an IRV method) and  
> not suffer the manifold consequences that occur when IRV elects  
> someone else than the CW.
>
> You might also be able to pass Borda-runoff (Nanson or Baldwin) - it's
> like IRV but with Borda counts, and because the CW always has an
> above-average Borda count, that method passes Condorcet. It's not
> monotone (but IRV isn't either). Nanson has actually been used in
> governmental elections (Marquette, Michigan in the 1920s) and has also
> been used by Australian universities.
>
>
> does "BTR" stand for "bottom two runoff"?  and who first suggested  
> this method?  is it published anywhere?  Jameson first mentioned it  
> here, AFAIK.  the advantage of this method is that is really is no  
> more complicated to explain than IRV, and it *does* resolve directly  
> to a winner whether a CW exists or not.  i am curious in how, say  
> with a Smith Set of 3, this method would differ from RP or Schulze.
>
> Anthony O'Neal first suggested BTR-IRV, and yes, BTR stands for  
> "bottom
> two runoff" (but also a reference to/pun on "better"). BTR-IRV is
> probably the simplest single modification you can do to IRV to get
> Condorcet, but it's not that good a Condorcet method. It's not  
> summable,
> for instance. I thought that Nanson/Baldwin weren't either, but
> apparently it's possible to derive Borda scores from a Condorcet  
> matrix.
>
> With three candidates, the "modified IRV" Condorcet methods differ  
> from
> RP and Schulze mainly in that they don't elect the one whose worst
> defeat is the least. With a Smith set of three but more candidates  
> in total, the methods may differ further, since BTR-IRV doesn't pass  
> Smith.
>
> ----
> Election-Methods mailing list - see http://electorama.com/em for  
> list info
>
> ----
> Election-Methods mailing list - see http://electorama.com/em for  
> list info

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20100903/4800fc35/attachment-0003.htm>