[EM] Fwd: (6) To Kristofer and everyone: MJ best to ‘tolerate’
Jameson Quinn
jameson.quinn at gmail.com
Sat Sep 17 14:05:14 PDT 2016
2016-09-17 0:42 GMT-04:00 C.Benham <cbenham at adam.com.au>:
> On 9/16/2016 4:22 AM, steve bosworth wrote:
>
> MJ simply asks each voter to grade each candidate when judged against each
> voter’s own criteria of what an EXCELLENT candidate would look like. Any
> candidate judged to be less than EXCELLENT must be graded either as VERRY
> GOOD, GOOD, ACCEPTABLE, POOR, or REJECTED. Balinski and Laraki refer to
> each voters own criteria for grading candidates as being ‘absolute’ (but
> this is only in the sense that these criteria should be independent of any
> one set of candidates that might be seeking election).
>
>
> C: To begin with superficial aesthetics, the grades should have simple
> neutral names (like A B C D E F) and the ballot "request" should be
> something like:
> "Give your favourite candidate or candidates an A and and your least
> preferred candidate or candidates an F and any intermediate candidates
> whatever
> grade you see fit. Default rating is F."
>
> As it is if in a given election A , by my "criteria that should be
> independent of any one set of candidates that might be seeking election" ,
> such as those doing so in election A,
> I rate my favourite candidate as being merely "Acceptable" I would resent
> having to either (a) accept that my vote will have less influence on the
> result than voters who rate
> their favourite as "Excellent" or (b) "lie" and falsely indicate that I
> rate my favourite as "Excellent".
>
To me, this parses as either an overgeneralization or a misapprehension of
how MJ works.
But before I get to my strongest criticism in that regard, let me discuss
why I feel the issue is less important than one might thing. Saying that
the criteria should be "independent of any one set of candidates that might
be seeking election" does not mean that it should be "independent of the
population distribution of candidates that typically seek such elections,
weighted by win probability". A reasonable calibration for 5 ratings levels
(ABCDF) and a metric under which that distribution is normal might put the
cutoffs at something like +1.4, +.7, 0, -.7 standard distributions; or in
other words, at the 90th, 75th, 50th, and 25th percentiles of historical
winners and close second-place finishers.
Assuming that the distribution of people who run for an office is at least
1/3 wider than the distribution of those who can win it (reasonable) and
that it's centered at the same place for our voter (probably not true for
all, but a reasonable assumption in order to avoid pesky constants), then
those cutoffs would be at about 83, 67, 50, and 33 percentiles of the full
set of candidates. Assuming independence (which is an ABSOLUTELY
unjustified worst-case assumption, purely for illustrative purposes), a
5-candidate race would have nobody at B or higher under 10% of the time. If
you want to talk only viable (potentially-winning) candidates, that's still
under 20%. So the case where a voter would have to be "dishonest" by two
grade levels in order to use a full spectrum is unusual.
OK, now the real objection.
You say that anybody who honestly votes their favorite below top must
"accept that [their] vote will have less influence on the result than
voters who rate their favourite as "Excellent". But if their favorite
candidate gets a median rating below the rating they give, then their
honest vote is in fact fully strategic. Generally, given the calibration
above, this will be the case with very high probability. In order to put
specific numbers on that probability, I'd have to make a thicket of tenuous
assumptions about how voter's candidate utilities distribute and correlate,
so I won't do it, but I would be more surprised if the probability were
under 90% than if it were over 95%.
> MJ poses as being somewhat like a jury in a trial, or a panel that judges
> say a competitive performance of Gymnastics or Diving.
>
> But elections for powerful public political elections are very different.
> In those cases the jurors/judges are more-or-less "disinterested", i.e. it
> doesn't really make any
> possible difference to their lives who wins the competition or whether the
> accused is jailed or set free. In elections who wins the election could
> have a big effect on the
> lives of voters.
>
> Also in those other cases there is usually general agreement what an
> excellent sporting performance looks like and what a terrible sporting
> performance looks like and
> what constitutes clear proof of guilt or innocence. In elections voters
> often have opposing ideologies, i.e. very different ideas of what policies,
> priorities, political philosophy,
> diplomatic/military strategies the election winner should have.
>
> Another difference is that in those other cases the people on the jury in
> a trial or the panel judging a sporting performance base their decisions
> the same evidence. The jurors
> all hear the same evidence and arguments and base their verdict on that.
> Likewise the judging panel all closely watch the same performance and give
> their scores based purely
> on that.
>
> But voters in public elections vary widely in terms of what information
> they get, and the quality and quantity of that information. And of course
> much of the "information" they
> use might be false or misleading, generated by those with a big interest
> in who wins the election.
>
> Leaving aside the strategy incentive for voters to only use the very top
> and very bottom grades, suppose that all the voters rate the candidates as
> sincerely as they can
> in the way MJ "invites" them to. Suppose that that there are only two
> candidates with any hope of winning, X and Y. Suppose I (Chris) think that
> X is clearly better than Y
> and you (Jameson) think the opposite. Suppose my rating of X is B and
> of Y is D, and your rating of Y is A and of X is F.
>
> Your pairwise preference will have greater weight than mine.
>
No, it won't. Either they will both have equal weight (most probably), or
Chris's vote will have no weight at all (improbable). There is no case
where they will both have an impact but Jameson's impact will be greater.
First off, you yourself say you prefer letter grades over verbal
categories. So I've edited your senario above to use corresponding letter
grades, and to use X and Y for the candidate names.
So the hypothetical votes are:
Chris: X:B, Y:D
Jameson: X:F, Y:A
If the winning median is anything from B- to D+, then our votes had exactly
the same impact. A winning median in this range is probable if people
calibrate as I suggested above. And when it falls outside of that range, it
will probably be predictable beforehand, giving Chris a chance to decide to
be more strategic for one election only.
Is that fair? According to the MJ philosophy your vote should have
> greater weight because you are more
> "enthusiastic" in your support for B over A. Does this greater enthusiasm
> mean that your opinion that B is better than A is more likely to be correct
> than my opposite
> opinion?
>
> Also, MJ seems to offer less scope for manipulative voting than any other
> method.
>
>
> C: As Kevin has pointed out, sincere voters are less likely to be at a
> disadvantage than with Range (aka Average Ratings) but in both the
> voter's best strategy is to only
> use the two most extreme ratings. If all the voters do that the method is
> just Approval.
>
> What exactly is your definition of "manipulative"?
>
> I still favor MJ even though it is theoretically vulnerable to
> ‘Later-no-harm’ (LNH).
>
>
> C: I don't particularly care about Later-no-Harm. It encourages the
> expression of preferences that may be very weak, and to the extent that
> they are decisive they
> would tend to lower the "Social Utility" (SU) of the winner. And the
> expressed preferences are also more likely to be the result of unprincipled
> mutual back-scratching
> deals between candidates.
>
> I put a greater value on Later-no-Help (which MJ, along with MTA and MCA
> meets). Ideally there should be weak zero-info truncation incentive.
>
> But MJ has a very strong truncation incentive. It's compliance with LNHelp
> is in practice useless if the voters should all truncate.
>
> IRV meets both of Later-no-Help and Later-no-Harm, and in my opinion it is
> the best of the methods that meet Later-no-Help.
>
> Other methods I like fail both. That is better than only meeting LNHarm
> and so having a random-fill incentive, or only meeting LNHelp and having a
> very
> strong truncation incentive.
>
> For reasons I might give in another post, I don't much like MAM. A
> simpler Condorcet method I like is Smith//Approval:
>
> Voters ignore candidates they don't approve and rank the rest.
> Equal-ranking allowed. Elect the most approved member of the Smith set.
>
> The "Smith set" is the smallest set of candidate/s who all pairwise beat
> all (if any) outside-the-set members. A single-member "Smith set" is the
> Condorcet winner.
>
> Compliance with both FBC and Condorcet is impossible. MJ meets FBC.
>
> A MJ-like method that is simpler and in my view better is Majority Top
> Approval (MTA).
>
> It uses 3-slot ratings ballots. Default rating is Bottom. If any
> candidate is rated above bottom on more than half the ballots, elect (if
> there is more than one) the
> one of those with the highest number of top ratings. Otherwise elect the
> candidate with the highest number of above-bottom ratings.
>
> The voters' best strategy is to normally use only the top and bottom
> ratings slots, but the middle slot is handy if there is one or more
> candidate the voter is
> unsure how should rate on a 2-slot ratings ballot, or if the voter is
> prepared to maybe take a small strategic risk for the sake of being more
> expressive.
>
> But a more complex method I much prefer is IBIFA.
>
> http://wiki.electorama.com/wiki/IBIFA
>
> Chris Benham
>
>
>
>
>
>
>
>
> ________________________________________
>
> From: Kristofer Munsterhjelm <km_elmet at t-online.de> <km_elmet at t-online.de>
>
> Sent: Wednesday, June 1, 2016 9:14 PM
>
> To: steve bosworth; election-methods at lists.electorama.com
>
> Subject: Re: [EM](6) To Kristofer and everyone: MJ best to ‘tolerate’
>
> ___________________________________________
>
>
>
> To Kristofer and everyone:
>
> Kristofer, while our most recent exchange was on June 31st, I want to
> thank you for forcing me to think more carefully and completely. As a
> result of the clarifications you offered me, when combined with my recent
> EM discussions with Kevin and Jameson, I am currently favoring MJ over all
> other methods for electing a president.
>
> MJ simply asks each voter to grade each candidate when judged against each
> voter’s own criteria of what an EXCELLENT candidate would look like. Any
> candidate judged to be less than EXCELLENT must be graded either as VERRY
> GOOD, GOOD, ACCEPTABLE, POOR, or REJECTED. Balinski and Laraki refer to
> each voters own criteria for grading candidates as being ‘absolute’ (but
> this is only in the sense that these criteria should be independent of any
> one set of candidates that might be seeking election). These criteria
> are not rankings but they can be used to make rankings. MJ’s winner is
> the candidate whose majority ‘median-grade’ is higher than any other
> candidate.
>
> For example, I see MJ as both better than IRV and MJ because:
>
> 1.
>
> I believe that ordinary citizens would more easily understand both how
> to mark MJ’s ballot and how the winner is discovered using its count. They
> would find it even more difficulty in understanding how MAM’s ballots are
> counted.
> 2.
>
> At the same time, both MJ and MAM have the following similar advantage
> over IRV: All MAM voters’ ‘rankings’ and MJ voters’ ‘gradings’ of all
> candidates continue to count until their respective winners are discovered.
> Some of the IRV voters’ ‘rankings’ are not counted after any of its
> candidates are eliminated and before IRV’s winner is discovered.
> 3.
>
> However, MJ has another advantage over MAM: because ‘grades’ are more
> evaluatively clear than are ‘ranks’, MJ’s ‘‘majorities of grades are …
> considerably more discerning decisions than are [MAM’s] majorities of preferences’
> (Belinski & Laraki, Majority Judgment, p.283)?
> 4.
>
> Moreover, unlike the different intensities of each MJ voter’s ‘grades’
> (evaluations) for each candidate, the different intensities of each MAM
> voter’s ‘rankings’ recorded on her ballot are not directly counted. For
> example, an MAM voter’s different intensities of preference for A over B
> and A over G in the following list of preferences are not differently
> counted: A>B>C>D>E>F>G. In contrast, all the ‘grades’ given to all
> candidates by all MJ voters continue fully to count until the highest
> ‘majority-grade’ winner is discovered. Thus, I see MJ as the most
> democratic method because it both wastes no citizen’s vote and is most
> likely to elect the candidate who is supported as enthusiastically as
> possible by the larges majority of citizens.
> 5.
>
> Also, MJ seems to offer less scope for manipulative voting than any
> other method. B&L argue that MJ is entirely strategy-proof with
> regard to grades. At the same time, if B&L’s mathematical proofs are
> correct, even when and if voters might instead use MJ’s ‘gradings’ as
> ‘rankings’, MJ structurally cuts by almost ‘half’ all the
> opportunities for all the manipulative strategies that are offered by all
> the methods other than MJ (Belinski & Laraki, Majority Judgment,
> pp.14, 15, 189-198, 212, 245, 282-292). (I assume that your
> mathematical knowledge is currently much superior to mine. Consequently,
> I’m hoping that you will be able to tell me whether the above B&L claim of
> ‘almost half’ is fully justified by B&L’s mathematical proofs.)
>
> If it is fully justified, MJ would have this advantage over IRV and
> MAM. I say this even though it still seems to me that, in any case,
> it would also be extremely unlikely that strategic voting would be
> successful when using IRV or MAM to elect one winner by millions of voters.
> 6.
>
> I still favor MJ even though it is theoretically vulnerable to
> ‘Later-no-harm’ (LNH). I am currently persuaded by B&L that this
> could occur in practice only in an election composed of voters and
> candidates whose numbers could be counted on one person’s hands. This
> is confirmed by Jameson in his 1st informative and recent EM posts: 1):
> ‘there is the failure of the later-no-harm (LNH) criterion. But note: MJ
> actually does pass a weaker version of LNH: rating additional candidates at
> above bottom will not harm the winner as long as those candidates are
> ranked below the winning median. My claim is that over time, the winning
> median grade will mostly fall in a given band of grades; for instance,
> using letter grades, between B- and D+. In that case, making distinctions
> between A and B at the top or D and F at the bottom are strategically
> safe.’
> 7.
>
> Similarly, I am aware that MJ fails the strong Independence of
> Irrelevant Alternative (IIA) candidates test. However, B&L explain
> how MJ does at least pass a weaker IIA test. As Jameson put in his 2nd
> recent and informative EM contribution: ‘Note that this strategy will
> almost certainly not affect the medians, and thus will not change the
> winner. Though technically it breaks IIA, it only does so in bizarre cases
> where both voter and candidate distributions differ severely from
> historical norms.’
>
> S: I would very much appreciate anyone explaining why you think any of the
> above views are mistaken. Please try to correct any mistakes you have
> noticed in the above.
>
> I look forward to hearing from you,
>
> Steve
>
>
>
> ----
> Election-Methods mailing list - see http://electorama.com/em for list info
>
>
>
> No virus found in this message.
> Checked by AVG - www.avg.com
> Version: 2016.0.7797 / Virus Database: 4656/13022 - Release Date: 09/15/16
>
>
>
> ----
> 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/20160917/f9bc2838/attachment-0001.htm>
More information about the Election-Methods
mailing list