> You continued:
>
> (Also it looks like you have
> some other method in mind
>
> [endquote]
>
> How so? As I said, I'm referring to MDDTR.
You replied:
Because in the description of your example you referred to information
that MDDTR
ignores:
> Say the method is MDDTR, and your favorite candidate is F. F doesn't
> have a winning approval (top + middle) score,
> because x has significantly more approvals.
[endquote] Ok yes, MDDTR doesn't count top + middle ratings. MDDTR counts top
ratings. And, of course, contrary to what I said elsewhere in my posting,MDDTR
elects the undisqualified candidate with the most top ratings, anddoesn't look
at whether someone gets top rating from a majority. But those errors of mine
don't affect my Mono-Add-Plump scenario. It's possiblefor your plump vote to
take away x's majority defeat disqualification. Andif x, now undisqualified,
has more top ratings than your favorite, thens/he might take the win from your
favorite. How could x have moretop ratings than your favorite, and still have a
majority defeat before you arrive tovote? Voters' don't agree on whom to
top-rate, and no candidate has very many top ratings. A relatively large number
of people have middle or top rated a few other candidates,but not x, making a
bare majority pairwise defeat against x, which ceases to exist when you
arrivewithout voting those other candidates over x. _That_ is my MDDTR
Mono-Add-Plump scenario. But my point was that the difference betweenmy
scenario, in which the newly-undisqualified candidate wins by beating your
favorite bysome higher vote-total (top-ratings total); and your scenario in
which s/he beats your favorite bybeing the only undisqualified candidate, is an
insignificant difference. Either way, you made your favoritelose by voting that
you're indifferent between everyone other than your favorite.
> But I'll post an example of that particular kind of Mono-Add-Plump
> failure within
> the next few days.
You replied:
I look forward to seeing it.
[endquote] Why?? As I said, I don't assert that MDDTR fails Mono-Add-Plump; I
admit it. And I tell whyit isn't a genuine fault or wrong result. So, without
knowing if I'm correct to do so, I'll concedethat MDDTR can't fail
Mono-Add-Plump in any way other than the way in your example. By the way, I
don't say that you're wrong, to value the criteria that you value. It's just
thatI value only criteria that have signfificant practical importance. To me,
this subject isn't an aesthetic art or aesthetic game. It'sabout practical
improvement, solving serious practical problems of voting. With completely
different goals and purposes, it's hardly surprising that we use different
criteria, leading to preference for different methods. I fully admit that
MDDTR, MDDA and MDD,ABucklin can look unaesthetic from the Plurality-accustomed
point of view.Likewise MMPO and every method that meets Condorcet's Criterion
(And no, I'm not saying that MMPO meets Condorcet's Criterion). So, MDDTR,
MDDA, MDD,Bucklin and MMPO could be "controversial" because they differ too
much from Plurality. Such objections canbe disposed of, as I have done in my
previous postings. But nevertheless, maybe one woudln't want to have to give
opponents theopportunity to use such distractions. For one thing, opponents
have more money to make their arguments than proponents haveto answer the
arguments. So maybe it would be better to not propose potentially controversial
methods. That's why, when proposing voting reform to people lately, I've been
proposing Approval, MCA, & MTA; and not MDDTRor MMPO. But I don't give up on
MDDTR and MMPO. I merely would offer them when there's more time to discuss
their differences fromPlurality. It would be necessary to establish that people
consider their FBC, LNHa and CD compliance to be more important thantheir
non-resemblence to Plurality, before making a public proposal about them.
People would have to be willing to genuinelybreak with the Plurality voting
system and its results. The Approval bad-example is a genuine problem, and one
that can andshould be avoided if we're going to use a rank method (in which
category I include 3-slot methods). Mike Ossipoff
----
Election-Methods mailing list - see http://electorama.com/em for list info
