[EM] Markus: RP & BeatpathWinner/CSSD

[Markus Schulze]

Dear Mike,

you wrote (9 March 2003):
> I'd propose RP as a public proposal, due to its briefer
> definition. And I'd offer BeatpathWinner to organizations
> & committees, due to its elegantly simple & brief algorithm
> & computer program. I promote BeatpathWinner/CSSD & RP.
> BeatpathWinner/CSSD to organizations & committees, and RP
> for public proposals.

I wrote (9 March 2003):
> Do I interpret you correctly?:
>
> You consider the beat path method to be better than Ranked Pairs
> because of the "elegantly simple & brief algorithm & computer
> program" of the beat path method. But you believe that the
> average committee member is significantly more intelligent
> than the average voter so that the beat path method could be
> too difficult to understand for the average voter. Therefore,
> you suggest that Ranked Pairs should be proposed for public
> elections because of its briefer definition and despite of
> its disadvantages (i.e. not having an "elegantly simple &
> brief algorithm & computer program").

You wrote (10 March 2003):
> Did I say that BeatpathWinner is better than RP? No.
> I said that its simpler & briefer algorithm & computer
> program make it a better proposal to committees &
> organizations.
>
> I also said that there's no significant merit difference
> between BeatpathWinner/CSSD & RP.
>
> I didn't say anything about committee-members being
> more intelligent than voters, but I also won't deny
> the possibility that that could be true.
>
> But yes, there seems a good chance that the members
> of committees & organizations who are involved in the
> choice of a voting system are more likely to understand
> BeatpathWinner's longer definition than voters are.
> Aside from that, the members of committees & organizations
> seem more likely to tolerate longer definitions than the
> average voter would be.
>
> If you want to know who says that the voters are stupid,
> it's the IRVists, who believe that IRV is the only
> thing that people will understand.
>
> Isn't that pretty much what I said?
>
> Likewise, I suggested that BeatpathWinner/CSSD seems
> a more practical proposal to committees & organizations
> despite its disadvantage of not having as brief a
> definition as RP.
>
> You could say that both methods have an acceptance
> disadvantage that the other doesn't have. I prefer to
> compare their advantages. But it could be worded either
> way of course.
>
> I don't understand what your point is in that last
> sentence that I quoted above. Is it that you believe
> that RP has as "elegantly simple & brief algorithm &
> computer program" as BeatpathWinner does?
>
> Or is it that you believe that such an advantage
> couldn't make BeatpathWinner a more practical proposal
> to committees & organizations?

So you say that there is no significant merit difference
between Ranked Pairs and the beat path method, that its
"elegantly simple & brief algorithm & computer program"
make the beat path method a better proposal for committees,
and that its brief definition makes Ranked Pairs a better
proposal for the public.

However, according to Steve Eppley, there is a merit
difference. Steve, who uses the term "MAM" for Ranked Pairs
and the term "PathWinner" for the beat path method, writes:

> MAM may be preferable to PathWinner for a couple of reasons:
>
> 1. MAM (but not PathWinner) satisfies immunity from majority
> complaints (IMC), immunity from second-place complaints (I2C)
> and other criteria described in the document Immunity from
> Majority Complaints.
>
> 2. Computer simulations using randomly generated profiles of
> voters' orderings suggest the alternative chosen by MAM will
> beat pairwise the alternative chosen by PathWinner more often
> than vice versa, and that over the long run more voters will
> prefer MAM winners over PathWinner winners than vice versa.

For more details:

http://www.alumni.caltech.edu/~seppley/Strategic%20Indifference.htm
http://www.alumni.caltech.edu/~seppley/Immunity%20from%20Majority%20Complaints.htm
http://www.alumni.caltech.edu/~seppley/Comparison%20of%20MAM%20and%20PathWinner.htm

In so far as Steve sees a merit difference while you see
no significant merit difference, I conclude (a) that you
consider his arguments insignificant or (b) that you have
arguments for the beat path method against Ranked Pairs
that compensate Steve's arguments for Ranked Pairs against
the beat path method. Which of these two conclusions is
correct? Please explain!

Markus Schulze