[EM] So I got an email... / IIA

[Kristofer Munsterhjelm]

On 11.04.2022 06:34, Forest Simmons wrote:
> As I remember, Maskin's Sci. Am. article advocated Copeland completed
> with Borda.
> 
> 
> Simpler and no worse would be ...
> 
> Elect the CW if there is one, else instant pairwise runoff between top
> two first place candidates.  Precinct summable ... monotonic ... more
> clone free than Copeland or Borda. 

That's not monotone, is it? Take the IRV example from
https://www.rangevoting.org/Monotone.html:

6: C>A>B
2: B>A>C
3: B>C>A
4: A>B>C
2: A>C>B

The Plurality order is C=A>B. C beats A pairwise and wins.

Let two ACB voters raise C:

8: C>A>B
2: B>A>C
3: B>C>A
4: A>B>C

Now the Plurality order is C>B>A. B beats C pairwise and wins.

My modification would be this, which happens to pass single-candidate DMTBR:

Elect the CW if there is one, otherwise if at least two candidates have
more than 1/3 first preferences, elect the pairwise winner of the two,
otherwise elect the Plurality winner.

That might be a little complicated, though, and I'm not sure it's
different enough to ordinary Plurality to make a difference (I could
easily see "three party rule" forming due to the 1/3 qualifier).

Kevin Venzke's rule that also passes single-candidate DMTBR might be
more well-behaved yet more complex still:

Elect the CW if there is one, otherwise each candidate's score is the
sum of that candidate's first preferences plus the first preferences of
every candidate he beats. Elect the candidate with the highest score.

-km