Kathy Dopp wrote:
Notice how typically, Fair Vote claims that they found majority
winners by manipulating the definition of majority to mean only those
voters left standing by the final counting round.
I wonder whether, if one were to make a "maximally wrong" IRV-type
method that eliminated the candidate most people voted for until only
two were left (and then picked the one who beat the other), FV would
still claim the "winner" to have been elected by a majority.
E.g
30: A > D > C > B
20: D > C > A > B
20: C > A > D > B
23: B > A > C > D
Plurality counts: 30: A, 20: D, 20: C, 23: B
Eliminate A.
50: D > C > B
20: C > D > B
23: B > C > D
Plurality counts: 50: D, 20: C, 23: B.
Eliminate D.
70: C > B
23: B > C
C wins by 70/93 = 75.3% of the votes. What a landslide!
(Schulze and MAM gives A > D > C > B, and IRV gives A > B > C = D.)
-
Less facetiously, if all elimination methods have this property, then so
does BTR-IRV or Borda-elimination. Either of the two aforementioned
methods may be better than IRV itself - I would say "would" but if I
remember correctly, Warren claimed Nanson and Baldwin are
extraordinarily easy to manipulate.
----
Election-Methods mailing list - see http://electorama.com/em for list info
