[EM] acceptable Chicken Proof methods (was chain climbing)

[Forest Simmons]

Kristopher,

you suggested ratings or grading, including MJ as possibilities.

But these cannot (at least in their present forms) be chicken proof.

For example MJ would give the win to B as long as the A faction gives B any
grade x less than or equal to the top grade but strictly greater than the
bottom grade:

40 C(100%)
34  A(100%), B(x)
26  B(100%)

So the B faction could defect with impunity.

Range does better since it would elect C for sufficiently small x and would
elect B for sufficiently large x, but it may fail to to elect A even when
the B faction cooperates:

40 C(100)
34 A(100), B(20)
26 B(100), A(20).

Also, if we go for explicit (as opposed to implicit) approval cutoffs, I
withdraw my support for TACC, since with explicit approval it fails the
plurality criterion:  In fact it elects A from the ballot set

40 C
35 A>>B
25 B


So now, my top choices are Chris Benham's Min(LV)(equal ranking
Whole)Margins in the case of no approval information, and
Condorcet(approval margins) or Majority Enhanced Approval when explicit
approval cutoffs are allowed.

Forest
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20140428/6cb668aa/attachment-0002.htm>