[EM] Post-count options help voting-systems, especially rank methods.

[Michael Ossipoff]

These options have mostly already been discussed at EM, but we tend to
forget how helpful they can be.

Steve Eppley proposed proposed the Candidate-Withdrawal-Option, whereby,
after the count, any candidate(s) can withdraw from the election, and call
for a new count, with the original ballots.

Of course the candidates could negotiate or discuss, among themselves, or
with their supporters, regarding which candidate(s) will withdraw.

Candidate-Withdrawal-Option would benefit the ideal majoritarian methods
(MAM and Beatpath), and any other pairwise-count methods, or IRV. IRV's
worst problems are pretty-much eliminated by Candidate-Withdrawal-Option.

In one poll that I conducted at EM, I stipulated that, after the results
are known, voters could have the option to truncate their rankings. That's
a protection against anti-CW offensive strategy, in wv Condorcet methods.

Of course, in an election by secret-ballot, that could only be achieved by
holding a 2nd balloting in the event that there's a top-cycle consisting of
all majority defeats, a sign of possible successful anti-CW strategy.

But if it's desired to also allow voter to penalize chicken-dilemma
defection (to which the ideal majoriarian methods are vulnerable), then the
2nd balloting could be triggered by _any_ top-cycle, instead of only by a
majority top-cycle.

But truncation isn't the only counterstrategy for anti-CW offensive
strategy in wv Condorcet methods. Additionally, there's the counterstrategy
of equal-ranking, to protect a perceived CW. That, too, could be allowed
after the count, or could be done by voters in that automatically-triggered
2nd balloting.

A 2nd balloting could be automaticlly routinely conducted even in Approval,
if it were felt needed to allow voters to penalize chicken-dilemma
defection. I don't suppose that there's any indicatation that should
trigger that 2nd balloting, and so, given the perceived need for it, it
should be routine.

If the ballot-modification were allowed, or the re-voting were
conditionally-triggered, and it were desired to combine that with the
Candidate-Withdrawal-Option, then that seems feasible--The
ballot-modification opportunity would be offered after any
candidate-withdrawals. And, in a system where a top-cycle or a majority
top-cycle triggers a re-vote, that would of course happen only if the
top-cycle or majority top-cycle remains after the candidae withdrawal(s).

Michael Ossipoff
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20140124/0ec636a4/attachment-0003.htm>