On Nov 16, 2010, at 12:03 AM, Kevin Venzke wrote:
Hi Forest,
--- En date de : Lun 15.11.10, [email protected] <[email protected]> a
écrit :
De: [email protected] <[email protected]>
Objet: Re: [EM] Why I Think Sincere Cycles are Extremely Unlikely in
À: [email protected]
Date: Lundi 15 novembre 2010, 15h38
From: Juho
[snip].
On the other hand we know that all Condorcet methods
are
vulnerable at
least to the burying strategy.
All Condorcet Methods? Or all deterministic Condorcet
Methods?
I would say all... LNHelp is basically a subset of a "burial
resistance"
criterion and you can't get a Condorcet method to satisfy LNHelp even
using randomness.
You may come up with something that in practice is very difficult to
try
to game (and we do try), but there's no invulnerability.
Kevin
My thoughts go to this example.
1: A>B>C
1: B>C>A
1: C>A>B
This is a three-way tie for all methods that satisfy some very basic
criteria.
The first voter is the only strategic voter. Her sincere vote would be
A>C>B. C is the sincere Condorcet winner. The first voter may thus
improve the expected outcome by burying C under B if she prefers
random choice between A, B and C to electing C.
Juho
----
Election-Methods mailing list - see http://electorama.com/em for list info
