[EM] serious strategy problem in Condorcet but not in IRV?

[Dave Ketchum]

I see topics worth commenting on:

If a method other than Plurality is to be given to/inflicted on Plurality
voters, there is MUCH need for adequate education as to need to get beyond
the minimal knowledge suitable for bullet voting, and how to express
yourself on a ballot.  This is NOT properly the business of parties or
candidates.

Similarly, parties have to get beyond "vote for us" and into who is the
worst enemy and why this party's candidates should get high rankings even
among voters unwilling to give top ranking.  Even candidates get in on
some of this.

Polling is a painful subject:
       Accurate polling is needed to help candidates decide whether current
strategy needs amending.
       Accurate polling can be desired by those who would use it 
destructively.
       INaccurate polling has been used in the past to trick voters into
voting against their own interests.
       Not clear to me whether forbidding polling immediately before
elections would help or hurt.

Dave Ketchum

On Thu, 21 Aug 2003 15:51:01 -0400 James Green-Armytage wrote:

 > Dear election methods fans,
 >
 > In the sort of strategy scenarios that I have been talking about, one
 > fairly positive possibility occurred to me. Take my ABC examples for
 > instance. (Let's assume that the method being used is beatpath Condorcet,
 > using defeat magnitudes rather than margins.)
 >
 > Example 1: Sincere preferences.
 > 46: A>B
 > 44: B>A
 > 5: C>A
 > 5: C>B
 >
 > Example 2: Sincere preferences.
 > 30: A>B
 > 25: B>A
 > 23: C>A
 > 22: C>B
 >
 > 	What I am hoping is that maybe candidates and parties in the position
 > that A and B are in will be moved to broker some sort of truce agreement
 > whereby neither party instructs their supporters to use a burying /
 > offensive order reversal strategy against the other party.
 > 	Or even a truce where the parties actively encourage their supporters to
 > rank the other fellow second rather than truncating, assuming that this is
 > their sincere preference.
 > 	Such a truce would make good sense if neither candidate knew exactly
 > which one of them would win out of the two of them, which is a pretty
 > common situation in real elections.
 > 	A truce would be especially in the interest of both candidates in example
 > #2, where they may know that neither of them can beat C without each
 > other's second choice votes.
 >
 > 	(Also, from a political perspective, candidates who engineered such a
 > truce would be better situated after the election even if they didn't win
 > than a candidate who had gone all out in a serious smear campaign /
 > burying strategy drive and then still lost.)
 >
 > 	Let me see if the use of strategy tables will clarify
 > 	Here is the strategy table for example #2 as I expressed the scenario in
 > previous postings, where all candidates have a precise notion of voters'
 > sincere preferences, especially concerning the result of the pairwise
 > competition between A and B.
 >
 > 		A buries	A truncates	A sincere
 > B buries	C wins	C wins	B wins		
 > B truncates	C wins	C wins	A wins
 > B sincere	A wins	A wins	A wins
 >
 > 	Here is a strategy table where the result of this pairwise competition is
 > uncertain. "AB equal" means that there is a roughly equal probability that
 > A or B will win the comparison.
 >
 > 		A buries	A truncates	A sincere
 > B buries	C wins	C wins	B wins
 > B truncates	C wins	C wins	AB equal
 > B sincere	A wins	AB equal	AB equal
 >
 > 	It looks to me like A and B have more incentive to cooperate with each
 > other in the second table. (If this is the case, then banning polls two
 > weeks before the election might not be a bad idea...) Is there anyone here
 > who is good with game theory?
 >
 > my best,
 > James
 >
 >
 >
 > P.S. Please let me know if I have made any calculation errors.
 > P.P.S. I wonder why none of the IRV advocates on the list aren't arguing
 > that the Condorcet strategy flaws are serious. I think that this is the
 > point on which Condorcet is most vulnerable, and it would help Condorcet
 > advocates sharp if they came up against the strongest available arguments
 > on the subject. I am basically a Condorcet advocate myself, so it is not
 > especially easy for me to be the one pressing for a harsh look at
 > strategic incentives under Condorcet...

-- 
davek at clarityconnect.com  http://www.clarityconnect.com/webpages3/davek
   Dave Ketchum   108 Halstead Ave, Owego, NY  13827-1708   607-687-5026
             Do to no one what you would not want done to you.
                   If you want peace, work for justice.