[EM] Comparison summary

[MIKE OSSIPOFF]

For the purpose of my critreria, I have a definition of sincere voting. It 
doesn't involve ratings. Here I'm using "sincere" with a different meaning. 
For the purposes of this message, sincere voting means voting that expresses 
the voter's true ratings to the extent possible with CR balloting or 
Approval balloting.

Because Approval is a CR method, "sincere voting" means the same in Approval 
as in CR. There of course are just two point assignments in Approval, and a 
sincere Approval ballot is considered to be one that gives a point to every 
above-mean candidate.

By "Approval strategy" I mean Approval's expectation-maximizing strategy, 
whether used in Approval or CR.

It's important to emphasize that sincere CR voting is possible in Approval 
too, by voting for the above-mean candidates, and maybe flipping a coin when 
undecided on a candidate; or by assigning ratings from 0 to 1, and drawing 
numbers from a bag, as described in my previous posting.

1.Some voters votng sincere CR ratings  vs all using  Approval strategy

  A. When there's no CW, or the CW (as rarely happens) doesn't maximize SU, 
or Approval strategy
      is based on mistaken information:

     The fact that some CR voters rate sincerely might increase the SU of 
the outcome. If there's
     a CW, and Approval strategy is based on false information, then sincere 
rating by some voters
     could help the CW win where s/he othewise wouldn't.


  B.. When there's a CW who (as is usually the case) maximizes SU,. and 
Approval strategy is based
       on accurate information:

      The sincere CR voters could be had by strategizing voters, defeating 
the CW and reducing the
      SU of the outcome.


As I was saying, in our current political system, progressives are basing 
their strategy on information that's unreliable, and probably false. So, for 
now, CR is better than Approval, since it encourages the sincere  rating 
that can save a CW from misinformed strategy.

But, after one Approval  or CR election, that election's result could inform 
future strategy so that Approval, which encourages strategy more, would be 
preferable. Better that everyone use well-informed strategy than that some 
vote sincerely and let the CW lose to strategists.

If, in the 1st Approval or CR election, Nader outpolls the Republican, then 
progressives should notice that they needn't vote for the Democrat.

2. Some voters using sincere CR stratgegy vs Condorcet:

  A. When there's no CW, or the CW (as rarely happens) doesn't maximize SU:

     Then, sincere rating by some CR voters could result in a better SU than 
that given by Condorcet.


  B.. When there's a CW who (as is usually the case) maximizes SU:

      Then, Condorcet will maximize SU when it elects the CW, and SU would 
be less if some
      sincere voting in CR elected someone else, by letting the sincere 
voters be had by
      strategists.

  I emphasize that these advantages & disadvantages are insignificant 
compared to the consideration of what is more likelly to be accepted by the 
public and enacted.

Smith's CR polling gives information that we should have gotten a long time 
ago. Now someone must do polling about which method people would be more 
likely to accept: Approval, CR, or Condorcet.

For that purpose, the Condorcet should be PC. Maybe, so as not to ask people 
for more time than they're willing to give, it would be better to ask each 
person "Yes or No" on their acceptance of one method, rather than defining 
all 3 methods for each person and asking them for an opinion on each one. 
But that could depend on where and whom one is polling.

Mike Ossipoff


.

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today - it's FREE! 
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/