[EM] James doesn't think people will favorite-bury in Condorcet

James Green-Armytage jarmyta at antioch-college.edu
Wed Jun 1 21:30:01 PDT 2005


James replying to Mike...

Mike:
>
>I shouldn't start repeating things for you again, but I observed that 
>behavior with someone voting in a BeatpathWinner election. Though I said 
>that I wouldn't favorite-bury with that method, and it isn't really 
>necessary, the person did it anyway, ranking all of the Democrats over
>their 
>favorite. But I didn't assure that person that there couldn't possibly be 
>any gain from favorite-burial, because that guarantee couldn't be made
>for 
>BeatpathWinner.

	I answered this part of Mike's argument as well, in the earlier message.
In short, I don't think that sweeping conclusions about voter strategy
should be made based on the behavior of one person in an internet poll. In
general, I don't think that voting behavior in internet polls is a good
indicator of voting behavior in serious elections, because in internet
polls: (1) the outcome is totally unimportant, and (2) the electorate is
more technically inclined than average.
>
James:
>, capable of
>outweighing concerns such as Condorcet, Smith, MMC, CL, etc.
Mike:
>You forgot to explain why you think that compliance with Condorcet,
>Smith, 
>MMC, or CL will reassure people to not favorite-bury better than FBC
>will. 

	Of course that wasn't my point. My point is that when comparing
beatpath(wv) to minmax(po) there is a tradeoff between the above criteria
and FBC, and if the benefit of FBC is not very large, the above criteria
should take precedence, and beatpath(wv) should be considered to dominate
minmax(po) from a merit standpoint.

Sincerely,
James





More information about the Election-Methods mailing list