[EM] Re: Condorcet strategy

[Chris Benham]

CF,
You  wrote (Fri.May 28):

>As far as finding a time share goes, here's what I had in mind: Divide each
>term into sub-terms of equal length (say, six months).  For X and Y, if X
>wins by Schulze, and Y is the candidate who, apart from X, loses the fewest
>comparisons of the strongest beatpaths, X and Y are the winners of the
>election.  For X and Y, if X and Y are winners, use the strength of the
>beatpath from X to Y and the strength of the beatpath from Y to X as input,
>and use the Sainte-Lague formula to allocate the sub-terms.
>
Surely sharing up the time between the top-cycle candidates is the same 
sort of problem as sharing voting power
between proxies, nothing complicated. Each ballot contributes one vote 
to the single top-cycle candidate it ranks
above the others, or equal fractions of a vote that sum to 1 to each of 
the top-cycle candidates it votes in equal
place above the others.  Then the time in office for each of these 
top-cycle  (or Shwartz set) candidates is allocated
in proportion to these tallies.

Chris  Benham