[EM] MDD,ER-Bucklin (whole)

Chris Benham chrisbenham at bigpond.com
Sun Oct 23 10:01:32 PDT 2005


Kevin,
A  while ago I suggested modifying  ER-Bucklin(whole)  so that after the 
round on which one or more candidates
reach the majority threshold, those candidates that haven't reached it 
are eliminated and then any ballots
that haven't yet contributed a vote to any of the remaining candidates 
now do so if  they make any distinction
between them (giving a whole vote to their most preferred, or a whole 
vote to each of their equal-most preferred).
After that elect the candidate with the highest tally.

I am now sure that this fails FBC, and so isn't good. 
Maybe  voting  Favourite equal to Compromise could cause Favourite to 
make the threshold along with Compromise
and Worst, and then maybe Worst wins because Favourite attracts some 
votes that would have gone to Compromise.

Also, changing the subject a bit, it now seems to me that FBC is not 
just incompatible with Condorcet but also with
Dominant Mutual Third.  (Do you agree?)

10: A>B
39: B
03: B=C
48: C

The DMT winner is B,  but  ER-Bucklin (whole)  elects C  on  the first 
round (with or without MDD).


Chris  Benham





>  
>



More information about the Election-Methods mailing list