[EM] Remember Toby

fsimmons at pcc.edu fsimmons at pcc.edu
Fri Jun 3 11:34:36 PDT 2011


Dear Kathy,

Great idea!  Why didn't I think of that?  As you indicated, it could be done with an ordinary Plurality 
ballot, and would amount to a simplified version of DYN:

If a voter "bullet" votes, then the ballot is interpreted as submission of a replicate of the Approval ballot 
that the marked candidate ends up submitting. Otherwise it is interpreted as an ordinary Approval ballot.  
The candidate with the most approval wins.

To rule out the smoke-filled-room scenario, candidates' Approval ballots must be consistent with their 
rankings of the candidates, which they are required to publish at least a three (?) days before the voting 
takes place.

This method is indeed uniformly better than Plurality, Asset, and Approval!

What should we call it?

Can anybody think of a better deterministic voting method for a single winner public proposal?

Forest

>Kathy Dopp kathy.dopp at gmail.com responded to

>> Date: Thu, 02 Jun 2011 19:14:12 +0000 (GMT)
>> From: fsimmons at pcc.edu
> >To: election-methods at lists.electorama.com

> I>s DYN too complicated?  If so, we are stuck with ordinary Approval or ordinary Asset Voting.  They 
are
> >the only choices simpler than DYN that dominate Plurality.

>Great analysis. I like the "uniformly better" concept. I agree with
everything you said - very logical.

>Question: Wouldn't there be a third option that is uniformly better
than plurality - an Approval/Asset hybrid?  I.e. allow voters to rank
only one candidate - then if that candidate loses, those voters who
rank only one candidate transfer the right to their losing candidate
to cast their vote for another candidate. Or if voters choose more
than one candidate, their own extra approval votes are counted.

>Just a thought that might alleviate the problem that some Judges have
found with some electoral methods on the basis of voters having
unequal amount of votes if voters may choose either two candidates or
one candidate and their candidate gets to cast their other vote, but I
haven't really thought it through completely. It seems like in this
case, an initial loser could end up the winner if enough of the asset
votes were cast for that person.  But it's probably not a good idea -
just a passing thought.





More information about the Election-Methods mailing list