[EM] margin vs winning votes
C.Benham
cbenham at adam.com.au
Thu Jul 24 10:25:18 PDT 2014
On 7/24/2014 8:39 AM, Toby Pereira wrote:
> I'm a bit late to the whole winning votes / margins / losing votes
> thing, but my immediate thought here is how well losing votes would
> handle clones. If C is a clone of B but the less favoured in the
> party, in A v B, B will rarely, if ever, be ranked fully bottom,
> whereas the uncloned A could be. So would this favour B in a cycle?
Toby,
No.
It sounds like you have misunderstood Losing Votes (equal-ranking
whole) in some way.
If you can give a specific example election (set of ballots) to
illustrate your concern, I'll respond in more detail.
Chris
>
> *From:* C.Benham <cbenham at adam.com.au>
> *To:* election-methods at lists.electorama.com
> *Sent:* Wednesday, 23 July 2014, 20:00
> *Subject:* Re: [EM] margin vs winning votes
>
>
>
> BTW, I don't like either of Winning Votes or Margins. Of the
> alternatives that are equivalent to both when all the voters give a
> full strict ranking I think the best is Losing Votes (erw). Ballots
> that equal-rank candidates A and B both above at least one other
> candidate are counted in the pairwise contest between A and B as
> giving
> a whole vote to each. Ballots that rank A and B equal-bottom
> (or truncate A and B) count as zero to both in their pairwise
> contest.
> The method considers that the greater the number of votes
> on the losing side, the weaker the pairwise defeat.
>
> And not limiting ourselves to alternatives that are equivalent to
> Margins and Winning Votes (and Losing Votes) when all the voters
> give a full strict ranking, I think better still is "MinMax Losing
> Votes (erw) Margins", which I first suggested in April this year.
>
> http://lists.electorama.com/pipermail/election-methods-electorama.com/2014-April/097935.html
>
> Chris Benham
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20140725/38966056/attachment.htm>
More information about the Election-Methods
mailing list