[EM] Modification of STV to guard against favourite betrayal

[Jonathan Lundell]

On 9 Sep 2013, at 7:37 AM, Peter Zbornik <pzbornik at gmail.com> wrote:
> There was no strategy voting as far as I know.
> There was an election to the party list (primaries), using essentially Top-down STV (see article: http://www.votingmatters.org.uk/issue9/p5.htm), where a less preferred candidate got a higher place on the list, instead of getting a more preferred candidate a lower place on the list, which is a well-known weakness of this method.
> 

I see. 

If you have an independent means of distinguishing a "more preferred" from a "less preferred" candidate, why not use that directly?

Offhand, if you're using the mechanism you refer to, I think you're better off using the bottom-up approach, in which you don't have to "protect" already-elected candidates, the source of considerable distortion. It may require more counts, depending on how many candidates you have, but OTOH the earlier counts needn't be carried to completion; only to the point where you have the desired number of candidates left standing.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20130909/e9ef9dcc/attachment-0004.htm>