[EM] Approval Margins example

[C.Benham]

Forest,

Originally I conceived Approval Margins as using ballots with an 
approval  cutoff/threshold, but since then I've largely gone off
allowing voters to rank among unapproved candidates (and the somewhat 
more complicated ballot).

But I agree that with that type of ballot Approval Margins is  a good 
method.

Still, in the example the chicken dilemma is to some extent revived. 
Maybe the A voters can be scared into approving B, in which
case maybe the B voters can again get their favourite elected by 
"defecting".  Arguably it's better for the method to have a kind
of  (Strangegloveian)  "doomsday machine"  defection deterrence.

Chris Benham


On 4/30/2014 6:04 AM, Forest Simmons wrote:
> Chris,
>
> just change 35 A>B to 35 A>>B, and C wins, unless the B faction gets 
> wise and changes to 25 B>>A or 25 B>A, making A the ballot CW.
>
> The explicit approvals are C40>A35>B25, and since pairwise C beats A 
> beats B,  Condorect(approval) whether total approval or approval 
> margins, elects C.
>
> MEA and Smith//Approval also elect C.
>
> Your instincts were right to begin with!
>
> Forest
>
>
>
>     Date: Tue, 29 Apr 2014 15:53:05 +0930
>     From: "C.Benham" <cbenham at adam.com.au <mailto:cbenham at adam.com.au>>
>     To: em <election-methods at electorama.com
>     <mailto:election-methods at electorama.com>>
>     Subject: [EM] Correction: Approval Margins fails Chicken Dilemma
>     Message-ID: <535F4549.8000908 at adam.com.au
>     <mailto:535F4549.8000908 at adam.com.au>>
>     Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
>
>     Oops!  I've just discovered that my recent claim that Approval Margins
>     (and Approval Margins Sort) meets the CD criterion is wrong.   Sorry.
>
>     35 A>B
>     25 B
>     40 C
>
>     B>C>A>B   Approvals: B60 > C40 > A35   (Approval Margins Sort
>     elects B)
>
>     Approval Margins: A>B -25,   B>C +20,  C>A +5.  B's defeat is the
>     weakest so B wins.
>
>     The combination of CD and Plurality  says that C must win.
>
>
>
>
>
> ----
> Election-Methods mailing list - see http://electorama.com/em for list info

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20140515/249a95a8/attachment.htm>