<DIV>I was hurried on that example. Here's a better one:</DIV>
<DIV> </DIV>
<DIV>35 A>X>Y>>B</DIV>
<DIV>15 X>Y>>A=B</DIV>
<DIV>35 B>Y>X>>A</DIV>
<DIV>15 Y>X>>A=B</DIV>
<DIV> </DIV>
<DIV>Note that both X and Y have 100 percent approval, while neither A nor B has more than 35 percent approval, yet in a two winner election STV elects {A,B}.<BR><BR>----- Original Message -----<BR>From: <FSIMMONS@PCC.EDU><BR>Date: Friday, June 10, 2011 2:31 pm<BR>Subject: MRSODA (Mr. Soda), a SODA-inspired PR method (NP-complete???)<BR>To: election-methods@lists.electorama.com,<BR><BR>> I second the idea. Any PR method based on approval ballots <BR>> could be implemented in this way. <BR>> <BR>> The main advantage of Range or Approval based methods for PR <BR>> when compared to STV is shown in the <BR>> following two winner example:<BR>> <BR>> 30 A(100)>X(90)>><BR>> 20 B(100)>X(90)>><BR>> 30 C(100)>Y(90)>><BR>> 20 D(100)>Y(90)>><BR>> <BR>> with utilities in parentheses.<BR>> <BR>> STV picks {A, B} as the winning set.<BR>> <BR>> Any RRV, PAV, etc. pick the set {X, Y}.<BR>> <BR>> The extra simplicity of approval style ballots in SODA is the <BR>> icing on the cake.</DIV>