> Message: 1
> Date: Sat, 13 Nov 2010 22:17:49 +0200
> From: Juho <[email protected]>
> To: Election Methods <[email protected]>
> Subject: Re: [EM] breakdown of Oakland mayor ballots
> Message-ID: <[email protected]>
> Content-Type: text/plain; charset="us-ascii"; Format="flowed";
>        DelSp="yes"
>
> - I decoded the ballot images few days ago since there was some
> interest on the rangevoting list => 
> http://groups.yahoo.com/group/RangeVoting/message/14474
>
> - Warren Smith wrote down some notes on the results => 
> http://groups.yahoo.com/group/RangeVoting/message/14483

Looks like IRV lucked out and dodged a bullet for a change when it
differed from plurality for the first time in San Francisco and got it
right.  As Warren Smith points out, if a little over 3,100 more voters
had ranked Kaplan 1st and Quan 2nd, Perata dead last, then Perata
would have won instead.    We won't have to wait long until IRV gets
it wrong when it differs from plurality voting, according to the odds
Warren calculates.

http://rangevoting.org/Oakland2010Mayor.html

explains. Juho's data processing appears to be correct.
Quan was the Condorcet and IRV winner.
Perata was the plain plruality winner.

PARADOX:
If 3135 extra Kaplan>Quan>...>Perata votes are
added, all ranking Perata dead last and Quan 2nd, that
causes Perata to (then) win.

It is fortunate for them that those Perata-hating voters stayed home!

--------
IRV is is hardly worth all the extra costs and complexity and lack of
auditability when the chance is so high of not getting a good outcome
like this whenever the result differs from plurality.

Kathy Dopp
http://electionmathematics.org
Town of Colonie, NY 12304
"One of the best ways to keep any conversation civil is to support the
discussion with true facts."

Fundamentals of Verifiable Elections
http://kathydopp.com/wordpress/?p=174

Realities Mar Instant Runoff Voting
http://electionmathematics.org/ucvAnalysis/US/RCV-IRV/InstantRunoffVotingFlaws.pdf

View some of my research on my SSRN Author page:
http://ssrn.com/author=1451051
----
Election-Methods mailing list - see http://electorama.com/em for list info

Reply via email to