I'd like to find an analysis of how a ranked choice voting system has
played out in a real election, preferably with well more than 3
candidates.  Perhaps something from a political science treatise or
publication, perehaps from a news analysis post election.  The
purpose:  San Francisco district supervisor elections in November 2004
are using a new ranked-choice voting system that some claim has never
been used before.  I would like to find previous examples that include
some analysis of how it actually turned out.  I don't care if the
election is in the US or not.  I don't care if the method is exactly
the same as the one being used in San Francisco.  The important
elements are that voters can select more than one candidate, that
there is some difference between how first and second (and third and
fourth?) choices are ranked, and that there were preferably a large
number of candidates in the election.

A brief synopsis of the method being used in San Francisco:  The
candidate who receives the fewest first-choice votes is eliminated and
his second-choice votes are added to the appropriate candidate.  This
process is repeated until some candidate has more than 50% of the
votes cast.

I believe discover magazine has an article on this including about
john mccain in the 2000 elections

Sorry, erick2, that article is just about a standard election and why
a popular candidate might not win it.  I'm talking about a very
unusual type of election where the losers votes get allocated up the
chain until eventually somebody has >50% of the votes.

Ranked Choice Voting or Intant Runoff Voting (IRV) has been used in
the past.  Very recently the exact methodology SF is planning to use
was used in a Utah GOP primary to pick the candidate for an open race.

http://www.fairvote.org/irv/utahindex.html


But there wasn't a complete analysis of this election.  However the
Illinois assembly on political representation did complete an
analysis:

http://www.igpa.uiuc.edu/publications/specPubs/ExeSumry.pdf

Hope this helps

At http://www.demochoice.org you can find results from several real
ranked-choice elections.  These include student government elections
from Stanford and UC-Davis, as well as multi-winner elections from
Cambridge, MA and a few districts in Ireland now that those places
have electronic voting and the ballot files are available.  It is
possible to re-tally these ballot sets with different numbers of
winners, with certain candidates excluded, or with restrictions on the
number of rankings; send email to the address on the DemoChoice site
regarding that.  If you are interested in the context of the Cambridge
elections, you can find news articles by googling, or rwinters.com may
help.  Context for Stanford elections can be found if you look up the
Stanford Daily.

I have posted an analysis that may answer some questions about this on
the Cal IRV list (http://groups.yahoo.com/group/instantrunoffca) - see
messages 1582, 1585, and 1586.  Also, I recommend the new San
Francisco ranked-choice web poll:

http://www.demochoice.org/sf.html