On Dec 16, 2011, at 6:16 PM, Ted Stern wrote:
On 16 Dec 2011 13:29:30 -0800, David L. Wetzell wrote:
---------- Forwarded message ----------
From: Kathy Dopp <kathy.d...@gmail.com>
To: election-methods@lists.electorama.com
Cc:
Date: Fri, 16 Dec 2011 09:11:11 -0500
Subject: Re: [EM] Egg or Chicken.
Date: Wed, 14 Dec 2011 14:59:14 -0600
From: David L Wetzell <wetze...@gmail.com>
if we push hard for the use of American Proportional Representation
it'll give third parties a better chance to win seats and they will
prove great labs for experimentation with electoral reform.
This is also a good reason to strategically support IRV, since we
can trust that with changes, there'll be more scope for
experimentation and consideration of multiple alternatives to FPTP.
This is precisely the kind of game theory that leads to the two party
problem with FPTP: we need to coalesce behind the strongest contender
in order to have some kind of voice, be it only a compromise. So no,
I don't think it is a good reason.
While IRV offers ranked choice voting - a big improvement over FPTP,
It fails to have a defendable way to count the votes - and, by that
incompleteness, can reject the true choice of a majority of voters -
see Burlington as a widely heard example. See Condorcet, a method
that is a good reason for dumping IRV - by accepting the same votes as
IRV, but then actually reading what the voters vote, Condorcet is a
major improvement.
KD: Actually, if we support the adoption of proportional
representation, it is a good reason to strongly oppose IRV and STV
which will sour the public on any notions of changing US electoral
systems for decades and greatly hinder any progress towards
proportional systems.
dlw: That is what is in dispute.
PR makes sense for legislatures - but is no help for electing such as
governors or mayors.
KD:We've already seen this occur in jurisdictions where IRV has been
tried and rejected when it was noticed how overly complex,
transparency eviscerating, and fundamentally unfair IRV methods are.
Right now there is a push to get rid of it in San Franscisco. IRV
was tried decades ago in NYC and stopped progress there for decades.
dlw: Unfair? Why because it emulates the workings of a caucus by
considering only one vote per voter at a time?
Yes, precisely. The traditional Robert's Rules method of taking only
a single vote at a time is at fault. It produces a suboptimal result
by segmenting the problem too much.
IRV does allow the voters to make a complete statement of their
desires, with no segmentation, which means no information from other
voters (as would happen in a caucus) as to what the other voters are
doing in what is called above "a single vote".
IRV does segment the vote counters' work by restricting their reading
of each ballot to what is, for the moment, the top rank.
It is similar to the less optimal result you get from dividing space
by partitioning in each dimension separately to get bricks, instead of
hexagons in 2D or truncated octagons in 3D.
dlw: If a 2-stage approach is used then it's less complex and the
results can be tabulated at the precinct level.
Could he be thinking of Condorcet, which tabulates the same ballots
intelligently at precinct level?
dlw: I'm sure the Cold War red scare stopped progress in NYC and
elsewhere a lot more than "IRV"....
KD: IRV/STV methods introduce problems plurality does not have and
do not solve any of plurality's problems, so it's a great way to
convince people not to implement any new electoral method and show
people how deviously dishonest the proponents of alternative
electoral methods can be. (Fair Vote lied to people by convincing
them that IRV finds majority winners and solves the spoiler problem,
would save money, and on and on...)
dlw: It's called marketing. FairVote wisely simplified the benefits
of IRV. IRV does find majority winners a lot more often than FPTP
and it reduces the spoiler problem considerably. It does save money
compared with a two round approach and its' "problems" are easy to
fix.
But when marketers lie and get caught, potential customers get
suspicious as to future marketing.
I do not understand the above claim about majority winners - true that
FPTP voters cannot completely express their desires, but the counters
can, accurately, read what they say with their votes.
Dave Ketchum
That is debatable. I happen to think that the goal/object of IRV is
different from what one wants to achieve in a single winner election.
If you model your government on a natural system (and the US Founders
based their arguments by appealing to "Natural Law"), then you do best
when you create a diverse and representational set of options (hence
PR for legislatures) and only then apply selective pressure using a
centrist single winner method.
IRV is not based on centrism. As the single-winner limit of STV, it
is better (not "best") at finding a representative of the majority,
not the best representative of the entire population.
As for STV, one can keep patching to deal with its many problems, but
at its core it also make a number of false choices:
* why can't a voter say that they prefer several candidates equally?
* why must choices be ranked?
* why do candidates have to be eliminated?
* why can't lower rankings be considered?
Ted
dlw
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."
"Renewable energy is homeland security."
Fundamentals of Verifiable Elections
http://kathydopp.com/wordpress/?p=174
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
