[EM] range versus condorcet & others; practical purposes

[Rob Lanphier]

On Sun, 2005-08-14 at 17:20 -0400, Warren Smith wrote:
> >robla: Condorcet has zero chance in 2005.  It has a small chance in 2010, and
> >better than even odds in 2050.  That's assuming we ignore your advice
> >and actually continue our work.
> 
> --what is your strategic plan?  One can make statistical estimates of "chances"
> based on polls and one can estimate chances based on who will support it.  Third parties
> will support range because range experimentally gives third parties the most support by a lot.

If you had the kind of backing that CVD has, I might believe you.
However, in terms of popular voting reforms, only CVD can make the claim
that they've got the political organization and the momentum to follow
through right now.  CAV/AAV is making encouraging progress, but it's
nowhere close to CVD in raw political power, as near as I can tell.

That's not to say that I think the CVD is right (I don't; I think
they're dead wrong).  Only that CVD is the only one that can credibly
throw their clout around the way that you are right now.

> That gives range a support base who need it for their very survival.  If
> we can convince them.  Now if anybody whatsoever works to try to convince third parties to
> adopt any system whatsoever other than range, that will simply slow down those parties'
> ultimate realization that range is the best for them.  I will not countenance
> any attempt to claim that any system besides range favors US third parties more.  Because
> I have done the world's only study of that issue.  Period.  Any such attempts
> to confuse third parties will (if third parties are really a key ingredient
> in the struggle for voting reform) therefore result in the opposite of benefit for society.
> There are only two reasons to try to confuse third parties in that way (a) if you are
> a covert operative of the 2 major parties, or (b) if you believe third parties will
> not be useful players in the voting reform struggle.

Convince me that Range Voting is superior to Condorcet, and I'll be an
advocate for Range Voting.  I've made a jump once already after an email
debate (from STV to Condorcet, back in 1995), and I'm willing to
consider another jump.

However, I myself will admit that I'll be very, very difficult to
convince.  I've been thinking about the subject for a decade, and
watched discussion on this list and others.  I'm not going to change my
mind over the course of a weekend...it just doesn't work that way.
Especially when you make faulty assumptions that I must either be (a) or
(b) by advocating something other than Range.

> >Right now, small associations are using Condorcet methods (the Debian
> >project being the flagship), and are giving people valuable real-world
> >experience in how it works.  I suspect that Condorcet methods will
> >continue to be relegated to computer-savvy organizations for a while
> >longer.
> 
> --good. However, I doubt this actually counts as "experience" because it
> is hard to assess strategic voting effects in the absence of advertising
> and political parties, and probably nobody is trying to keep
> assessing that.  Am I wrong?  I'd like to be wrong.  I have seen studies
> of trade union elections and the IEEE adoption of approval trying to draw conclusions, and
> I was never very convinced of anything by any of those studies.

It's not perfect.  I'd like to see more experience where the stakes are
higher and the field is wider.  But I'm willing to live with what we see
today for a while longer.  There's plenty of software to be written, and
lots of small glitches to work out of the system before we get in front
of a white-hot spotlight.

> >That's not to say that compromise can't be a smart strategy.  I believe
> >that Approval voting, for example, would be a /great/ reform for
> >primaries.  In fact, I'd go so far as to say that the ideal method would
> >be a form of Approval for a unified primary (where either all candidates
> >exceeding a certain Approval threshold move on to the general election,
> >or some fixed number of candidates move on), followed by a
> >Condorcet-compliant general election.
> 
> --actually approval is at its weakest in large elections with more-honest and 
> less-strategic voting - as primaries often are.  Range is maximally superior
> to approval in that regime as is shown by my sim study.  That is one reason
> I want range in Iowa 08.

Interesting.  I'm reconsidering my position on this one.  I starting to
see where approval can be weak in large elections.  Some anecdotal
evidence I'm mulling over from the Wikipedia Arbitration Committee
election in 2004 makes me think this is true.

> >However, telling people to forget looking for the best system,
> 
> --actually, if you read my post, I in fact said I *WANTED* EM to continue
> forever looking for the best system.  Thank you for saying I said
> exactly the opposite of what I said.  I really enjoy that.

My apologies.  I reread it, and it looks like I did misconstrue it.  In
particular you said: "So while I want the mental masturbation to
continue unabated, I think we have to unify as a practical matter behind
range voting."

However, it's still condescending as hell.  Let me paraphrase: "please
continue to discuss whatever silly little things you want to, but oh, by
the way, stop actually advocating what you believe in and unite behind
what *I* believe in".  Let me respond:  "ummmm....no".

Look Warren, I'm not trying to pick a fight here, but you're going to
have to have a little patience.  We're not just going to drop what we've
been doing (in my case, a decade...others possibly longer) because you
come onto this list and declare it to be so.  You *may* actually
convince some people here if you take some time and humor our questions.

Thanks
Rob