There’s a strong chance that Condorcet would be held unconstitutional.  It doesn’t mimic a primary and runoff elections like IRV does.  Rather, it mimics a sports season where the team with the best record wins the title.  All its (virtual) contests are one-on-one.  Condorcet counting has no phase when all the candidates are considered together, which I think is at the heart of the traditional (framers’?) notion of elections.

On the other hand, our society’s norms are moving closer and closer to treating politics as sport, which is legally a form of entertainment.  When competition is valued over representation, or when making politicians “work harder” for votes becomes a dominant desire (that was the top reason offered by campaigners for a yes vote in the AV referendum - why?), courts or constitutional reformers might favor single-winner Condorcet elections.  Or we might start enjoying the fun of having 162 election days a year (or 9, so the top 10 “teams” can face each other once).  Play ball!

  - dah

Can I add to Rob's second dot point a third - that if there's a circular tie (a stone-scissors-paper "cycle" or "loop"), there are various ways of breaking it, but the one most favoured by Condorcet supporters seems to be a "benchmark" method - ie, each candidate's "benchmark" is the lowest s/he polls in any pairwise contest, and the winner is the candidate with the highest benchmark (eg, "Blue is the only candidate whose vote never dropped below 47% against any opponent").

This can mean that Blue wins because (say) Orange beat him/her by "only" 53% to 47% (and no one else beat Blue by a wider margin). However, this seem to nullify one of the main pro-Condorcet arguments, which is that AV-IRO might elect Blue (after exhaustive elimination) even though Orange would have defeated Blue one-on-one had Orange not been eliminated early in the count. After a series of exhaustive pairwise comparisons, we end up back with electing a candidate who is preferred by fewer voters than a defeated rival anyway.

Worse, Condorcet would seem to risk undermining public confidence in the voting system. With IRO-AV, the winner wins because s/he defeats his/her main rival (say) 53-47% (or even 48-42% if a lot of voters bullet and their ballots exhaust). By contrast, the "decisive contest" under Condorcet, the one that clinches the race, the one that public attention focuses upon, will be (if there is no beats-all candidate) the Orange-Blue throwdown where Orange gets 53% and Blue 47% but Blue nonetheless wins. Note how IRO-AV was vulnerable to the "the loser wins" soundbite (a candidate with 35% can beat a candidate with 43%) and multiply this by ten.

Even when a beats-all candidate does exist, Condorcet is still vulnerable to the "so what?" objection. Saying that a candidate who is the first preference of 1-2% could beat every other rival head-to-head is like saying that Molly Ringwald [*] might agree to marry me if she and I were marooned alone on a desert island. Well, yes, but in the real world there are a lot of other rivals in the field first. IRO-AV narrows that field in a way that gives high, but not sole, weight to first preferences and greater probabilistic weight to second and third preferences than to ninth and tenth preferences.

[*] Showing my age here. Feel free to insert a different name with more meaning to the reader...


At 03:55 8/05/2011, Rob Richie wrote:

I'd take Condorcet voting over plurality voting in a heartbeat, but I suspect it will be very hard to convince people to adopt it -- certainly it will be harder than winning IRV-alternative vote, which can be hard as it is. A couple things Condorcet voters have to explain:

* A candidate can win who is nobody's first choice.

* You can have a cycle where no candidate defeats every other candidate (e.g, A defeats B, B defeats C and C defeats A) - -and then have to get to tiebreaking procedures. Due to this, your selection of a second choice can count against the chances of your first choice, which will trouble people.

More broadly, Condorcet discounts the "tribal" reality of politics as it is. Take Larry Levine's, bitter messages yesterday against Jean Quan's win in the mayoral race in Oakland last year. I think Quan's victory was quite clear. She defeats Don Perata in a one-on-one comparison (both citywide and in well over half the precincts) and won more votes than any Oakland mayoral candidate had won in decades, but Larry can't get over the fact that she had fewer first choices than Perata. Imagine if instead of Quan winning it was some candidate who trailed much further behind in first choices as in 4th or 5th place in 1st choices. Accepting that result will take a real change in what people expect winners to achieve.

I'll use one real example of another aspect to consider: the 2009 Burlington mayoral race with IRV. The plurality vote leader in first choices was the Republican candidate. He lost to the second place Progressive Party candidate in instant runoff voting, just as he would have lost to him in a runoff if all the same voters had come back to vote.. He also would have lost to the third place Democrat in a one-on-one race - and so would the Progressive. So the third-place Democrat was the Condorcet winner, but observers of that race would generally say that the Democrat ran a lousy, uninspiring campaign. Condorcet would reward him just by his positioning on the ideological spectrum (between the Progressives voters and the Republican voters). Even when working clearly, therefore, Condorcet would create a kind of "tyranny of the center", with that centrist candidate winning even if running a lame campaign. (And as a coda, all the political energy against IRV came from backers of the Republican candidate, the "condorcet loser," rather than the Condorcet-winning Democrat who backed keeping IRV despite his defeat).

Again, this critique is all relative, but I like the fact that IRV winners have to go out there and earn first choices by inspiring some real support -- even as they also get value out of earning second and third choices of backers of others candidates. An intern did a thoughtful piece on this issue last year -- see:
http://www.fairvote.org/why-the-condorcet-criterion-is-less-important-than-it-seems/

Rob
On Sat, May 7, 2011 at 12:29 PM, Scarberry, Mark < Mark.Scarberry@pepperdine.edu> wrote:

If you require mandatory ranking then you could use a Condorcet approach, right? My math whiz friend says that you only need very minimal assumptions to get a Condorcet ranking out of the voting, if each voter ranks each candidate.

 

I’m not a Condorcet expert by any means, and this subject may have already been beaten to death on this list. But my best understanding of it is that you run each candidate against every other candidate in head to head races, and the one who beats all the others in these head to head races is the Condorcet winner. It makes sense to me to treat such a Condorcet winner as the winning candidate.

 

If every voter ranks every candidate then you can derive the head to head results as long as each voter’s preferences are transitive (if that’s the term). But you still have the possibility that A will beat B who beats C who beats A. We’ve never had that happen in faculty voting on new hires, but you need a way to break the tie if you get such a “cycle.” Perhaps someone can tell us whether there is a jurisdiction that uses Condorcet voting, and how likely it is that you will get a cycle.

 

Mark Scarberry

 

Mark S. Scarberry

Professor of Law

Pepperdine Univ. School of Law

Malibu, CA 90263

(310) 506-4667

 

From: election-law-bounces@mailman.lls.edu [ mailto:election-law-bounces@mailman.lls.edu] On Behalf Of Rob Richie

Sent: Saturday, May 07, 2011 8:22 AM

To: JBoppjr@aol.com

Cc: David@holtzmanlaw.com; election-law@mailman.lls.edu

Subject: Re: [EL] a "thumping" in the U.K. AV referendum

 

Hi, Jim,

 

There are a couple issues here.

 

1. Some American jurisdictions have implemented instant runoff voting with a limitation on rankings tied to current voting equipment inflexibility -- three rankings in such cities as Oakland, Minneapolis and San Francisco. That means it's possible that you could rank three candidates who don't end up in the final two in the IRV tally.

 

This limitation on three rankings is subject of a federal lawsuit against San Francisco, but it is a losing case -- by summary judgment at the district level and soon from the 9th circuit, based on the tenor of oral argument. The plaintiffs' argument is dependent on seeing each round as a "separate election," which is not the case. If plurality voting is okay, with voters limited to one ranking, then giving voters to backup opportunities to cast a decisive vote is okay too.

 

2. More broadly, most implementations of IRV don't limit rankings, but allow voters to abstain from ranking every candidate. If you rank some candidates and then choose not to rank remaining candidates, you are expressing indifference to those remaining candidates -- essentially saying you wouldn't have voted if those were the only candidates running. If we grant voters the right to abstain, you can see how that decision to abstain from ranking is no more of a dilution of the will of the voters than saying that all the people who don't vote at all should be counted against the winner.

 

Australia has mandatory voting -- and mandatory ranking in its IRV elections. So if you want to force people to express an opinion, you can -- but I suspect a lot of Americans might resist that notion, as abstention is one means of expressing a political opinion.

 

Given Jon. Huntsman's potential presidential candidacy in a very divided presidential field next year in which "winners" of early contests may have low percentages of the vote, there's value in taking a look at an IRV election in which he participated: at the 2004 Utah state convention with several thousand votesrs See coverage of that race here:

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

 

And the actual count, round by round, here -- there were eight candidates and the field narrowed to two, with those two going onto a primary because neither earned 60% to win the nomination outright:

http://archive.fairvote.org/irv/utahresults.htm

 

best regards,

Rob Richie

 

On Sat, May 7, 2011 at 10:13 AM, <JBoppjr@aol.com> wrote:

     As I understand it, the result of "ranked-choice" is that some voters are left out, even potentially a lot of them, if they did not "rank" any of the remaining candidates. How can this result in an election that reflects the will of the voters? Jim Bopp

 

In a message dated 5/6/2011 6:51:56 P.M. Eastern Daylight Time, David@holtzmanlaw.com writes:

In Los Angeles, the League of Women Voters advocates using the “alternative vote.” We call it “instant runoff” or “ranked-choice” voting. Many politicians do resist it, since the existing system has served them well. Nevertheless, some members of the L.A. City Council, just two shy of the number needed, were willing to put it to a vote in March.

Voters here would likely have embraced the change, because ranked-choice ballots allow better expression of voters’ preferences, and instant runoffs (elimination of last-place candidates until a candidate receives a majority of the votes for candidates who remain) make elections fairer — and spare everybody the costs of a separate runoff election day.

While the British Prime Minister belittled the proposed election method as “only used by Australia, Fiji, and Papua New Guinea,” he left out neighboring Ireland, and paid no attention to the method’s success in San Francisco and Oakland, and at UCLA, where the Undergraduate Students Association Council recently voted unanimously to retain it.

  -- David A. Holtzman

President, League of Women Voters of Los Angeles

(I wrote this in response to the L.A. Times story, http://www.latimes.com/news/nationworld/world/la-fg-britain-election-20110506,0,3383863.story )

On 5/6/2011 2:15 PM, Rob Richie wrote:

Yes, indeed. British voters sent a message that they don't want a voting system that will cost them 250 million pounds, give some voters more votes than other voters and give the currently despised Liberal Democrats ongoing power to control government. Now what they would think about the alternative vote as it actually is, who knows...

 

One must accept the voice of the people on this, of course, and it does demonstrate that new voting rules can have difficulty withstanding over-the-top miscategorization, particularly when coming from people (like the David Cameron for the Conservative Party and some of the "old lions" of the Labour Party that opposed AV) that people want to believe in.

 

For folks recognizing the origins of our electoral rules in Britain, however, there is an important story to be told in reading the article Rick circulated to the bottom. For one, the genie is out of the bottle as far as a neat-and-tidy two-party system there. The Scottish National Party won a majority of seats in Scottish assembly elections, for example, and the two major parties (just like in last year's elections) continue to share a growing share of votes with other parties (last year, a third of voters didn't vote for the two major parties, and more than half of districts were won with less than 50%).. Plurality, "top of the heap" voting isn't meant for such a political reality, so the conversation about what to do about it will continue whether the Tories want it to or not.

 

Second, the UK is far ahead of us in using alternative voting systems in key elections. Scotland and Wales used "mixed member" proportional representation (MMP_ yesterday, for example. MMP is an intriguing method developed with American leadership in Germany after World War 2 that combines winner-take-all districts elections with proportional voting, one also adopted in a national referendum in New Zealand in 1993. Northern Ireland yesterday used the choice voting, AV-type single transferable vote system to elect its regional assembly and local governments, as Scotland does in its local elections as well. Next year, London will elect its mayor with a form of the alternative vote (simplified to voters having two rankings and candidates needing to finish in the top two to win) and MMP for city council.

 

So onward, despite a lot of disappointment in our reform world,

Rob Richie

 

On Fri, May 6, 2011 at 4:21 PM, Rick Hasen <rhasen@law.uci.edu> wrote:

http://www.guardian.co.uk/politics/2011/may/06/av-over-yes-campaign-routed

--

Rick Hasen

Visiting Professor

UC Irvine School of Law

401 E. Peltason Dr., Suite 1000

Irvine, CA 92697-8000

949.824.3072 - office

949.824.0495 - fax

rhasen@law.uci.edu

http://law.uci.edu/faculty/page1_r_hasen.html

William H. Hannon Distinguished Professor of Law

Loyola Law School

http://electionlawblog.org

_______________________________________________

election-law mailing list

election-law@mailman.lls.edu

http://mailman.lls.edu/mailman/listinfo/election-law

--

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

"Respect for Every Vote and Every Voice"

Rob Richie

Executive Director

FairVote 

6930 Carroll Avenue, Suite 610

Takoma Park, MD 20912

www.fairvote.org rr@fairvote.org

(301) 270-4616

Please support FairVote through action and tax-deductible donations -- see http://fairvote.org/donate. For federal employees, please consider a gift to us through the Combined Federal Campaign (FairVote's CFC number is 10132.) Thank you!

_______________________________________________

election-law mailing
list

election-law@mailman.lls.edu

http://mailman.lls.edu/mailman/listinfo/election-law

 

--

David A. Holtzman, M.P.H., J.D.

david@holtzmanlaw.com

Notice: This email (including any files transmitted with it) may be confidential, for use only by intended recipients. If you are not an intended recipient or a person responsible for delivering this email to an intended recipient, be advised that you have received this email in error and that any use, dissemination, forwarding, printing or copying of this email is strictly prohibited. If you have received this email in error, please immediately notify the sender and discard all copies.

 

_______________________________________________

election-law mailing list

election-law@mailman.lls.edu

http://mailman.lls.edu/mailman/listinfo/election-law

_______________________________________________

election-law mailing list

election-law@mailman.lls.edu

http://mailman.lls.edu/mailman/listinfo/election-law

--

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

"Respect for Every Vote and Every Voice"

Rob Richie

Executive Director

FairVote 

6930 Carroll Avenue, Suite 610

Takoma Park, MD 20912

www.fairvote.org rr@fairvote.org

(301) 270-4616

Please support FairVote through action and tax-deductible donations -- see http://fairvote.org/donate. For federal employees, please consider a gift to us through the Combined Federal Campaign (FairVote's CFC number is 10132.) Thank you!

--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
"Respect for Every Vote and Every Voice"

Rob Richie
Executive Director

FairVote 
6930 Carroll Avenue, Suite 610
Takoma Park, MD 20912
www.fairvote.org rr@fairvote.org
(301) 270-4616

Please support FairVote through action and tax-deductible donations -- see http://fairvote.org/donate. For federal employees, please consider a gift to us through the Combined Federal Campaign (FairVote's CFC number is 10132.) Thank you!

_______________________________________________
election-law mailing list
election-law@mailman.lls.edu
http://mailman.lls.edu/mailman/listinfo/election-law

_______________________________________________
election-law mailing list
election-law@mailman.lls.edu
http://mailman.lls.edu/mailman/listinfo/election-law