[EM] Rob: Condorcet's Criterion vs FBC. Will people favorite-bury?

[Rob Lanphier]

Hi Mike,

On Mon, 2005-10-03 at 05:12 +0000, MIKE OSSIPOFF wrote:
> You said that you don't think that significantly many people would 
> favorite-bury in a public BeatpathWinner election.
> 
> A voter will favorite-bury if, for that voter, the important goal is to keep 
> an unacceptable candidate from winning, and if favorite-burial will increase 
> the probability of accomplishing that.

I posit that very few voters will be in that predicament.  It is my
experience that most voters prefer candidates with a chance to win.
People that like to be part of quixotic movements are exceedingly rare.
That's why we've got our work cut out for us in the first place ;-)

Moreover, the cases where sincere voting regret are rare.  Combine that
with the innate desire that I think people who vote contrarian ballots
have to "make a statement", and I suspect we're talking about a
diminishingly small number of people who will reverse order.  Those that
have strong enough believes to vote against the tide, and yet are
fearful enough to reverse ordering on a ballot "just in case" I'll bet
are rare.

> 
> Because MDDA meets FBC and SFC, MDDA provides assurance to the voters who 
> otherwise would need to favorite-bury, and also to the voters who don't need 
> that assurance, and who, therefore, can benefit from the assurance that SFC 
> compliance gives.
> 
> If it were up to me, public elections would use MDDA. 

I have to think about this more.  For those who have lost track of the
acronyms, here's what's being discussed:
MDDA (Majority Defeat Disqualification Approval):
http://wiki.electorama.com/wiki/Majority_Defeat_Disqualification_Approval
SFC (Strategy-Free criterion):
http://wiki.electorama.com/wiki/Strategy-Free_criterion
and FBC (Favorite Betrayal criterion):
http://wiki.electorama.com/wiki/Favorite_Betrayal_criterion

In particular, I hadn't fully grokked SFC, and I'm still not sure I do.
In particular, I'm trying to see how this:
"If a Condorcet candidate exists, and if a majority prefers this
candidate to another candidate, then the other candidate should not win
if that majority votes sincerely and no other voter falsifies any
preferences."

.../always/ leads to this:
"SFC requires that the majority of voters who prefer the Condorcet
candidate to another particular candidate vote sincerely (neither
falsify nor truncate their preferences), and it also requires that no
other voter falsifies preferences"

MDDA would seem to encourage truncation in rare situations.  I don't yet
have an example, but I think one could be devised based on the mechanics
of the system.

I agree that SFC is a very important criterion, if perhaps misnamed.
I'm sure there were long discussions about naming it during the long
period where I wasn't very active on the list.  I was actually trying to
arrive at this criteria, doing many searches for "Majority Winner" and
so on, and was surprised to find SFC to be the name.

This may be a seemingly minor quibble, but I raise this because I
consider the two SFC quotes above to both be very important criteria,
and I'm trying to figure out how both apply to MDDA.

> Realistically, I 
> propose RV, with more rating-levels than Approval.

Range is a political stillborn.  This example kills it:

100 voters, two candidates, scale of 0-10:

90 voters: A=7, B=6
10 voters: A=0, B=10

A:630
B:640

B wins, even though 90% of voters prefer A to B.

There is no possible way Range will ever get serious support, given that
weakness.  If it manages to pass constitutional muster, it goes against
what I suspect is the instinct of most voters out there, including
myself.  I cannot be brought to recommend a system that suffers from
such a glaring defect.

Rob