[EM] Remember Toby

[fsimmons at pcc.edu]

I hope that everybody gives careful thought to the three parts of Jameson's letter, (1) the heuristic 
plausibility argument for the high degree of Condorcet compliance, (2) the near impossibility of winning 
candidates throwing the election through back room deals, and (3) a great call to focus on something 
that we can all support.

Just a few random loose ends.  

If you delegate your approvals to some candidate, and that candidate approves your most detested 
candidate X, automatically everybody he publicly ranked ahead of X gets approved by him as well.  So if 
he ranks X high on his list, you should probably do your own approvals by simply approving the 
candidate that you would vote for under Plurality as well as everybody you like better.

If the candidates' public rankings are truncated, then they can only approve down to the truncation mark, 
but no more.  Under this rule most voters will find that their favorite did not even rank their most despised 
candidate X, so they don't have to worry about that.

Kathy was worried that courts might consider Approval in violation of one-person-one-vote requirements.  
There are many ways to get around this:

(i)  Interpret the intent of the law as meaning that each voter gets copies of the exact same anonymous 
ballot except possibly for randomization of the candidate order to avoid giving any candidate special 
advantage from being listed first or last.

(ii) Each voter gets one vote on each question, and for each candidate the question is do you or do you 
not approve of this candidate.  A mark is a positive response, while a blank is a negative response.  So 
each voter gets exactly one vote per candidate.

(iii) (suggested by Martin Harper ten years ago) Complete the Approval election with the following 
transfer of votes, so that in the end each voter's vote is cast to only one candidate in the race:  your vote 
is transferred to that candidate who received the greatest approval among the candidates that you 
approved of.  This transfer step does not change the Approval winner.


With regard to the problem of "chicken" in the presence of nearly equally matched clones, I suggest that 
in the process where the candidates cast the votes delegated to them, they all simultaneously cast their 
first delegated approval votes, then their second, etc. until all of them (with votes left to cast) vote the 
same twice in a row.  Then their remaing votes must be identical to those two.

It is well known that repeated play tends to yield a satisfactory solution to the games of prisoner's 
dilemma and chicken when the players are rational.  

> From: Jameson Quinn 
> To: EM 
> Subject: Re: [EM] Remember Toby
> Message-ID: 
> Content-Type: text/plain; charset="iso-8859-1"
> 
> Message contents:
> Section 1. When *isn't* SODA "more condorcet compliant than condorcet
> methods"
> Section 2. Smoke-filled rooms?
> Section 3. What are we looking for, anyway? (in this thread)
> 
> 
> *Section 1. When isn't SODA "more condorcet compliant than condorcet
> methods"*
> 
> 2011/6/4 
> 
> > For the benefit of those who are advocating ranked ballots in 
> order to
> > achieve Condorcet Compliance,
> >
> > note well that Jameson has a good argument that SODA, a simple 
> method that
> > uses only a Plurality
> > style ballot, is more Condorcet Compliant than most well known 
> Condorcet> methods.
> >
> 
> I believe that is true in a practical sense. However, I should 
> note that I'm
> not claiming that SODA achieves the impossible. As with other methods,
> (attempted) strategy could spoil its condorcet compliance. I'll 
> explain how,
> and why I think that wouldn't happen.
> 
> First off: I'd like to note that I'm mostly worried about burial 
> strategyhere. Generally, favorite betrayal strategy is useful to 
> break an honest
> Condorcet cycle to your advantage, while burial is useful to 
> create a false
> cycle which gives you some advantage. Since I think that honest 
> cycles will
> be rare, I'm more worried about the latter. Also, 
> psychologically, most
> people have a much higher propensity for burial than for 
> favorite betrayal.
> At any rate, in all the discussion that follows, I will assume 
> that there is
> an honest pairwise champion (CW).
> 
> So, in SODA, burial/truncation is still possible in several ways:
> 1. A candidate could cause a cycle through burial, and thus 
> avoid the
> (known, unique, strong, stable) Nash equilibrium for the honest 
> pairwisechampion. However, that can only work to their advantage 
> if the other
> candidates actually believe that the false, strategic preference 
> order, or
> if they manage to exploit a dishonest strategic mistake by another
> candidate. I believe that high-profile frontrunner candidates 
> could not
> plausibly claim a false preference order, so this strategy would be
> ineffective.
> 2. Individual voters could use truncation (not burial, because it's
> approval). For instance, voters could prevent their votes from being
> delegated in order to engage in games of chicken to ensure that their
> preferred near-clone was elected. This is a fundamentally 
> unnecessary risk,
> however. The near-clone with an honest lead does not need such 
> tactics, and
> the near-clone who is behind will probably need a dangerously 
> large number
> of voters to do such tactics. I don't see how this could be 
> coordinated on
> an effective scale in real life without backfiring.
> 3. Individual voters could vote for "false flag" minor 
> candidates whose
> honest preference ordering happened to be the strategic burial 
> ordering that
> they sought. I find this totally implausible, though; this 
> requires a level
> of cold-bloodedness and sophistication that only a tiny portion 
> of people
> have.
> 
> So when *is *SODA voting more condorcet-compliant than Condorcet 
> methods?When there is an honest pairwise champion; most voters 
> bullet vote, allowing
> delgation; and no frontrunner candidates (those with a chance of 
> winning)can plausibly claim a false preference order. *I believe 
> that these three
> conditions will hold most of the time.*
> 
> 
> *Section 2. Smoke-filled rooms?*
> *
> *
> Some people on this list have said that they don't like asset-inspired
> methods like SODA because of the "smoke-filled room" scenario. 
> That is, what
> if you voted for someone who eventually, in some crooked deal, 
> ended up
> giving your vote to your least-favorite candidate? Certainly, 
> I'm sure some
> UK Lib Dem voters might feel that way about Cameron, so it's not 
> a crazy
> idea. There are at least 4 reasons that I think this fear is 
> unrealistic;I'll list them from weakest to strongest.
> 
> The weakest reason first: hopefully, your favorite candidate 
> will be someone
> you can trust. Sure, Nick Clegg might have betrayed some part of 
> his base;
> but in SODA voting, that part of his base who didn't trust him 
> would have
> been free to choose a different, more-trustworthy candidate, 
> without fear of
> FPTP making their votes irrelevant. I find some comfort in this 
> argument,but this reason alone wouldn't convince me to trust SODA.
> 
> Second, there is the fact that candidate's preference orders 
> must be
> announced in advance. Sure, that doesn't stop candidates from being
> deliberately unstrategic in order not to help a higher 
> preference beat a
> lower preference, but that would be rare, and despite this, you 
> can still
> absolutely guarantee that your vote will not actually provide 
> the winning
> margin for that lower preference.
> 
> Third, the point of SODA is that candidates' post-election 
> strategy is done
> with perfect knowledge of the number of delegable votes and preference
> orders of the other candidates. This perfectly transparent 
> situation is the
> exact opposite of a smoke filled room. As I've already said, it 
> means that
> if there is an honest pairwise champion, then there will be a 
> known, unique,
> strong, stable, Nash equilibrium where that champion wins. 
> That's the gold
> standard of game theory. With such clarity, I don't see how back-room
> deal-making has a chance.
> 
> But fourth, and most important of all: in SODA, the vote 
> delegation is
> optional for the voter. If you don't want your vote to go to your
> least-favorite candidate, then don't bullet vote. If you only 
> want to vote
> for one, you can make your vote non-delegable by voting for a 
> blank or
> invalid write-in. End of story.
> 
> And delegation is not all downside. Remember, by voting a 
> delegable bullet
> vote, you are not just giving your voting power to a candidate, 
> you are in a
> real way letting them speak for you too. A leader/spokesperson 
> with 20% of
> the vote would be much harder for the media, and the eventual 
> winner, to
> ignore, than 20% of the voters who (in plurality) simply hold 
> their nose and
> vote for one of the two main parties. Again, the point is that 
> SODA gives
> you more voice and more transparency, not less.
> 
> *Section 3. What are we looking for, anyway? (in this thread)*
> 
> If this whole "remember Toby" thread is going to be productive, 
> I think we
> need to define the question we're trying to answer here. It 
> should be a
> question that could have a clear answer. Is it:
> 
> 
> - "What is the best system in practical terms?" This is an 
> unanswerable question, without first agreeing on the 
> definition of "best" --- something
> we're not about to resolve.
> - "What systems are clearly better than plurality?" This is an
> improvement over the above, because the bar is so low that, 
> even if our
> definitions of "better" aren't the same, we should be able to 
> find a good
> degree of agreement. However, by the same token, we'll end up 
> with a list of
> systems that's so broad that it makes a worse-than-useless 
> target for
> activism. A normal voter's eyes will glaze over before we're 
> done even
> defining the terms we need in order to start defining the systems.
> 
> 
> (Note: If I were answering this second question, I would 
> certainly include a
> number of Condorcet systems. I'm not going to spend time making 
> argumentsagainst those systems, even if I have such arguments, 
> because I don't think
> that's helpful. And I ask others to do the same; as much as 
> possible, this
> thread should focus on the positive.)
> 
> 
> - "What systems are better than plurality, for any reasonable 
> definition of the word 'better'?" I think that this is the 
> question we should be
> focusing on, because we can reasonably expect to agree on an
> acceptably-short list of answers.
> 
> That's why I think that Forest's idea of pareto dominance is 
> useful, not
> just for when we want to promote a proposal with ordinary 
> voters, but also
> in this discussion right here. I can accept that our goals and 
> values are
> not exactly the same. While we all share a basic desire to improve
> democracy, and would probably agree on the desirability of a 
> number of
> criteria considered independently, we won't always agree when it 
> comes to
> trading off one desirable trait against another. So if we are 
> looking for a
> practical proposal that we can all get behind, it should be one 
> that doesn't
> require any trade-offs, one that is strictly better than the 
> current system
> in every way. I think that SODA is the answer. (Though I'm 
> keeping an open
> mind, and would certainly happily support a number of other systems).
> 
> .....
> 
> Sorry for the long message; I hope I said enough to make it 
> worth it.
> 
> JQ
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: > electorama.com/attachments/20110604/91d92c73/attachment.html>
> ------------------------------
> 
> _______________________________________________
> Election-Methods mailing list
> Election-Methods at lists.electorama.com
> http://lists.electorama.com/listinfo.cgi/election-methods-
> electorama.com
> 
> End of Election-Methods Digest, Vol 84, Issue 12
> ************************************************
>