[EM] Sincere methods

[Juho Laatu]

Hello Kevin,

On Apr 27, 2005, at 00:14, Kevin Venzke wrote:

>> And if one changes the winner based on a false clone
>> assumption, then one may violate the rights of the candidate that 
>> would
>> have won without the clone assumption.
>
> I think this is kind of silly. What "rights" could be violated?

What I meant is that if some rule (e.g. defending clones) is applied 
without proper reason (false cones instead of true clones) and the 
winner is changed from one candidate to another, then the winner 
candidate that was not elected may be unhappy because of this.

(I obviously defined clones based on what the situation is in real 
life. It is possible to define clones also so that preferences in the 
given votes define what is a true clone set without considering the 
situation in real life, in which case there would be no separation 
between false and true clones.)

> But my point is that, other than simplicity, I can't see why 
> "additional votes"
> gives more useful answers than some other imaginary experiment. A 
> better argument
> seems necessary to oppose Smith.

Fair enough. I have had problems justifying Smith. I accept that many 
different methods could be sincere methods (maybe for different 
purposes, maybe for different people). We are on a good track as long 
as someone believes that certain method produces the best winner with 
sincere votes and good justification is provided. My justification of 
additional votes / minmax(margins) is not foolproof or one that could 
be proven right, but I think it has some good points in it anyway.

>>> 20 A>B>C>D
>>> 20 B>C>A>D
>>> 20 C>A>B>D
>>> 13 D>A>B>C
>>> 13 D>B>C>A
>>> 13 D>C>A>B

>> If there was a second round, A, B and C supporters would have an
>> interest to join forces against D. The needs of A, B and C supporters
>> would be served better. But needs of voters in general would be served
>> worse (if MinMax is used as the SVM). Better not tell to A, B and C
>> about this possibility :-). Maybe the voting method should not help
>> them implementing the strategy.
>
> No, because the method should have the sense to see that this strategy 
> is
> what the ABC voters would attempt in a second round.

I don't like methods with second round since that increases the risk of 
strategic voting considerably. If there is no second round, then it is 
easier to avoid strategies (=> hide or otherwise try to discourage 
their use).

(Also opinion polls before the election have similar effects than 
second round has but I don't dare to propose to banning them :-).)

>> If A, B and C are true clones and all belong to party "ABC", then the
>> situation changes a bit. We could expect that A, B and C are friendly
>> competitors and they now regret that they didn't arrange a 
>> pre-election
>> where one of A, B and C would have been elected as the only "ABC" 
>> party
>> candidate for the main election.
>
> It is *bad* if candidates wish they had had a pre-election. The method 
> should
> not fail to work just because some candidates are added in.

Agreed.

> I don't think it matters whether ABC are "true clones." Whether 
> candidates
> are considered "clones" is based on how the votes are cast. The ABC 
> voters
> don't have to cooperate with each other if the method has the sense to 
> see
> that {a,b,c} would beat {d} in a one-on-one election.

It is maybe not too bad (although not ideal) to overdo the defence and 
defend also groups that are not really clones.

>> If we do not elect D, we may violate one of the principles of basic
>> ranking based election, namely the fact that only relative preferences
>> should be considered, not their strengths. Vote A>B>C>D should mean
>> A>B, A>C, A>D, B>C, B>D and C>D and nothing more (not e.g. A>B>C>>>>D
>> which could be a good guess if A, B and C were true clones).
>
> Eh? You don't need to make assumptions about preference strength to 
> elect
> from the Smith set.

Yes, Smith set is a clearly defined set with not relationship to 
preference strengths.

> What would justify MinMax as
> a sincere voting method? Being "natural" and easy to explain, right? 
> Then, sure,
> I don't see why not. But I don't yet understand why such methods are
> interesting to consider.

The key property of a SVM is that it elects the best candidate (with 
certain criteria) with sincere votes. Naturalness and easy explanations 
are just typical characteristics of such methods. Minmax thus has to 
have the capability to pick the best winner according to some 
understandable criteria in some type of elections. I may have not been 
able to provide good enough explanations why I find it to be one good 
SVM.

>> 101: A>B>X>C
>> 101: B>C>X>A
>> 101: C>A>X>B
>> 100: X
>
> X is on the wrong side of a majority in every contest. For that reason 
> I
> consider it contrary to the purpose of an election to elect X.

Some SVM might emphasize the number of contests where certain candidate 
loses. Others may emphasize the strength of the worst defeats. Maybe 
these two are both good SVMs for two different purposes.

Best Regards,
Juho