[EM] Cloneproofing Random Pair and Random Candidate?

[Jameson Quinn]

> Hay voting seems to have been invented to encourage the expression of
> sincere utilities, as distinct from von Neumann-Morgenstern utilities, and
> there is a whole practically knee-jerk assumption in voting system theory
> that "strategy" is "bad."


You've made that assertion, that people are looking for low-strategy
methods only out of an inherent aversion to strategy, before; and I
disagree. There are several problems people could have with strategy, and
different people (theorists) focus on different ones:
1. Some say it's inherently bad because it's dishonest. (I don't care).
2. Widespread strategy can have a disastrous impact on utility/BR. (Think
Borda, where this phenomenon is pretty well-documented, at least
anecdotally.)

2a. If different factions have different inclination or capacity for
strategy (for instance, due a take-no-prisoners attitude or better access
to polls) it could lead to "unfairness". This would impact legitimacy as
well as utility/BR, and the mere whiff of this could be enough to cause a
system to be repealed.

3. Strategy is a cognitive burden on voters.

You seem to be saying that 1 is the only possible reason, but I weight
reasons 2 and 2a heavily, and 1 almost not at all.


> In Range voting, strategic voting refers to something radically different
> from what was strategic voting in ranked systems: in the latter, it was
> reversal of preferences. That reversal occurs to me as radically unfair,
> that is, a system that encourages such reversals is unfair. But Approval
> and Range don't do that.


Neither do most graded Bucklin methods (ie, any of them that don't rely on
an inherently ranked method like Condorcet for the tiebreaker).


> I think there are some weird circumstances with multiple majorities in
> Bucklin that could lead a voter to think they should reverse preference,
> but ... that is a rare circumstance, difficult to anticipate, and of
> questionable value.


There is no such circumstance for rational voters. Obviously irrational
voters could wrongly conclude this, but that could happen for any method.
Range, Approval, and graded Bucklin methods (like GMJ) are all 100%
semi-honest if voters have either perfect or zero information about each
ballot. (That is, if for instance everyone knows that 20% of voters each
are A>B, B>C, and C>D, and nobody knows about the remaining 40%, then all
rational strategies are semi-honest)

However, there are crazy circumstances of incomplete information where
non-semi-honest strategies are rational, for all three kinds of methods.
For instance, if I know that all other voters will vote purely based on
party, and I am the only voter who cares about gender (I want to elect a
woman), BUT I have no idea which party is preferred by more voters, I might
be rational to approve women who I prefer less over men I prefer more. That
is an unrealistic situation, though, because it very quickly breaks down if
I think that there are even a few other gender-noticing voters (whether or
not they agree with me on either party or gender preferences), OR if I have
an idea of which party is in the lead. Usually, both of the latter
circumstances pertain.



> Runoff Bucklin fixes or ameliorates possible problems on this.
>
> (For starters, it would make multiple majorities rare, unless there
> *really are clear multiple approvals without strong preference between
> them,* and I usually write about this that "we should be so lucky.")
>
>
>  The multiround stuff breaks the provable strategy-proofness, and what
>> remains is largely handwaving. Even if that handwaving is in some deeper
>> sense "correct" that strategy is ineffective, it could still fail to matter
>> if people believe that strategy will pay off somehow. So the benefits are
>> dubious; indeed, I very much doubt that they're worth the complexity.
>>
>
> It's a long way to go to try to eliminate voters considering practicality
> when they choose how to vote.
>
> What we want, I suggest, is to encourage sincere designation of favorites,
> when they are clear to the voter. That's considerate of voters. It is also
> considerate of voters to *allow* equal ranking at all ranks. And, then, it
> is simple and has utility benefit to not eliminate unexpressed ranks. I.e,
> Borda becomes Range if equal ranking and empty ranks is allowed. And why
> not do this? It reflects how voters actually think. There are *choices* to
> be made in this process, and those choices probably will reflect "strategic
> consideration," but if it's a zero-knowledge election, they will purely
> represent an estimate of preference strength.
>
> Bottom line, a simple ballot design that collects preferences and
> preference strengths from voters is ... a Range ballot. If resolution is
> adequate, it can be voted as a ranked ballot -- and it might be so
> analyzed, for part of a system.
>
> But that is the basic information.
>
> I have argued that when voters shift their preference *strengths* away
> from what would be "purely sincere," it's because they *care* about the
> effectiveness of their vote. Often, we pretend that someone with a slight
> preference for A over B, in Approval, *should* approve both (or neither).
> Yet if it matters to them which is elected, they do not have a weak
> preference, they have a strong one. Maybe their strong preference is based,
> not on the candidate, per se, but on party affiliation. And that's
> important to them!
>
> Essentially, voters make these choices based on their own preferences and
> judgements, and a good system will analyze the ballots and determine what
> candidate will maximize social utility, and if that's not clear, the system
> may refer the matter again to the voters.
>
> When an explicit majority support for an election has not been obtained,
> it is *not clear* that the electorate would approve of that result. What we
> have are rules that have been designed to force a conclusion, that being
> considered "desirable." Yet this is what I call the past binding the
> future. By setting up such a rule, today, we are telling tomorrow's
> electorate what is best for them. Bad Idea.
>
> However, in the real world, we also make certain compromises.
> Nevertheless, the goal of majority approval remains highly desirable. These
> highly abstracted and complex voting systems don't address us where we live.
>
> We may use stochastic methods, on occasion, but generally only when we
> have narrowed options to a few, considered to be roughly equal in value.
>
> However, Asset Voting bypasses the need for all these complex, a-priori
> fixed, methods. Terminally simple, yet *totally flexible,* once we
> understand that the Electoral Assembly that Asset creates could do whatever
> it decides, by majority vote, is needed. If it needs to toss a coin, it can
> agree to do that. Asset can also nominate new candidates in-process,
> whatever will find full representation in the Assembly to be elected. It
> also sets up benefits that are far broader than mere completion of an
> election.
>

Asset is an awesome system, but it has one flaw: it is not guaranteed to
resolve in any finite time. This is one of the rationales for more-complex,
Asset-inspired systems like SODA.

Jameson
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20130404/a20f0006/attachment-0002.htm>