[EM] Fwd: Pseudo-election reform in California

[Dr.Ernie Prabhakar]

[Forwarding as directed... - dr.e]

> You may quote me on the E-M list, but please keep me anonymous.

Begin forwarded message:

> I was the unnamed quoted respondent to Ernie yesterday.  Yes, I'm an 
> amateur
> newbie wading into the discussion ;-).  And probably everything I 
> state below
> has been suggested already ...
>
> On 1 Jun 2004 at 16:00 PDT, Ernie Prabhakar wrote:
>>>
>>> STPV reduces tactical voting in the primary since one places one's
>>> most-preferred candidates highest to prevent their elimination.  It 
>>> is also
>>> unlikely that, say, a Republican voter would rank a unelectable 
>>> Democratic
>>> candidate higher than his/her own favored candidate.
>
> I slept on this and realized that STPV-PV doesn't reduce tactical 
> insincere
> voting at all.
>
> If a voter of one party wants to spoil the election for the other 
> party and
> the first party candidate has a comfortable lead, that voter could 
> easily vote
> for the worse of the two alternative party candidates, or even a third 
> party
> candidate, in order to reduce the likelihood of realistic competition.
>
>>
>> So, I interpret that -- if one assumes the existence of a primary --
>> as  recommending STPV-PR for the first round, then a three-candidate
>> Condorcet for the final round.
>>
>> In the California case, if we can't get Condorcet for the general
>> election, we'd have to pick two candidates, and it still sounds like
>> either Approval or STPV-PR would be the optimal primary.
>>
>> Thoughts?
>
> I think as Mike pointed out, one has to add some incentive to encourage
> sincere ranking.  The technique used currently in US state and local 
> elections
> is to have a way for the primary winner to be the actual winner.
>
> That leads me to think that the primary election should be a single 
> winner
> election (Condorcet wv), with a run-off between the top 2 or 3 Approval
> candidates (Condorcet if 3) as a fall back only in case of a cycle.
>
> A voter could rank candidates (equal ranking allowed ) and add a line
> demarking the approval cutoff.  Note that this has attributes of AERLO 
> and
> ATLO, but would be far simpler to implement.  If no line were added, 
> ranked
> candidates would be considered to be above the line and unranked 
> candidates
> would be below it.  [Mike -- Is this different from ATLO?]
>
> Condorcet would encourage more centrist appeal in the primary, while 
> the
> threat of an Approval run-off to winnow the field would have the same 
> effect
> as AERLO and ATLO options in discouraging tactical insincere ranking.
> Approval-based run-off would reduce the controversy of a large 
> Condorcet cycle
> (and possibly contradictory cyclic resolution results) and also 
> increase the
> popular mandate of the eventual winner -- even if there is another 
> cycle
> (possibly even the same one as in the primary) that needs to be 
> resolved using
> Ranked Pairs / Beat Path.
>
> AERLO and/or ATLO or some other technique might eventually have to be
> introduced in the run-off if 3 candidates are allowed. But for 
> simplicity, the
> same approval line option could be used as in the primary, with the 
> effect of
> a combined AERLO & ATLO -- drop back to "approval" (equal ranking 
> approval
> above the line, no ranking below the line) in case no unambiguous 
> Condorcet
> Winner exists.
>
> I think the potential expense of a second election is justified 
> because a
> cyclic "tie" really means that the voters deserve a more detailed look 
> at the
> candidates.  And that would certainly happen with the focused campaign 
> that
> would result after a controversial primary.