[EM] [CES #9172] An idiosyncratic 4-D chart of voting system quality

[Jameson Quinn]

2013/7/26 Clay Shentrup <clay at electology.org>

> On Friday, July 26, 2013 7:47:41 AM UTC-7, Bruce R. Gilson wrote:
>
>> To me, voter satisfaction also includes some other elements: especially,
>> as we've debated, the question of "could I have done better by voting
>> differently?" And of course, Approval, as I've pointed out, is seriously
>> deficient in this regard.
>>
>
> This is simply irrational. I'll demonstrate.
>
> Say you have two choices:
>
> 1) Use a voting system (e.g. Score Voting) where you get a utility of 5 if
> you're sincere, or 7 if you're tactical.
>
> 2) Use a voting system (e.g. random ballot) where you get a utility of 3
> (tactical and sincere voting is the same)
>
> Now if you tell me that #2 is preferable because you didn't have some kind
> of suffering associated with the knowledge that you could have done better
> after the fact, then I point out to you that you are WRONG. Because you
> COULD have done better—if you had chosen to live in a world that used Score
> Voting.
>

I agree with Bruce in many regards.

I agree with Clay that Bruce is being irrational to focus on strategic
regret.

But Clay: why Score? The best system for Bayesian Regret is approval with
runoff. The best system for descriptive simplicity is approval. The best
system for voter and strategic simplicity and for fairness is SODA. The
best system for an evenly-weighted average of all of those is probably MAV.
Score is only best if you value BR and descriptive simplicity at about a
2:1 ratio. So OK, sure, it's technically on the Pareto frontier, but not on
a particularly convex part of it.

Jameson
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20130726/f92dd733/attachment-0003.htm>