[EM] Venzke's election simulations

[Kevin Venzke]

Hi Warren,

--- En date de : Mer 9.6.10, Warren Smith <warren.wds at gmail.com> a écrit :
> I admit "reasonable behavior" is somewhat subjective and
> depends on the
> scenarios you are setting up... you've inherently
> introduced a length
> scale the minute you actually set up any scenario..., but
> you can try
> a few scenarios and make sure to use A,B that do something
> reasonably
> sane in them, given the length scale you chose.

I agree that is unavoidable to implement *something*. But if I ignore
an issue like this and just use the simplest thing possible, it looks
less like cherry picking than if I decide what I think performs reasonably
well and use that.

> And by the way, there really IS a length scale, set by the
> parameters
> defining "a human life." I contend human utility simply is
> not
> scale-free.

I think you must be right. Though exploring this issue adds more elements
when there are already a lot.

> IEVS by the way uses my f-formula and also has Lp distance
> with
> user-choosable p.
> I'd recommend Venzke also do that, plus if he wanted to
> keep his
> present thing he could rename it "expected distance" rather
> than
> "utility."

Well, if I become picky about what models can be called "utility" I
will probably be conservative and say none of them can.

> >Venzke:
> It's pretty clear to me that if you just toss out
> candidates randomly,
> RangeNS will usually win. It just happens that in the
> scenarios I pick
> out as being of interest to me, RangeNS isn't usually
> winning. So I would
> like to investigate this to find exactly what are the
> circumstances that
> cause methods like Bucklin or DAC to prevail.
> 
> --well, if you can somehow understand what is "interesting"
> and bias
> your distribution toward interestingness (and somehow
> justify that
> biasing as being realistic, otherwise it is just data
> fudging) then
> you might find RangeNS is no longer best.  I cannot
> say at present.
> I still think at some point you have to settle on some
> distribution
> also you can make a recommendation of some voting system.

I think you misunderstand me. I am not sure it will be possible to
settle on a distribution, or make a recommendation, and that doesn't
bother me. I really just want to describe why/where there is variation.

It seems to me quite possible that the numbers just aren't different
enough to want to make a recommendation. The utility doesn't take some
things into consideration (e.g. the public's opinion of how understandable
or fair the method is) and there are many assumptions being made (e.g.
regarding polling, voter distribution, nature of utility, perception of
strategic incentive) that could be flawed.

Certain methods' results give me enough anxiety to recommend *against*
them though...

> But anyway, glad to hear Venzke's sort-of-endorsement of
> range voting.

Well, I fully endorse sincerely voted normalized Range when three
candidates are nominated at random, but I will have to be careful and
at this time only endorse it for the case that there are only two issues.

Kevin Venzke