[EM] Range Voting vs Condorcet (Greg Nisbet)

[Raph Frank]

On Fri, Oct 17, 2008 at 5:24 PM, Jobst Heitzig <heitzig-j at web.de> wrote:
>> I know, but it does have randomness.
>
> I includes a chance process just as many sophisticated things in our life
> do. It does not include arbitrariness. It will most often lead to a certain
> winner (one option getting 100% winning probability).

I am not sure it would in practice.  It is likely that a few percent
would bullet vote.

I think having thresholds at both ends would be a good idea, i.e.
eliminate all options with less than 1/3 support and automatically
elect any option which achieves greater than 75% probability.

> Here's some evidence that the perceptions that chance processes are evil and
> that deterministic processes cannot lead to random results is wrong:

I think it is that random methods have the potential to be easier to
corrupt.  If someone with 5% support wins the draw, there is likely to
be many accusations of it being rigged.

Also, it could have stability problems.

> 1. Some time ago I challenged you all by asking for a method which elects C
> with certainty in the 55/45-example. The only methods which achieved this
> seeminly simple goal included a chance process.

I actually do think that it is a reasonable idea, but having it
implemented would be an uphill battle.

I think that a system that results in a 100% winner would be a
reasonable target unless, say more than 1/3 of the voters, refuse to
compromise would be a reasonable target.

Using it for something like a legislature where it is possible to
repeat votes is also potentially a problem, as a losing majority can
'toss the coin' over and over.

> 2. Every majoritarian method leads to a severe kind of randomness when
> there's no Condorcet Winner! This is because in all these situations there
> is no group strategy equilibrium, that is, whatever the winner is, there
> will be some majority having both the incentive and the means to change the
> winner to an option they like better. So, where the strategic process will
> end is mostly random since it cannot settle on an equilibrium.

I think it is likely that there would be an honest condorcet winners
in most real cases.  Also, the Smith set should contain candidates
that are at least reasonably similar and anyway, condorcet completion
methods are rarely random.

Btw, again, can you put on the web a full description of the method.
It would be helpful to be able to type "FAWRB" into google and see the
current version.

You have produced software that implements the method, so you should
include a description of the method it implements.