[EM] Range Voting Strategy

[Juho]

On Nov 12, 2006, at 23:44 , Jobst Heitzig wrote:

> Juho wrote:
>> Some more proposals:
>>
>> 1) In E the average value was used to make the division to max and
>> min. Also 50% of the candidates getting max and 50% min could be one
>> strategy.
>
> I guess that's what Forest meant with "above median utility approval".

Yes, seems so. (I just replied without reading all the relevant mails.)

> Another question: Don't these results show that from the strategical
> point of view Range Voting is actually only Approval Voting with some
> added possibilities that only naive voters will use in the first  
> place?

Yes, Range voting easily becomes just Approval Voting. Simulations  
where all voters have same strategy don't show this very strongly yet  
but I'd expect simulations that have voters with different strategies  
would show more clearly that it pays off to be more strategic /  
"approval oriented" than voters in average are.

I think Range Voting is quite ok for opinion polls and corresponding  
fully non-contentious elections (e.g. www.imdb.com), but as soon as  
there is an element of competition Approval style voting is the  
likely outcome. Maybe we'll learn something about Approval and  
Approval style voting with the simulations.

Juho Laatu

> Jobst
>

Send instant messages to your online friends http://uk.messenger.yahoo.com