[EM] Anyone got a good analysis on limitations of approval andrange voting?

[Juho]

On Nov 8, 2009, at 8:43 PM, Raph Frank wrote:

> On Sun, Nov 8, 2009 at 5:22 PM, Terry Bouricius <terryb at burlingtontelecom.net 
> > wrote:
>> A somewhat more accessible (and available online for free) analysis  
>> of
>> strategic vulnerability of various methods is in this doctoral  
>> paper by
>> James Green-Armytage ("Strategic voting and Strategic Nomination:
>> Comparing seven election methods"). He found that Range and  
>> Approval were
>> just about the worst in terms of manipulability.
>> http://econ.ucsb.edu/graduate/PhDResearch/electionstrategy10b.pdf
>
> His assumptions about how people vote are not very realistic.

There are many small things in the paper that can be discussed but I  
think he gives a good solid basis for further discussions and  
improvements to make the results more accurate from real life  
modelling point of view. This is a good attempt to describe the  
practical properties of various methods from a neutral scientific  
point of view.

>
> Plurality -> vote for favourite
>
> Top 2 - run-off -> vote for favourite and then best of top 2
> Alternative vote -> honest ranking
> Minimax -> honest ranking
> Borda -> honest ranking
>
> Approval -> vote for better than mean

Two short comments on this one.
- One could alternatively assume that sincere voting in Approval would  
already be (at least partially) based on applying the basic strategy  
of trying to make a difference between the (two, three) most probable  
winners. Strategic voting would mean additional strategies on top of  
the basic one.
- Unfortunately the model that James Green-Armytage used was not yet  
able to handle strategic nominations in Approval. I think cloning has  
an impact in Approval (clones may easily harm you).

Juho


> Range -> give max to favourite and min to least favourite and scale  
> the rest
>
> His analysis is pretty interesting and he has created a condition for
> each method that is needed for the winner to be vulnerable to loss if
> a group of voters switched their vote.
>
> In practice, people are able to handle the basics of plurality  
> strategy.
>
> It would be interesting to see how approval scores with a pre-election
> poll being performed.  For example, each voter approves based on a
> mean threshold and then use the result of that election to work out
> the top-2.
> ----
> Election-Methods mailing list - see http://electorama.com/em for  
> list info