[EM] Meta-criteria 6 of 9: Heuristics. #1, simplicity

[Juho]

On May 10, 2010, at 5:08 AM, Kevin Venzke wrote:

> In particular I don't understand why you use the term "utility  
> function."
> Does this just mean "method"?

I used terms utility function and method so that
- utility function => what kind of winner would be good for the  
society and why is that winner good
- method => what kind of a procedure should one use to implement or  
approximate the chosen target utility function

>> Plurality criterion is a special criterion since it assumes
>> that ranked candidates are somehow approved. I think it
>> should not be used on methods that do not make such
>> assumptions. In margins pure rankings based approach seems
>> to be the normal way to read the ballots.
>
> The difficulty is that I think voters make assumptions more so than
> methods.
>
> Anyway, wasn't what I was trying to do, is explain the principle  
> behind
> WV? Are you going to reject anything I come up with because it doesn't
> work with margins?

Plurality criterion could be used to justify winning votes but it is  
less useful to analyze margins using that criterion. The intended  
message to voters should be clear. When should they not rank  
candidates. Does it mean "ranked equal" or "categorized as ... in  
order to influence the outcome so that ...". If there is a difference  
then the ballots that the voters give may have different meaning. What  
is the guidance to voters with respect to truncation when they walk to  
the polling station? In margins I'm aiming at request "just mark all  
your sincere preferences".

>>>>>> 10: A>B=C
>>>>>> 20: A>B>C
>>>>>> 16: A>C>B
>>>>>> 01: B>A=C
>>>>>> 01: B>A>C
>>>>>> 26: B>C>A
>>>>>> 03: C>A=B
>>>>>> 03: C>A>B
>>>>>> 20: C>B>A

>>> Why all this talk about B:C, why don't you talk about
>> A:B?
>>
>> Most methods and we two seem to agree that A should not
>> win. (There is the cycle still and that could maybe be used
>> to argue something else.)
>
> Yes, I'm talking about the A:B contest, not the idea that A should  
> win.
> If the public has such one-sided complaints (and I'm sure they will)  
> we
> are in trouble no matter what we put out there.

Ok, I missed your point there. There is a cycle and therefore all  
alternatives and pairwise comparisons should be checked. A wins B with  
one vote so there is also some ground to claim that B should not win.

>> This scenario is important to me since these votes could
>> well materialize in a real life election. Many other example
>> threat scenarios on the EM list are ones that are likely to
>> occur only in the minds of the election method experts but
>> this one could really happen in real life, and regular
>> voters and media could start wondering why the method failed
>> to see the obvious looking widely spread opinion that B is
>> better than C. (This could be a bit like IRV failing to
>> elect the Condorcet winner, although maybe not as obvious
>> case to argue about because of the involved cycle.)
>
> And A is preferred to B. Yes, the margin is different, but it doesn't
> sound like the critics you're talking about are clever enough to even
> see that. If we can't take criticism like that then we should give up
> now.

There is a cycle and one must decide which defeats are the weakest/ 
strongest. This is where margins and winning votes differ (cyclic  
cases). In this example the margins decision seemed more natural to  
me. (It could look better also to uneducated journalists.)

> You don't think margins' results are immune to criticism do you?  
> Without
> an autofill option or something I doubt margins as a proposal could
> even get off the ground.

What do you mean with autofill? Maybe to randomly generate full  
rankings. I think that voters should be encouraged to indicate all  
opinions that they may have, but allowing equal rankings (and not  
voting at all) is ok as well.

> But in general, in itself, truncating the worse of two
> frontrunners doesn't hurt anything.

One counterexample. Left wing has one moderate candidate and one  
extreme candidate. The extreme candidate has more first place support.  
If right wing voters truncate the moderate candidate (frontrunner  
together with the moderate candidate of the right wing in sincere  
opinions) and left wing has majority then the extreme left wing  
candidate may win.

Juho