[EM] Classifying 3-cand scenarios. LNHarm methods again.

[Dave Ketchum]

On Apr 18, 2010, at 6:39 PM, Kevin Venzke wrote:
> Hi Robert,
>
...

> I somewhat admire FPP incentives, as they do a pretty good job of
> delivering to the median voter two reasonable choices. A lot is
> accomplished with a simple method that mostly lacks potential for
> voter strategy.

There are many elections with only one reasonable choice - such as a  
good qualified worker trying for re-election.  Here even FPP would be  
fine, and we hope for nothing that makes voting unreasonably labor  
intensive.

The many with two reasonable choices are also doable with FPP, though  
needing TTR for help when losers prevent FPP from seeing a majority.

But, while many elections have no reason for more than the above  
complexity, we must not plan when setting up for an election that this  
particular one will fit.  For example the one worker mentioned above  
may no longer be available and there may be half a dozen wannabes.   
While TTRs can help FPP, it is worth some effort in picking a method  
to avoid the expense and time consumption of runoffs when possible.
>
> I think that a modest goal would be to have a method that provides
> incentive to coalesce behind three candidates.

While we should be prepared to serve such, I see no reason for trying  
for any particular count of serious candidates.
>
>
> The Burlington votes are inspiring. I'm amazed at how close the
> first preference counts were, and that a fourth candidate even got
> 15%. Unfortunately the resolution is so stereotypical you could
> think it was contrived to make a point.
>
> What worries me is the possibility that every time we succeed in
> implementing an election method which can handle any number of
> candidates that we throw at it, we will mostly see scenarios with
> one or two strong candidates and a half-dozen losers that never
> coalesced into anything, so that we mostly will not be able to tell
> the difference in effect from just using FPP.

This brings me back to Condorcet.  Bullet voting is no more painful to  
the voter than FPP, and counting is trivial extra effort - but it is  
ready when some, or even all, voters wish to do ranking.  It is the  
ability to do ranking when voters want this that makes it better than  
FPP.

Also Condorcet can survive having many candidates, while a voter truly  
desiring to vote more than 3 ranks should be rare enough that  
supporting such is debatable.

Dave Ketchum
>
> Kevin Venzke