Manually counting IRV (was RE: Truncated preferences ok for Condorcet (was Re: [EM] Ignorance))

[Steve Eppley]

James Gilmore wrote:
> Steve Eppley wrote:
> > The third explanation is a hold-over from elections that 
> > tally Instant Runoff (or the proportional representation 
> > version, Single Transferable Vote) by hand.  To quickly 
> > tally Instant Runoff by hand: 
> > 
> >    Distribute the ballots into piles, each according to
> >    its top-ranked candidate.  From the height of each pile,
> >    you can see at a glance which pile has the fewest
> >    ballots.  So, if there are more than two piles, take 
> >    the shortest pile, eliminate its candidate, and
> >    redistribute its ballots to other piles according to
> >    their highest-ranked non-eliminated candidate. 
> >    Repeatedly eliminate & redistribute the shortest pile 
> >    in this way until only two piles remain and elect the
> >    candidate of the taller pile.
> 
> I don't know where you found this description of IRV, but it could
> result in a lot more paper shuffling than necessary.  The piles
> should be counted at every stage and papers transferred only so
> long as necessary, ie until some one candidate has as many votes
> or more votes than all of those remaining.  That is often achieved
> before all but two have been eliminated. James 

That's true, except there's no need to count the ballots in 
each pile.  It's quicker to compare the heights of each 
pile by eye to see which is shortest, unless the shortest 
piles are very similar to each other in height, in which 
case you might want to count the shortest piles if the 
group is concerned with accuracy.  Some were and some 
weren't concerned with accuracy in Florida 2000.  
Personally, I'm not so concerned.  Minor inaccuracies could 
elect a so-called "second best" alternative, but the 
incentives for candidates to try to be the best compromise 
and for incumbents to be accountable to the public would 
still be intact (assuming a voting method much better than 
IRV) and that's what I do care about.

To decide when a pile is tall enough to be more than half 
of the votes (at which point you could stop and declare a 
winner, as James rightly pointed out) you'd want to add the 
heights of every pile once, at the beginning, and construct 
a reference stick which is half that height, to compare 
periodically with the tallest pile.

I hope this is all moot.  Aren't we living in the computer 
age yet?

---Steve     (Steve Eppley    seppley at alumni.caltech.edu)