[EM] Ballot Pairs Revisited by Donald

[Mike Ossipoff]

donald at mich.com writes:
> 
> Dear Methods List,
> 
>      I would like to thank everyone that read my post on Condorcet Pairs on
> the Ballot - and extra thanks to the ones that gave a reply.
> 
>      I will yield to the chorus - you were in good voice and all together -
> I concede that this would not be the best thing to do in a real election.
> 
>      Having said that I hold that this is a good thing to do for testing
> purposes. This method gives Condorcet results. If Condorcet is better than
> Run-off then it follows that Condorcet Pairs on the Ballot is better than
> Run-off.


Anything would be better than Instant Runoff. But asking voters to
vote on each possible pair of alternatives, even in a test,
would be a really lousy idea. That test would reveal that 
voters don't want to have to vote between 45 pairs when there
are 10 alternatives. It would be difficult to devise a sillier
way to introduce Condorcet's method to the public. I should
add that anyone presenting Condorcet's method to a voting 
group, such as the public, in the form of separate votes on
each possible pair would be a liar if he said that such balloting
is necessary to Condorcet's method, is part of Condorcet's method
or is proposed by any Condorcet's method proponents. The only
person to propose it is someone who has repeatedly & continually
demonstrated that he still doesn't know what Condorcet's method
is.

> 
>      We should be able to prove that Condorcet Pairs on the Ballot is
> better than Run-off? Do you think this can be proved without using
> rankings? Condorcet Pairs on the Ballot has no rankings.

As I said, Condorcet, regardless of the balloting details would
be incomparably better than Runoff or Instant Runoff. The point
here is that your balloting suggestion isn't any good. So of
course you'll keep on pushing it.

> 
> Donald,
> 
> 
> .-
> 


--