[EM] Reply to Markus' Approval criticism.

MIKE OSSIPOFF nkklrp at hotmail.com
Sun Dec 3 13:21:01 PST 2000


In another message today, I apologized that I misunderstood what
Markus meant when he criticized Approval. Now I'll say what he
did mean, and I'll answer that:

>When this information isn't exact, [the voter] will probably approve
>all potential winners or disapprove all potential winners so that
>his vote has no influence on the election result.

I now realize that Markus meant that the voter will often
_accidentally_ vote for both or neither of the 2 frontrunners.


He said "Is this really what you want?"

Ok, Markus, what you're saying is that, because Approval doesn't
let us vote all of our pairwise preferences, we sometimes won't
vote our preference between the two frontrunners. And you're saying
that therefore you'd rather have a method that lets us sincerely vote
all of our pairwise preferences. What you're saying is that you want
a method that complies with GSFC, or at least SFC. Those criteria
are about letting us sincerely express all of our pairwise preferences,
when that goal is attainable.

As you know, I quite agree that it's good to comply with GSFC, or
at least SFC. That's why I say that I'd prefer Condorcet to Approval.
I've always made that clear.

But IRV, while it lets us vote all of our pairwise preferences,
will often not count them. And IRV doesn't let us sincerely vote
all of our pairwise preferences. It will often force us to reverse
preferences, even to the extent of dumping out favorite.

Your criticism of Approval for not being Condorcet is quite unfair,
and it's one that we often hear. You're asking too much of Approval.
When you criticize it for not letting us vote all of our sincere
pairwise preferences, you're criticizing it for not meeting SFC or
GSFC. You're criticizing it for not being Condorcet. But compare
Approval to the methods other than Condorcet: It does better than all
of those by the defensive strategy criteria. For instance, Approval
meets FBC & WDSC, and IRV fails both, with nothing to offer in return.

Mike Ossipoff





_____________________________________________________________________________________
Get more from the Web.  FREE MSN Explorer download : http://explorer.msn.com



More information about the Election-Methods mailing list