[EM] Lomax reply, 3/16/12

[MIKE OSSIPOFF]

Abd:

Regarding how AOC would look on the ballots, and how understandable it would be, I'd like to 
remind you that I don't suggest AOC as a first proposal. I suggest ordinary Approval. AOC could
be later added as an option. Various other options could be added too, maybe later, maybe together.
That question needn't be dealt with now.

The AOC option would be offered only if the C/D problem is already perceived. Voters wouldn't be new
to it when entering the voting booth, though there should be information there about it. And remember
that often, probably usually, a voter wouldn't need conditional appovals. Sometimes some people would
prefer to give some one or more candidate(s) a conditional approval.

It's useful to speak of primary and secondary C/D problems. Primary is the kind that a not defection-resistant
method has. The kind wherein defection consists of not supporting the other candidate. Defection-resistant
methods still retain a C/D problem of another kind. With defection-resistant rank methods, that consists of
burial. With the conditional methods, it consists of giving strategic insincere conditional support to candidates that
the voter wouldn't otherwise give any support to. I call that latter, 2nd-level kind of C/D problem, the kind
retained by defection-resistant methods, "Secondary C/D problem".

I was mistaken to say that the conditional methods' secondary C/D problem requires burial. But
the conditional methods complicate and alleviate the C/D problem--their secondary C/D problem.

Don't judge the conditional methods because they don't entirely get rid of C/D. IRV does, but at too high
a cost for public elections. I don't know of any FBC-complying method that doesn't have at least a secondary
C/D problem.

You were concerned about AOC's effect on SU. But, when AOC helps elect A, in the Approval bad-example (ABE), 
it's electing the sincere CW when there is one. Electing the sincere CW doesn't always maximize SU, but,
in general, it typically improves SU.

Now to reply to a few parts of your posting that I didn't get to yesterday. There won't be time to more than
start on this continued reply.

You asked how Bucklin handles the C/D problem. It doesn't. AOCBucklin does.

You mention runoffs. Runoffs can bring problems. Runoff doesn't bring FBC compliance
to methods that don't have it. Maybe could lose it in some methods that do. 
Runoffs about double the cost of an election.

Kevin pointed out why Approval-Runoff wouldn't be a very good proposal. Much
better just to propose the much more inexpensively implemented ordinary Approval.

You say that most people don't regard the Democrats as I do. I reply that, with the
enaction of Approval, the Democrats' illusory status, as favorite to a significant number
will evaporate.

Mike Ossipoff[ author ]
         
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20120316/041aeb6f/attachment-0002.htm>