[EM] Responses to some of Forest's ideas

[Forest Simmons]

On Wed, 18 Jul 2001, Richard Moore wrote:

<snip>

> 
> Note 1: Of course, the numerical solution for maximizing 
> utility expectation also shares the goal of minimizing 
> regret, assuming that utility and regret are opposites.
> But perhaps by "regret" we mean something that is not 
> quantified?
> 

I also believe that some kinds of regret are worth minimizing. In fact, I
once outlined a method for minimizing voter despair. 

Some kinds of regret are more worthy of our sympathies than others.

A man buys a thousand dollar cow from a widow for fifty dollars and
later regrets that he offered her so much when he finds out that she was
over a barrel. 

Another man obeys the speed limit all day and later regrets it when he
finds out the speed traps were not in operation that day. 

A man buys a ticket for the light rail and has it machine validated before
boarding. The ticket checker doesn't ask to see his ticket during the
entire ride.  The man regrets having bought the ticket.

A man wins a sealed bid to do a job for ten thousand dollars. He finds out
that the runner up bid twelve thousand dollars. He regrets that he didn't
bid eleven nine ninety nine.

Personally, I have little sympathy with these kinds of regrets.

The kind of regret Mr. LeGrand has been talking about (in the context of
CW versus Approval winner) is more or less similar to the kind of regret
illustrated by these examples (regretting one's inability to extract
strategic advantage due to lack of information about others preferences). 

Putting aside regrets and other emotions, let's think of results:

Approval does as well or better than Condorcet in zero information cases.

If the worst that can happen is for the Approval winner to be no better
than the CW due to voter strategy, what's to worry? 

Personally, I'm happy with the CW when one exists, but I like a method
that allows the possibility of improving on this, even if it cannot
always succeed, especially if that method has the CW as worst case (when
one exists).

I have a hard time understanding why people are afraid of the benign
strategic aspect of Approval.

Does anybody have an example in which reasonable strategy would tend to
make the Approval winner definitely less desirable than the Condorcet
Winner?

I haven't yet seen such an example. Or if I did, it didn't penetrate my
thick head ;-)

[Please include sincere approval and preference information in your
example, and remember, we're only talking about examples that have CW's.]


Forest