[EM] Approval vs Condorcet

Michael Ossipoff email9648742 at gmail.com
Thu Mar 21 21:58:34 PDT 2024


Robert—

Of course, in matters like Approval vs Condorcet, no one is wrong, because
different people have different valuations of the differing advantages &
disadvantages.

But the matter of one-person-one-vote (OPOV) is a sub-issue that is a more
objective exception.

To most everyone, OPOV is synonymous with the Plurality, Vote-For-1 (VF1),
voting-system.

…& yes I admit that, by that definition, Approval indeed violates OPOV.

…& I admit that Condorcet violates it too.

By that definition, OPOV most definitely should be violated.

But of course that isn’t how OPOV has been meant in electoral-law matters.
 …consisting mostly of district issues.

…& it certainly isn’t how OPOV is interpreted by us who discuss
voting-systems here.

It’s about not favoring anyone. Greater support /popularity can favor a
candidate or party, but the voting-system itself should not.

Here are some reasons why Approval doesn’t violate OPOV, as the
legal-system & we at EM have been interpreting it:

1. Every voter has exactly one ☝️ one vote for rating a particular
candidate as approved or not-approved.

2. No matter how many approvals you give, only one ☝️ of them can affect
the outcome.

3. Any voter can cancel out anyone else’s ballot.

e.g. There are 10 candidates. I approve 9 of them. You can cancel-out my
ballot by just approving the one that I didn’t approve.

I approved 9 times as many as you did, & you canceled out my ballot.
——-
But, other than OPOV, of course it’s a matter of individual subjective
preference.

You said you don’t want voters to have to use strategy. By strategy, you
mean anything other than indicating preference-order & letting the
rank-count do it all for us & shelter & insulate us from the choice.

I claim that there’s nothing wrong with voters doing it for themselves, &
that there’s nothing difficult about it.

In fact I claim that voting in Approval needn’t, & usually or always
doesn’t, resemble what could be called a “strategy” task.

How hard is it to mark the candidates you like?

,,,or some particular set (whatever one you feel like) as better than the
rest?

…or the ones you hope will win.

… or the ones you’d appoint instead of holding the election.

When you approve the candidates you like, you maximize the probability of
electing someone you like.

When you approve candidates at least equal to or better than expectation,
you raise your expectation.

Some of the abovementioned choices approve the candidates at or above
expectation.

So there are many ways of voting in Approval, & many ways of choosing how
to vote.

No that doesn’t make it complicated or hard. It just means that you have
great freedom to approve however you feel like.

Use whichever standard you like.

No dilemma.

It isn’t a dilemma. It’s a broad free choice determined by whom you feel
like approving, &/or how you feel like choosing.

A wide variety for whatever you like.

You’re not trying to guess the *objectively* optimal vote. (…the one that
would maximize your utility if the same situation were repeated many
times). That’s unknowable. …unknowable to the other voters too.  …so don’t
worry about it.

For me, because I regard some candidates as completely-unacceptable, it’s a
dichotomous, 2-valued, candidate-set. So approve all the acceptables & none
of the unacceptables.

This post is long already, & so I’ll post it as Part 1.

Part 2 will be along next.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20240321/730c8770/attachment-0001.htm>


More information about the Election-Methods mailing list