[EM] Comments on Approval posting

[Dave Ketchum]

On Sun, 23 Jan 2005 00:04:39 +0000 MIKE OSSIPOFF wrote:

> 
> 
>  I wasn't going to disagree with anything Russ posted, because I didn't 
> want to seem to be picking on him, or argumentative. But, after a few of 
> his postings, I finally decided that any posting that says misleading 
> things can fairly and properly  be commented on.
> 
> Russ said:
> 
> Like plurality, IRV tends to reach a stable equilibrium with two
> dominant parties.
> 
> I comment:
> 
> With Plurality, nearly any 2 parties can keep remaining the two top 
> votegetters at Myerson-Weber equilibrium. MW equilibrium is only defined 
> for point systems such as Plurality Approval, CR, & Borda, but it's also 
> true that IRV, lilke Plurality, would tend to stably protect the 
> perceived top-two, as Plurality does.

I claim rank systems (IRV and Condorcet) would:
     Do much like Plurality here.
     Not having Plurality's spoiler problem, react more gently as it 
becomes time to change.  If the 2 respond to the warnings, they remain 
top two; if not they get replaced for their sins.
> 
> Russ continued:
> 
> Approval, on the other hand, will not necessarily reach a stable
> equilibrium with only two dominant parties.
> 
> I comment:
> 
> ?With Approval, if there's a party that's always voter-median or CW, 
> there will be a stable equilibrium with _one_ party--the CW or voter 
> median party.
> 
> ERuss continued:
> 
> If I understand it
> correctly, it could allow any number of parties to reach a stable state
> of parity.
> 
> I reply:
> 
>  The party that is always CW or at the voter-median would keep winning 
> at MW equilibrium. Just that one party.
> 
> But it's true that Approval would fairly deal with any number of 
> parties. But it wouldn't include them all in a winning equilibrium, nor 
> should it.
> 
> Russ continued:
> 
> But I see a potential problem. If several parties reach parity,
> strategic voting becomes difficult, and the voter could end up with very
> difficult decisions. Do Greens approve the Democrat as a hedge against
> the Republican, or do they not? It's the same old problem, except now
> it's much more complicated.
> 
> I comment:
> 
> Same old problem, yes. It can be summarized by saying that Approval has 
> strategy to a greater extent than Condorcet does. That's the difference 
> between Approval & Condorcet.
> 
>  "Except that now it's much more complicated":   Much more complicated 
> than what? Than Plurality? It isn't much more complicated, and in fact 
> it isn't more complicated at all, to vote for the candidate that you'd 
> vote for in Plurality and also for those whom you like better, as 
> opposed to just voting in a Plurality election.
> 
> Russ continued:
> 
> Now Greens may need to worry about letting
> the Libertarian get elected if they don't vote for the Democrat.
> 
> I comment:
> 
> Again, this is nothing other than a re-statement of Approval's strategy 
> situiation, as it's always been discussed here.
> 
> I and some others have very recently posted here about strategies for 
> Approval. These strategies are ways of deciding the question that you're 
> asking. They're ways of dealing with the strategy situation that you 
> describe, the strategy situation that has always characterized Approval 
> voting.
> 
> Russ continued:
> 
> As a
> matter of fact, they may need to think about approving the *Republican*
> to prevent the *Libertartian* from getting elected! Or vice versa.
> 
> I comment:
> 
> Actually, a genuine libertarian isn't worse than a Republican. A genuine 
> libertarian is the opposite of a  Republican on authoritarian issues. 
> But that's a genuine libertarian, as opposed to an authoritarian.
> 
> Russ continued:
> 
> The bottom line is that the election could be tipped in any of several
> different directions depending on where voters decide to "draw the line"
> between their approved and disapproved candidates.
> 
> I comment:
> 
> Check the recent EM messages. There's been discussion about that very 
> question: Where to draw the line in Approval.
> 
> Russ continued:
> 
> Maybe this is
> obvious, but I wonder how many have really thought about the potential
> consequences.
> 
>  I comment:
> 
> Well,we've pretty much been discussing it quite a bit.
> 
> Russ continued:
> 
> I am not claiming that the disadvantages of Approval
> outweigh the advantages; I'm merely pointing out the potential problems.
> 
> I comment:
> 
> Condorcet has advantages over Approval, and many, for good reason, 
> prefer Condorcet to Approval.
> Approval has advantages over Condorcet, including some merit advantages 
> over Condorcet.
> 
> Approval also has disadvantages in its comparison with IRV. They have 
> already been pointed out, and much discussed on EM.  Of course discuss 
> them more if you want to , but they've already been pointed out. Or, of 
> course, point out new ones.

Said as if IRV is radically different from Condorcet.  Agreed that the 
methods of counting ballots differ in detail but, considering that they 
usually agree as to winner, are the result differences that big?  Note:
     IRV has a spoiler problem, though not as often as Plurality.
     Condorcet cycles can annoy when they occur.
> 
> Russ continued:
> 
> The Approval strategy "formulas" that have been put forth are
> interesting, but in many cases they will be of no practical use to
> voters. Why? Because they are based on how other voters are expected to
> vote, which is obviously uncertain. The best strategic cutoff point
> could depend critically on that uncertain information, in which case the
> formulas will be useless as a practical guide for voters.
> 
> I comment:
> 
>  Hardly. The fact that some Approval strategies are  based on estimates 
> of who will likely be the top-2 votegetters, or on other probabilities 
> (as opposed to certainties) doesn't lessen its usefulness as a practical 
> guide for voters.  Some Approval strategies require more to be estimated 
> than others do. Some require different things to be estimated. Check the 
> recent postings about Approval strategy.
> 
> Some ways of voting in Approval don't require probabilities to be 
> estimated. Those too are discussed in recent EM postings.
> 
>  It's obvious that if we only adopt for puiblic elections methods which 
> have been tested by use in public elections, we're pretty much stuck 
> with what is currently used in public elections. That's fine if you like 
> what's currently in use.
> 
> Anyone arguing that Approval or CR might be worse than the Plurality 
> currently in use would have to explain why Approval's or CR's added 
> freedoms, freedoms that they add to what's allowed in Plurality, would 
> make them worse than Plurality.

Arguments I see above look a bit painful:
     In EM it seems pretty well settled that we can and should find and 
promote something better than Plurality.
     If it is explainable to the public, something new is tolerable.
     BUT the voter's actions, such as strategy, have to be based on what 
is practical for voters to learn and use (it is too easy for EM members 
to design strategies that sound nice in EM debates, while not practical 
for public election voters to either get the data or process it).
> 
> Mike Ossipoff

-- 
  davek at clarityconnect.com    people.clarityconnect.com/webpages3/davek
  Dave Ketchum   108 Halstead Ave, Owego, NY  13827-1708   607-687-5026
            Do to no one what you would not want done to you.
                  If you want peace, work for justice.