[EM] wv criticisms for LNHe failure. The "group opinion". Frequency of failing a criterion.

[Michael Ossipoff]

On 1/30/2013 2:21 PM, Michael Ossipoff wrote:
> ...
> For instance, the LNHe failure of such traditional unimproved
> Condorcet (TUC) methods, such as Beatpath, Ranked-Pairs, etc. is
> admitted by most to be a disadvantage.

[endquote]

Richard says:

To anyone here who is isn't already aware, Michael Ossipoff makes
statements about what other participants here believe, yet frequently
those statements do not reflect what participants here actually believe.

To repeat what has been said here many times, different election-method
experts have different opinions about which criterion failures
("disadvantages") are more important than other criterion failures.

In other words, Michael's opinions about which criteria are more
important than others does not reflect the group's opinions.

[endquote]

Strategic need to rank all the candidates, and random-fill incentive,
have, on EM, long been criticisms of the the winning-votes (wv)
traditional unimproved Condorcet (TUC) methods.

Because of Richard's relatively much shorter experience with EM, I
forgive him for being unaware of that.

Sure, obviously TUC(wv) advocates must think that a strategic need to
rank all the candidates, and random-fill incentive are ok, because
otherwise they wouldn't be TUC(wv) advocates.

But I've never heard one of them cite random-fill incentive as an
advantage :-)  or claim that it isn't undesirable. Rather, they
advocate TUC(wv) in spite of it, because they've chosen to accept it,
as a price of TUC(wv).

I was there. Richard wasn't.

That we have differing opinions on the importance of different
criteria (and usually an unwillingness to answer regarding why we
claim a criterion is important or isn't), is quite irrelevant to my
statement.

As for whether my statements about the relative importance of criteria
reflect the group opinions, Richard's meaning, in that statement,t is
unclear (but it isn't the first time for that).

Is he saying that my statements don't reflect _all_ of the group's
opinions? How could they? The group's opinions differ :-)

Is he saying that my statements don't reflect _any_ of the group's
opinions? So Richard doesn't think that anyone here would agree that,
when there are 25 candidates, it would be nice to not have strategic
need to rank all of them, even rank them randomly if you have no
preference between them, even if most are unacceptable to you?

So what did Richard mean? Who knows? He'll never tell :-)

Aside from that, it may come as a shock to Richard that I'm not in the
least interested in duplicating or copying the "group opinion" (if
there is such a thing). Sorry Richard, I've never been much of a
group-copier or a follower of fashion. That's why I might sometimes
say things here that run afoul of current local fashion at EM. I make
no apology for that.

But I don't criticize Richard's role as a watchdog, to bark when he
perceives that someone violates group opinion

Richard says:

As a more advanced clarification, research has not yet been done to
identify _how_ _often_ each method fails each criterion.

[endquote]

1. Later-No-Help failure isn't a sometimes thing. Typically, methods
that fail LNHe consistently, not sporadically or occasionally,  give
strategic incentive to rank additional candidates, to help
higher-ranked ones. Ordinarily, that will take the form of strategic
incentive to rank all of the candidates, including the unacceptable
ones.

2. When I made my comment, I was specifically replying to someone who
had said that not failing LNHe is a source of problem.

Even if it meant something to speak of how often a method fails LNHe,
and even if a method didn't fail it often, a discussion of how often
methods fail LNHe wouldn't be relevant to my above-quoted reply.

I've amply explained (again and again) why the strategy need resulting
from FBC failure is an all-the-time problem. No, I'm not going to
repeat the explanation again for Richard.

Michael Ossipoff