I just read the referenced article, and while its conclusions are
disturbing -- primarily the release failure of the tow plane's
Schweizer hook -- the experiment itself was not well conducted.

First and foremost, identical circumstances were not created to test
the difference between nose hook and cg hook. The K-13 with nose hook
was taken "progressively higher..." while the same aircraft with cg
hook was "pitched up."

As noted in an earlier post, I think all of us are in agreement that a
nose hook is (qualitatively) a safer proposition, but if you're going
to raise an alarm, it should be valid and proportionate.

Chris Rollings has presented us with a compelling demonstration of
what can happen during aerotow. The conclusions based on outcomes for
the two release types, since the maneuvers were not identical, has
been exptrapolated and are weighted to some degree by the knowledge
that the nose hook is preferred. A good next step would be to repeat
the experiment, this time putting emphasis on measuring the loads and
effects of identical maneuvers using a nose vs. cg hook.

In the absence of valid statistical or empirical evidence, it's hard
to determine just how critical this problem is. I think most of us are
willing to spend dollars on our well-being, but we'd like to know that
we are, in fact, purchasing something of value. If the record shows
that we're simply buying better handling as opposed to a measurable
increase in safety, then we are looking at a much different
proposition.

Point of reference, I only have a nose hook in my Ventus 2. I am
perfectly content to aerotow gliders with a cg hook if there is no
other choice. And frankly, I'd rather aerotow on a cg hook that winch
launch, which I consider an unreasonable comprimise to safety. It's
all a matter of what you're used to. No tow plane? let's unwind the
winch. But the risk to reward equation is becoming less attractive.