No offence or insult taken.
I am ambivalent about the subject because I can see both sides of the
argument and I haven't seen any quantitative evidence that one or the other
is statistically better.

I think the post down the list about the army's cyclical including and
excluding full touch-downs says a lot. If there was a clear answer based on
evidence we would all be doing it. Until then there is a lot of gut feeeling
in it and nothing much to hang a hat on - one way or the other.
Guess it's like most things with some "damned if you do damned if you don't"
in it.

Interesting - about your comment about a "super instructor". In Oz that is
how it is - in a way.
As I recall from my student days (in the 90s) - our more junior instructors
can teach the students, the CFI (In Oz that's Chief Flying Instructor- the
super instructor) has to have interaction with each student on a regular
basis and he's (or she's) the one who signs off that you're ready for the
flight test etc. So it would be easy enough here to ensure that - say if you
were training in an R22 - you did power recoveries with the grade 2
instructors then did full downs with the boss-man. or something along those
lines.
Don't know how you system works in USA

Well doesn't that add to what I said about a machine that isn't safe for
training?

Maybe. Certainly it is a harder machine to train in and probably does exceed
the skill limits of some student pilots. But if UH-1s can exceed the skills
of some pilots then there's no hope for Robbies :-)

At some point we have to be practical and say "this is what we got - now
lets make the work we have to do with it safe as possible". If it were
legislated that R22's couldn't be used for training I would guess the costs
would go up so lots of people would be forced out when they would otherwise
accept the risk.