That's certainly an interesting way to asess safety. The implicit
assumption here is that it's OK to have more fatalities if you draw
sufficiently bigger crowds, as this will make your death/spectator
ratio lower. ...snip...

That was not the purpose of the comparison.

But it's a necessary assumption for the comparison to be meaningful.
So do you agree with it or don't you?

The purpose of the comparison was to indicate that their safety record
is
similar if not better than other aerobatic teams.

But it doesn't show that unless you accept the above assumption. For
all you know, the other teams have way fewer accidents but draw smaller
crowds (not that this is the case, but based on the numbers you quoted
it could be) or have way more accidents but draw larger crowds.

This made me wonder about
the safety record of "private" aerobatic teams, for which little data
is
available.

On the contrary - both in the US and in Canada, every accident
involving a private aerobatic team member is public knowledge, and is
investigated. It's just that no private aerobatic team ever racked up
five. Government teams, on the other hand...

And it also made me wonder whether it was fair to compare the
record of aerobatic teams with complex shows... whether private or
not....
against GA statistics.

Depends what you mean by fair. If you mean that the flying involved is
more demanding and thus can be expected to result in more accidents -
well, I agree. The same can be said of any accident that begins with
"Hey, y'all, watch this." The only difference is the size of the
audience.

As for government funding.... as I said before, governments fund many
other
way-less-useful projects

I'm not in favor of them either, and I don't think that's a good
argument.

Michael