I mean, by your definition, Flub is literally every new language, modulo a few that are too domain-specific. Hell, Flub need not even be a powerful language, just a language with crackpot users.
So, rather than about being any averages, it appears to me that your point is just to not-so-subtly poke at anyone who thinks any language is better than the status quo (if so, then why hasn't it made an impact?!).
I don't see that as being a difficult question in general; there are many real-world problems, such as technological inertia and corporate culture, that already skew the chances of any new language.
However, in this case, where you've broadly directed this question at, well, everyone, I'm not sure if an answer would do you any good. What is true for Haskell may not be true for Rust (which is way too new to have made a big impact), which may not be true for any given crackpot language, which, to avoid offending anyone, I'll call Crud.
Aside from ruffling feathers, was there any point to this?
Unfortunately, the data point doesn't mean much; we tend to, collectively, reach a platform that's "good enough" and then just stop looking. So just because Algol 68 is no longer used does not mean C++ will eventually go the same way; indications are that it's a "good enough" language that will continue to see massive use.
More to the point, everything, or at least most things, written in C++ would have to be rewritten. This is enormous inertia compared to Algol 68.
>I was just pointing out that adoption of new languages is very much "Crossing the Chasm" - after the enthusiastic early adopters, the gap until the mainstream starts listening.
I don't get your point (not trying to be rude; I literally don't get how it ties in to your article).
Don't worry about it. The other definition of Flub I've seen [0], in Doug Hoyte's Let Over Lambda, comes from a similar place.
[0] "A language that can only thread code by subroutines." [In this case, "thread" has an idiosyncratic meaning.] Basically, anything other than Forth. Or those Lisps with the right sort of macros to re-implement Forth.