This is a fair point, but I'd guess -- and I stand to be corrected -- that most of the later-round players who turn out to be better than replacement level still scored noticeably more in their draft years (in whatever level they were playing) than a lot of the guys who don't. Quite apart from the few (but by no means unicornish) ones who actually turn out to be stars.
Absolutely, this is generally somewhat true, although some weird stuff happens.
And I can't stress this enough : the USHL is *weird*.
Like, I've followed the CHL my whole life. I've been a WHL season ticket holder. I know those leagues inside-out and I understand well what I'm watching and what the stats mean and what tends to project and what doesn't. But the USHL I have no clue. It doesn't make sense. Guys like Aidan McDonough and Adam Gaudette were unpopular picks here and their production at their respective ages in the WHL or OHL simply wouldn't project at all. And the USHL is theoretically a worse league, but both guys ended up in the NHL. And that happens a lot in that league - we were discussing the Colton/Koepke/Perbix TB trio earlier and these were all OA/double OA guys with seemingly very iffy production in that league who reached the NHL, and two of them are legitimately good NHL players.
I think that the reason is that for non-elite/middling prospects the USHL->NCAA age 19-23->AHL is a more realistic/attainable progression that allows guys to keep developing year over year and keep improving for many years whereas the CHL->massive jump to the AHL and get destroyed thing just ends the development and progression of a lot of kids. Or they go to CIS and they might as well be playing on the moon at that point.
So I've seen this movie before and learned from it and I guess I don't have the same reaction when we pick a middling-producing USHL guy, because it seems inevitable that that player will be a pretty dynamic NCAA scorer in 3-4 years.