Of course there are. I said that in the last comment.
There are two points here:
* Estimates of "darknet" economies (and "criminal" economies in general) strongly express preferences for the mostly unfalsifiable LEO hypothesis that there's lots of crime just floating around out there, and they could do so much more about it if we just put up with a little more surveillance, etc.
* There's no particular evidence that there's a bimodal distribution between incompetent criminals who get caught and competent criminals who don't. There are probably lots of competent criminals who don't get caught, but there are also probably lots of incompetent ones who don't (and vice versa). The strongest predictor for successful interdiction (especially at borders) isn't competence, but sheer numbers: criminals have to succeed every time, cops only have to succeed once.