A-HA! We all lecture in our classes about the perils of naming too many species, but now the facts have been statistically proven! Well, at least for dinosaurs:
The more fossil species you describe, the less likely the names are to stick.
...
"I would have expected that more prolific, experienced authors might be better at recognizing genuinely new species, yet they were less successful than authors who name only a few dinosaurs," says [paleontologist Michael] Benton. "It is hard, and maybe impossible, to construct a case that experience in naming dinosaurs makes one better at the job."
OK, dinosaurs are a much bigger sample than hominins. But our rate of discard is actually much higher -- we must still be using only 10 or 20 percent of the names that have been proposed, and probably half the ones we're using now will bite the dust. By contrast, more than half the dinosaur names are still accepted.
It's the failure of the "prolific" taxonomists that deserves to be cited again and again. Oh, they got it right 41% of the time, and a lot of the sunken names date to before 1950, but the phenomenon has not abated with names of the last 50 years. Prolific namers seem to have a lower threshold for erecting a new taxon, and they get it wrong more than half the time.