Aboslutely. I'll admit to not having read the actual study, but saying that 33% of pilots improved is fairly meaningless. You could also say that 67% of pilots

Aboslutely. I'll admit to not having read the actual study, but saying that 33% of pilots improved is fairly meaningless. You could also say that 67% of pilots didn't improve. It might even be the case that those 67% that didn't improve had half of them get worse. In which case we'd have 1/3 improve, 1/3 no change and 1/3 get worse. Or in other words, a boring average distribution that means absolutely nothing...