I’d ignore the sample size. In this case, the largest predictor of wear is break-in and he has regularly cited this confounding factor in lab results. I was flabbergasted when he mentioned break in effect, then made the conclusion.
I was literally jaw dropped.