Maybe the young men measured have low enough testosterone ranges that the high/low cycles have less effect, but the more likely explanation is that the methodology of this study is pretty laughable...
For starters. No information about the subjects other than their ages. I cannot see any information about their base testosterone levels either.