I see a fair share of woo out there that claims increases in sports performance or pain relief, and in comes the endorsements from celebrity sports stars. Or at least I think they're supposed to be sports stars. I don't keep up with that sort of thing, so I don't know who's good or not. Anyway...
I'd take any endorsement for health or performance claims from sportsmen, and it may have to deal with a stereotype I have: Sportsmen are superstitious. I've read some commentary on one bit of flawed thinking that comes easily to a lot of people, and even I'm vulnerable to it at times: "The Zone" It's that place where everything clicks and you do everything right. And it's because you were wearing your lucky shirt.
Actually, no. Streaks of good performance as well as bad occur because of statistical clustering: When you're gathering data of something essentially random (and there's a lot that can influence physical skills over long periods), you can expect there to be some points where highs and lows gather, rather than a completely uniform distribution. When one of those coincides with some new behavior or object, a baseless superstition can form.
I don't know how much is exaggerated by the media, but there seems to be a lot of that mode of thought in sports, with all sorts of unusual warmup rituals, lucky charms, and taboos. I've heard about it with native American "Medicine" that would influence their combat ability, and recently with the "mana" of tribal Hawaiian warriors. This sort of thinking is pretty well ingrained in our heads: Pattern recognition is a skill that was useful in our evolution, and the costs of being overzealous with it were low enough for us to survive despite the associated flaws.
That's why we need the introspection of science and in particular, statistical mathematics. We're no longer fighting for mere survival, but for goals that require a greater precision with increasingly subtle and intricate principles. In medicine, of course, you've got one major source of that intricacy: The human body.