It might be plausible to assume that someone exploiting ESP or whatever to beat the house could keep the secret of how they're winning. But it's a lot less plau

It might be plausible to assume that someone exploiting ESP or whatever to beat the house could keep the secret of how they're winning. But it's a lot less plausible to postulate that they could keep secret the fact that they are beating the house or beating the market, unless the effect size is too small to be of much interest in the first place (ie. you have to largely abandon part of the original hypothesis: that the crazy phenomenon actually works).