Unlike being better-than-the-market at predicting the value of some asset so far, being better-than-the-market at predicting H didn’t help people to increase their control over the market (until they’re no longer better than the market at that).
Unless we assume that [being better at predicting values of assets so far] correlates with [being better at predicting H] (compared with random LW users), I don’t see why we should expect the market to be better than random LW users at predicting H.
Unlike being better-than-the-market at predicting the value of some asset so far, being better-than-the-market at predicting H didn’t help people to increase their control over the market (until they’re no longer better than the market at that).
Unless we assume that [being better at predicting values of assets so far] correlates with [being better at predicting H] (compared with random LW users), I don’t see why we should expect the market to be better than random LW users at predicting H.