> The other big issue is you tend to look at whatever patterns you find that are repeated, but you don't know why. To what extent? In my limited experience of

> The other big issue is you tend to look at whatever patterns you find that are repeated, but you don't know why. To what extent? In my limited experience of quantery, you want to have some idea of the mechanism behind an effect before you start trading on it. How much of the industry is building hypothesis-free models and trading on them?