> We can assume that playing lacrosse was predictive. I am not going to assume that without further investigation. I am rather skeptical of the robustness of p

> We can assume that playing lacrosse was predictive. I am not going to assume that without further investigation. I am rather skeptical of the robustness of predictive models in the absence of relevant causal relationships.