I believe I understand your point, but there these options (as I see things): 1. There is little correlation, but there is actual causation (e.g: a cricle) 2.

I believe I understand your point, but there these options (as I see things): 1. There is little correlation, but there is actual causation (e.g: a cricle) 2. There is correlation, but it turns out to be a fluke later during validation. So, yes. It suggests but in both cases I'd rather continue checkcing