I'm not making an empirical claim here. I am trying to explain why their results do not imply that "training to balance on one leg" causes good health outcomes.

I'm not making an empirical claim here. I am trying to explain why their results do not imply that "training to balance on one leg" causes good health outcomes. I'm saying that this experiment: 1. Does not define 'balance' 2. Does not show that the test they applied predicts or correlates with 'balance' 3. Does not show that 'balance' causes good health outcomes.