I agree “BMI gives lots of false negatives compared to more reliable measures of overweight” is not the same thing as “BMI is more prone to false negatives than by-eye estimation” – it could be that BMI underestimates overweight, but by-eye estimation underestimates it even more. It would be great to see a study comparing both BMI and by-eye estimation to a third metric (I haven’t searched for this).
But if BMI is more prone to false negatives, and less prone to false positives, than most people think, that still seems to me like prima facie evidence against the claim that the opposite (that by-eye will underestimate relative to BMI) is true.
I agree “BMI gives lots of false negatives compared to more reliable measures of overweight” is not the same thing as “BMI is more prone to false negatives than by-eye estimation” – it could be that BMI underestimates overweight, but by-eye estimation underestimates it even more. It would be great to see a study comparing both BMI and by-eye estimation to a third metric (I haven’t searched for this).
But if BMI is more prone to false negatives, and less prone to false positives, than most people think, that still seems to me like prima facie evidence against the claim that the opposite (that by-eye will underestimate relative to BMI) is true.