Somewhat related, I am puzzled that we still use BMI with a quadratic exponent for height. We are not flat creatures, but we also don't grow linearly either. The actual exponent is closer to 2.5 - this was even known to the creator of the BMI in the 19th century, but I guess that calculation was considered too tedious back then.
What it means is that people taller than the average are in fact less overweight than the BMI suggests, and shorter people more so. This mathematician suggests using 1.3*weight[kg]/height[m]^2.5 to align the corrected score with the established BMI scale: