« Popularizing tiny effects | Main | Choosing reference levels »


Feed You can follow this conversation by subscribing to the comment feed for this post.

Joshua Jendza

The problem with BMI is that it was never intended as a measure of Obesity, and definitely not intended for diagnosis of individuals.

An anecdotal story of how the scale can mislabel someone as obese includes my older brother. He was an athelete in High School, lettered in baseball, wrestling and soccer. He was in peak physical condition, but when he applied to the Naval Academy in Annapolis, they made him take a body fat test because his BMI said he was Obese. His test came back with a single digit body fat score if I remember correctly. All sorts of professional athelets are considered Obese based on BMI, and that decreases peoples faith in it as a metric, and as an extention their faith in recommendations made solely on how they rank on that scale.

Now, I know next to nothing about this proposed new metric, but it's newness and cost do not necessarily preclude it from being potentially more informative than BMI. That simply remains to be seen.


Joshua: thanks for your comment. Bear in mind that obesity is an invented idea for which there is no objective truth, and that any metric will have false positives and false negatives. Also see point 1, the proposed metric has not been linked to any health risks while the old metric has. I'm not advocating one or the other metric here; I'm just observing that the discussion points to the many social problems of defining metrics. You're certainly right about the problems with BMI.

Cody L. Custis

BMI is worthless for athletes, because muscle is more dense than fat.

The Navy uses a formula that takes the circumference of the neck to the circumference of the waist to estimate body fat, which is more accurate. The problem is that not everyone knows the circumference of their neck.

A simple estimate (especially for males) would be to compare waist length to pant length, which is well-known, easily chartable, and actually useful.

Stuart Buck

he proposed metric has not been linked to any health risks while the old metric has.

If BMI is linked to health risks only because it is a (rather poor) proxy for bodyfat -- and there's every reason to think that's the case -- then a better proxy for bodyfat should be more closely related to those same health risks.

Tom West

Joshua Jendza's comment indicates the Navy knows that BMI gives false postives for obesity, but use it because it gives few false negatives. So they run on the basis that low BMI = not obese, high BMI = obesity unknown.


Kaiser, this is a great post.

Tom West's comment is a good one. Inevitably, in any simple measurement system there is a tradeoff between false positives and false negatives. The common objections about athletes' BMIs (including some above) miss the point that these false positives seem unlikely to lead to any negative outcomes -- it is vanishingly unlikely that anyone seeing Michael Jordan in his prime ever recommended he lose weight, even though his BMI was high (or, if they did, that he would do anything other than laugh).

The comments to this entry are closed.


Link to Principal Analytics Prep

See our curriculum, instructors. Apply.
Business analytics and data visualization expert. Author and Speaker. Founder of Principal Analytics Prep, MS Applied Analytics at Columbia. See my full bio.

Next Events

Oct: 31 Webinar on Data Visualization, online at JMP

Nov: 1 NYU unCOMMON Salon Public Lecture, New York, NY

Nov: 8 Tufts Gordon Institute: A Conversation with Kaiser Fung, Facebook Live

Nov: 8 Tufts TGI Careers & Networking Night panel, Somerville, MA

Nov: 26 Data Visualization New York Meetup, New York, NY

Nov: 27 NYPL Data Analytics Resume Workshop, New York, NY

Nov: 30 Purdue School of Engineering Seminar, West Lafayette, IN

Dec: 1 Purdue Mathematics, Data Science, and Industry Conference, West Lafayette, IN

Past Events

See here

Future Courses (New York)

Summer: Statistical Reasoning & Numbersense, Principal Analytics Prep (4 weeks)

Summer: Applied Analytics Frameworks & Methods, Columbia (6 weeks)

Junk Charts Blog

Link to junkcharts

Graphics design by Amanda Lee


  • only in Big Data