Over the weekend, NPR provided a Stanford mathematician to argue the invalidity of BMI as a rough measure of fatness. His basis of argument hinges on the idea that BMI measures are not a true indicator of fatness. Mathmatically minded people don’t like BMI because it is, admittedly, inexact. While this may be true to a point, the rough measure is important and here’s why.

First some history. In the mid-1800s, Belgian mathematician Lambert Adolphe Jacques Quételet was hard at work measuring social physics, or in simpler terms just determining normal measurements of people. The BMI equation itself is based on the distribution of data that he found. The result is a distribution curve with normals distributed around the mean. This has been born out over time and is used by health organizations from the CDC to WHO.

BMI = (weight in pounds x 703) / (height in inches x height in inches)

There are plenty of sites that will do the math for you.

This results in a number somewhere between 1 and 100. The actual results tend to occur between 18 and 45. Healthy BMI is generally considered to be between 18.5 and 25. There is undoubtedly some room for slop. But since the original equations are based on social normals, the common argument that professional athletes have comparatively high BMI doesn’t really wash. The equations are based on typical people, leading typical sedentary lives with relatively low percentages of muscle and high percentages of fat. Ultimately, even the CDC acknowledges this and allows for a measurement of “normal BMI for age” which allows for some normalcy outside the 18.5-25 range.

Athletes don’t measure their fat content this way and using them to argue against BMI is just ridiculous. Instead, there are a variety of ways to measure actual fat percentage with a high degree of precision. Whether rough estimating with fat fold calipers or high precision fat measuring using a DEXA scanner, there are ways to determine one’s body fat percentage. For those who are curious about this, the American Counsel on Exercise looks at percentage of body fat this way:

So here we are back to ranges again. How can we depend on ranges as normals? Medicine does it all the time. Every child who has ever been to a pediatrician is graphed on a chart that is based entirely on statistics measured years ago and their growth is compared to a normal curve. If the child is either too short or tall, or thin or fat (greater than 2 standard deviations from the norm) then physicians look seriously at why. Are their hormones ok? Are they too fat? That kind of thing.

BMI works the very same way. If your BMI is > 25, ask why. Is it because you have bulked up and leaned down but your muscle mass puts you outside the normal range? Or, more likely, is it that too many donuts or poorly consider lunches combined with too little exercise have tipped you into the statistical realm of “overweight”?

Dismissing an entirely valid tool for rough measurement because you don’t like the results is foolishness. Asking why the tool isn’t working for you is the right answer.

For the record, I tip the scales at a BMI of slightly over 25, therefore on the cusp of overweight. I also have a body fat percentage between 12.5% and 14% depending on measuring technique. I am also no longer someone whom BMI was intended to measure.