I used to train with a guy named Mike at the gym near my apartment in Denver. Mike is 5'10", 205 pounds, and has maybe 8% body fat. The man is a tank. He deadlifts 500 pounds for fun. One day we got talking about health metrics, and I pulled out my phone to show him my BMI tracker. I typed in his numbers out of curiosity.
His BMI came back at 29.4. One decimal point away from "obese." He looked at the screen, laughed, and said, "So I'm almost obese?" We both knew how ridiculous that was. You could see the guy's abs through his shirt. But according to the same formula doctors use to flag health risks, Mike was borderline obese.
That moment stuck with me because it exposed something I had sort of known but never fully internalized: BMI is a population-level screening tool, not a personal diagnostic. It was invented in the 1830s by a Belgian mathematician who wasn't even studying health. He was looking for patterns in human growth. The formula — weight divided by height squared — doesn't know anything about muscle, bone density, or body composition. It treats a pound of muscle and a pound of fat as identical.
The thing that really got me was when I looked up the actual history of BMI. It was created by Adolphe Quetelet, a Belgian astronomer and mathematician, in the 1830s. Not a doctor. Not a physiologist. A guy who studied stars and wanted to find statistical patterns in human populations. He was looking for the "average man" — literally his phrase — to help with social statistics. Health had nothing to do with it.
Fast forward two centuries, and insurance companies started using BMI in the early 1900s to set premiums. They needed a cheap, fast way to classify people into risk categories. BMI was perfect for that: one number, no expensive testing, works on paper. Then in the 1980s, the World Health Organization adopted it as a standard, and suddenly it became medical gospel. But the underlying math never changed. It still can't tell muscle from fat. It still assumes everyone has the same body composition. It still thinks Mike the deadlift monster is borderline obese.
I ran the numbers on myself just to see how silly it gets. At 5'10" and 176 pounds, my BMI is 25.3 — technically "overweight." But my waist is 33 inches, my body fat is around 18%, and I can run a 5K without dying. If I gained 10 pounds of pure muscle from lifting, my BMI would climb to 26.7, deeper into "overweight," despite being healthier. The formula would penalize me for getting stronger. That's not a health metric. That's a math error.
And don't even get me started on how BMI handles different ethnicities. Studies have shown that Asian populations tend to have higher body fat at lower BMIs, while Polynesian populations tend to have lower body fat at higher BMIs. The single threshold of 25 for overweight and 30 for obese doesn't account for any of this variation. It's a one-size-fits-all number in a world where bodies come in infinite configurations.
For the average person with average muscle mass, BMI works okay as a rough signal. But for athletes, weightlifters, or anyone who carries significant muscle, it's basically useless. Mike's "overweight" BMI wasn't a health risk. It was a math artifact. The formula couldn't distinguish between 205 pounds of muscle and 205 pounds of fat, so it just flagged him red and moved on.
I see this all the time in online forums. Some poor guy posts his stats, says he's been lifting for three years, and asks if his BMI of 27 means he needs to cut. The comments are a mess. Half the people say "BMI doesn't matter, look in the mirror," and the other half say "trust the science, you're overweight." Both sides are partially right and completely missing the point.
The real issue isn't that BMI is "wrong." It's that we keep applying it to people it was never designed to assess. If you're an athlete, your doctor should be looking at body fat percentage, waist circumference, blood markers, and performance metrics. Not a two-hundred-year-old ratio that thinks Mike the deadlift monster is almost obese.
I've stopped recommending BMI as a primary metric to my gym friends. I still track my own because I'm not an athlete — I'm a software engineer who sits ten hours a day — and for me, it's a useful rough guide. But context matters. A lot. The number on the screen means nothing without knowing what's actually inside the body it's measuring.
I think what frustrates me most is how BMI gets used in healthcare settings. I've heard stories of athletes being denied health insurance or charged higher premiums because their BMI flagged them as overweight. I've read about high school wrestlers being told they need to lose weight when they're already at dangerously low body fat percentages. The formula has real-world consequences, and those consequences fall hardest on the people it was never meant to assess.
The alternative metrics aren't perfect either, but they're better. Waist-to-height ratio accounts for abdominal fat, which is more predictive of health risk than total weight. Body fat percentage, measured properly, tells you what you actually want to know. Even just tracking waist circumference over time gives you more useful information than BMI ever could. These measurements require slightly more effort than stepping on a scale, but the accuracy improvement is worth it.
What metrics do you actually trust for tracking your own health?
At this point, I treat BMI like I treat the weather forecast — useful for broad trends, useless for specific decisions. I wouldn't cancel a picnic based on a 10% chance of rain, and I wouldn't tell an athlete they're unhealthy based on a BMI of 27. Context is everything. The number is just a starting point for a conversation, not the end of one.