BMI is the most misused number in fitness. It is a simple ratio of weight to height that was built to describe populations, not individuals, and it says nothing about how much of your weight is muscle. Read as what it is, a rough population statistic, it has a use; read as a verdict on one body, it misleads, which is why a muscular lifter often reads overweight on it.
What BMI actually measures
BMI divides weight by height squared, producing a single number that correlates with body fat across a large population but not reliably within it. It cannot tell muscle from fat, so a lean, heavily-muscled person and an unfit person of the same height and weight get the same BMI, which is the whole reason it fails as an individual diagnosis.
Where it still has a use
As a population statistic BMI is genuinely useful: across thousands of people it tracks average body-fat trends and health risk well enough to be worth collecting. Reading your own BMI against the standard bands tells you roughly where the population would place you, which is informative context as long as you remember what it cannot see.
Why lifters read overweight
A trained lifter carries more muscle than the population average, and muscle is dense, so their weight-for-height is high without the fat BMI assumes. That is why the number flags many athletes as overweight, and why body-fat percentage and fat-free mass index, both on this site, are better individual measures than BMI for anyone who trains.
Questions people ask about bmi
Is BMI accurate for individuals?
No. BMI was built to describe populations and cannot tell muscle from fat, so it misjudges muscular and very lean people; body-fat percentage is a better individual measure.
Why does BMI say I am overweight when I lift?
Muscle is dense, so a trained lifter's weight-for-height is high without the fat BMI assumes, which is why it flags many athletes as overweight.
Is BMI useless then?
Not useless, just misused. Across a population it tracks average body-fat and risk trends well; the error is reading a population statistic as an individual diagnosis.