One of my pet peeves is the idea that BMI provides an accurate indication of individual health. It doesn't. It's useful across populations (and may be useful to individuals to monitor progress), but when it comes to indicating which individuals are "healthier", BMI fails miserably - and our new Sciblings at Obesity Panacea do a great job of explaining why.
If, as a policy matter, we want to differentiate between the healthy and unhealthy - which is a big if, depending on how libertarian you are - let's make such assessments meaningfully, using a workout that proves general endurance and cardiovascular health. BMI is far too influenced by genetics/body type. I look forward to more solid posts like this from Obesity Panacea to counter popular misconceptions about weight and health.
BMI is a way for insurance companies to jack up your rates. My BMI (based solely on height and weight) is 30, which is considered obese. I am in no way obese, not even close. But when I went to get a new insurance policy, the application had a place to put your height and weight. They approved the application but said they had to add a bit to the premium because of my blatant obesity. If you saw me, you'd realize this is laughable.
The BMI calculators also say that for my height I should be 55 lbs lighter. There's no way I could ever achieve that weight and not look anorexic. I'd have to start a nice crystal meth addiction and start auditioning for Schindler's List.
I've never met anyone who thinks BMI is a indicator of "individual health", that would be entirely insane as being normal-weight is in no sense a guarantee of being healthy.
It's just meant to give you a rough idea what weight-range would tend to be normal for a person of a certain height. No more, and no less. It's a very aproximate thing anyway, with large individual variations.
It says if you're a fairly average male at 180cm, for most people the healthiest range is somewhere in the 60-85kg area. That's a large area !
And yeah sure, it's likely that since people are different, there exist people for whom the healthy range is a little more, or a little less than this.
But on the other hand, weight isn't everything, not even close, so if your weight is somewhere *close* to the ideal, it's very likely that worrying about your weight is pointless, as the negative health-effects of a small deviation from the optimal, are likely to be nil.