What a formula calculator can and cannot tell you
Every tool in this category implements a published equation and nothing more. That is a narrower claim than most fitness calculators make, and stating it plainly is the point.
A formula fitted to a population describes a population. Applied to one person it gives a central estimate with real uncertainty around it, and the uncertainty is not small: individual resting metabolic rates vary by ten percent or more in either direction from what an equation predicts, and body composition varies enormously between two people with the same height and weight.
That does not make the numbers useless. It makes them a starting point to be adjusted by observation rather than a target to be obeyed. The honest use of a maintenance-calories estimate is as an opening hypothesis you revise against what actually happens over several weeks.
None of this is medical, dietary or coaching advice. These pages contain no diagnosis, no prescribed targets and no claims about disease risk. If you have a health condition, are pregnant, are under 18, or are considering a meaningful change to what you eat or how you train, that is a conversation for a qualified clinician or a registered dietitian who can assess your individual situation.
Where the equations come from
Body mass index is weight divided by height squared, devised in the nineteenth century as a population statistic and adopted much later as a clinical screening shorthand. It was never designed to assess an individual, and it cannot: it does not distinguish muscle from fat, says nothing about where fat is distributed, and reads differently across ages, builds and ancestral populations. It is a ratio, not a verdict.
Mifflin-St Jeor estimates basal metabolic rate from weight, height, age and sex, and is among the better-validated equations of its kind at the population level. Multiplying it by an activity factor produces a rough total daily energy expenditure -- and that multiplier is the least reliable step, because self-reported activity is consistently overstated.
Riegel's formula projects a race time at one distance from a time at another using a fatigue exponent of 1.06. It holds up reasonably for a doubling of distance and grows steadily more optimistic beyond that, which is why first marathon predictions from a 10 km time so often disappoint.
Reading the output
Treat every figure here as a range rather than a point. A BMI of 26.4 and one of 24.6 are not meaningfully different measurements of anything. A maintenance estimate of 2,410 calories means something closer to "somewhere in the low-to-mid two thousands". A predicted marathon time is a statistical extrapolation that knows nothing about the course, the weather, or whether the training behind it happened.
The arithmetic is exact. What it is measuring is not.