Why do TDEE calculators give you different numbers?

Most calculators are built on one of three equations: Mifflin-St Jeor, Harris-Benedict, or Katch-McArdle. Mifflin-St Jeor, published in 1990 and used here, is the most common modern default. Harris-Benedict dates from 1919 and was revised in 1984, and tends to read slightly higher. Katch-McArdle works from lean body mass instead of total weight, so it needs a body-fat percentage you probably do not have to hand.

For the same person, the gap between Mifflin-St Jeor and Harris-Benedict is usually in the region of 50–80 kcal. Real, but small and nowhere near enough to explain a calculator telling you 2,200 when another said 2,900.

This is where the large gaps come from. “Moderately active” is not a defined quantity; it is a label each site attaches to a number of its own choosing. One calculator may call moderate 1.55, another 1.46, another 1.6. Some sites use four buckets, some six, some ask for training days per week and derive a factor from that.

Because the multiplier acts on the whole BMR, a small difference in the factor produces a large difference in the result. On a BMR of 1,639 kcal, moving from 1.55 to 1.725 adds 287 kcal, more than three times the entire spread between the two BMR equations above.

The most common cause of a mismatch is not the calculator at all. Two sites describe their activity levels in different words, and the honest answer to “moderately active” on one phrasing is “lightly active” on another. If one site anchors on exercise sessions per week and another on how you spend your working day, a desk worker who trains four times a week can legitimately land in different buckets.

Before concluding that two calculators disagree, check that you gave them the same answer. It is worth being deliberate about the activity level, because it is the single input with the most leverage over the result.

A calculator that rounds BMR to the nearest 10 before multiplying by 1.9 carries that rounding into the final figure, magnified. It is a small effect, usually under 20 kcal, but it is why two sites using identical formulas and identical multipliers can still show figures a few calories apart.

FitQuota rounds only for display. Every downstream multiplication uses the unrounded value, which is also why the BMR shown on the BMR calculator, the TDEE calculator and the calorie calculator agree with each other rather than drifting apart by a few kcal per step.

None of them, exactly, and that is not a dodge. Every one of these calculators is a prediction equation fitted to population averages. It tells you what a typical person with your height, weight, age and sex tends to burn. It cannot know your body composition, your genetics, your sleep, your medication, or how much you actually move on a Wednesday.

The useful approach is to treat any estimate as a hypothesis with a test attached. Pick one calculator, eat near its number consistently for two to three weeks, and watch the weight trend. If you are holding steady, that intake is your real maintenance level which is better information than any equation can give you, because it was measured on you rather than predicted from a sample.

A useful sanity check: if two calculators are within about 200 kcal of each other, the difference is noise relative to the error in the method itself. Pick either and move on.