Your body spends energy on two things: staying alive, and moving around. The first is your basal metabolic rate, and for most sedentary people it accounts for around two-thirds of the daily total. The second covers deliberate exercise, incidental movement, and the energy cost of digesting food itself.
Add them together and you get total daily energy expenditure, which is what people usually mean by "how many calories do I need". This calculator estimates the first with the Mifflin-St Jeor equation and the second with an activity multiplier.
What the equation actually does
Mifflin-St Jeor was published in 1990 and has since been validated as the most reliable of the common prediction equations for the general adult population. It looks like this:
| Variant | Formula |
|---|---|
| Male | (10 × kg) + (6.25 × cm) − (5 × age) + 5 |
| Female | (10 × kg) + (6.25 × cm) − (5 × age) − 161 |
It is a regression line fitted to measurements from several hundred people. That means it describes the average person of your height, weight, age and sex — and you are not the average person. Studies comparing it against indirect calorimetry, the lab method that measures resting metabolism directly, typically find it lands within about 10 per cent of the true value for most people. Ten per cent of 2,000 calories is 200 calories, which is not nothing.
The activity multiplier is the weaker half
If the BMR estimate is imperfect, the activity multiplier is worse. It compresses an enormous amount of individual variation — fidgeting, job type, commute, how much you move on rest days — into five buckets. Two people who both describe themselves as "moderately active" can differ by 500 calories a day.
The practical takeaway: treat the output as a hypothesis. Eat around it for two or three weeks, track weight in a consistent way, and let reality correct the estimate. Your own trend line beats any equation.
Activity levels, described honestly
| Level | What it looks like in practice | × |
|---|---|---|
| Sedentary | Desk work, driving, under 5,000 steps most days, no structured exercise | 1.20 |
| Lightly active | Desk work plus 1–3 sessions a week, or a job with regular walking | 1.375 |
| Moderately active | 3–5 real training sessions a week, generally on your feet | 1.55 |
| Very active | 6–7 sessions a week, or manual work plus training | 1.725 |
| Extremely active | Physical labour, or twice-daily training blocks | 1.90 |
The most common error is picking one level too high. Three gym visits a week is lightly to moderately active, not very active. A single hard hour does not offset fifteen sedentary ones.
Why calorie counting drifts
Two well-documented effects pull real-world results away from the arithmetic.
The first is reporting error. Studies using doubly labelled water — the gold standard for measuring real energy intake — consistently find people underestimate what they eat, often by 20 per cent or more, and it isn't deliberate. Oils, sauces, drinks and the handful of things eaten standing up are the usual culprits.
The second is adaptation. When intake drops, expenditure follows it down: resting rate falls a little, and incidental movement falls a lot, often without you noticing. This is why a deficit that worked in week two stops working in week ten, and why patience matters more than precision.
Where the calories come from still matters
Total energy drives weight change, but composition drives how you feel and what you keep. Two anchors are worth holding onto:
- Protein. Around 1.6 g per kilogram of bodyweight per day supports muscle retention, and it is the most satiating of the three macronutrients by a wide margin. This is the single most useful thing to get right.
- Fibre. Roughly 30 g a day, mostly from vegetables, legumes, fruit and whole grains. It does more for how full you feel per calorie than almost anything else.
Beyond those, the split between carbohydrate and fat is largely a matter of what you enjoy and can sustain. The best diet composition is the one you are still following in a year.
A note on very low intakes. Sustained intakes below roughly 1,200 calories for women or 1,500 for men make it difficult to meet basic micronutrient needs and should only be undertaken with medical supervision. If you find yourself drawn to numbers well below your estimate, or if food and weight take up more of your thinking than you'd like, that is worth talking to your GP about — it's a common experience and a treatable one.
Common questions
How many calories should I eat to lose weight?
A gradual reduction of 10 to 20 per cent below maintenance is the usual guidance, which for most adults works out to roughly 0.25 to 0.5 kg a week. Faster is not better: steeper deficits cost more muscle and are far harder to sustain. If you have a significant amount to lose, or any medical condition, plan it with a GP or registered dietitian rather than a calculator.
Does the equation work if I'm very muscular?
Less well. Mifflin-St Jeor uses total bodyweight, and muscle is more metabolically active than fat, so it tends to underestimate resting rate in lean, muscular people. The Katch-McArdle equation, which uses lean mass instead, fits better — but it requires an accurate body fat measurement.
Do I need to eat back exercise calories?
Not if you used an activity multiplier, because your training is already counted in the total. Adding tracked workout calories on top double-counts them, and wearable estimates of exercise burn are often high by a wide margin anyway.