Getting an estimate that is about you rather than about a 70 kg man
1. Choose the activity you actually did
Each entry carries its own MET value from the Compendium of Physical Activities, the reference catalogue researchers use to put a number on the energy cost of human movement. Pick the closest match rather than the most flattering one — walking at 5 km/h and walking briskly uphill are genuinely different activities and the list treats them that way.
2. Enter how long you did it for
Actual moving minutes, not elapsed time at the gym. The rest between sets, the queue for a machine and the walk to the changing room are not part of the activity you selected, and including them is the single easiest way to double an estimate without noticing.
3. Enter your weight
Energy cost scales almost directly with the mass you are moving, which is why a heavier person burns more doing exactly the same thing. This is the input with the largest effect on the answer.
4. Add height, age and sex
These are what let the tool stop assuming you are the man the MET convention was built on. With them, the estimate is scaled to your own resting metabolic rate rather than to a fixed constant, and the page tells you how far the two answers differ. For some people that gap is small; for others it is more than half the total.
5. Read the net figure first
The large number is the energy the activity cost you beyond what you would have burned anyway sitting still for the same period. The gross figure is shown next to it. Which one you want depends entirely on what you are doing with it, and the next section explains why.
6. See what the session counts toward
The activity is classified as moderate or vigorous from its own MET value and set against the WHO and CDC weekly targets, so a thirty-minute session becomes a proportion of something rather than a floating number with no context.
Net versus gross, and why eating it back quietly ruins a calorie plan
Every calories-burned calculator faces the same question and most answer it silently. Do you report the total energy your body used during the activity, or only the extra energy the activity caused?
Your body burns calories continuously. Sitting on the sofa for thirty minutes costs something — typically thirty to forty calories for an average adult. If you spend those thirty minutes walking briskly instead and the gross cost is 160 calories, the walk did not add 160. It added about 125. The other 35 you were going to spend regardless.
On a brisk half-hour walk that overlap is roughly twenty per cent of the headline. For gentler activities it is a larger share still. And it compounds: if you eat back what a tracker or calculator tells you an activity burned, you are systematically eating back energy you never actually spent. This is one of the most common reasons a carefully planned deficit stops producing results while the person following it does everything they were told.
The previous version of this page knew about this. Its own footnote read that the figures “include the energy you would have burned at rest during that time” — and then displayed the gross number in fifty-two-point type as the answer. Naming a problem in small print underneath the number that has the problem is not the same as fixing it. The net figure now leads, with gross alongside it.
Use the net figure when you are reconciling activity against a calorie target. Use the gross figure when you want to know the total metabolic cost of a session, for instance comparing two workouts against each other.
The MET convention was derived from one man in his forties
The standard formula is MET × 3.5 × weight in kilograms ÷ 200 calories per minute. The 3.5 is where the trouble lives.
One MET is defined as the energy cost of sitting quietly, conventionally set at 3.5 millilitres of oxygen per kilogram of body weight per minute, or equivalently 1 kcal per kilogram per hour. Byrne and colleagues traced the origin of that figure in the Journal of Applied Physiology in September 2005: it was first derived from the resting oxygen consumption of a single individual — a 70 kg, 40-year-old man.
They then measured resting metabolic rate in 769 people: 642 women and 127 men, aged 18 to 74, weighing between 35 and 186 kg. Measured resting rates averaged closer to 2.6 ml per kilogram per minute than 3.5. The paper’s title says it plainly: the metabolic equivalent is one size that does not fit all.
The practical consequence is that the standard formula overestimates for most people, and overestimates most for the people furthest from that original reference man. Measured against each individual’s own resting rate estimated with the Mifflin-St Jeor equation, the conventional figure runs about seven per cent high for a 70 kg man of 30 — close to the reference, so close to right. For a 90 kg woman of 55 it runs 55 per cent high. For a 110 kg woman of 65, 65 per cent high.
Those are not edge cases. They describe a large share of the people who look up a calories-burned calculator in the first place, and they are being handed a number that is more than half again what they actually spent. Entering height, age and sex here switches the calculation onto your own estimated resting rate, and the page reports how far that sits from the textbook figure so you can see the size of the correction rather than take it on faith.
Where MET values come from
The MET values behind each activity are drawn from the Compendium of Physical Activities, a reference catalogue first published by Ainsworth and colleagues in 1993 and revised several times since, which assigns an energy cost to several hundred specific human activities. It is the standard source in physical activity epidemiology, and it is what allows “cycling” to be split sensibly into leisurely, moderate and vigorous rather than treated as one thing.
Two caveats are worth carrying. The values are population averages measured in study conditions, so an efficient, well-trained mover may burn less than the table suggests for the same output while a beginner burns more. And they describe steady-state activity; interval work, stop-start sports and anything with long rests are less well captured by an average multiplied by total time.
Reading your session against the weekly targets
A single session is more meaningful as a share of a week than as a calorie figure. The World Health Organization’s guidance, restated in its physical activity fact sheet updated on 26 June 2024, recommends that adults aged 18 to 64 accumulate at least 150 to 300 minutes of moderate-intensity aerobic activity per week, or 75 to 150 minutes of vigorous activity, or an equivalent combination — plus muscle-strengthening work on two or more days. The CDC’s guidance for Americans matches those figures.
Intensity here is defined by MET value: roughly 3.0 to 5.9 METs is moderate, and 6.0 and above is vigorous. The tool classifies whatever you selected on that basis and shows what proportion of the weekly target your session represents. It is a more honest measure of whether a workout “counted” than the calorie figure, because the health benefits of activity are substantially independent of how much energy it happened to cost.
Worth knowing: the guidance dropped its old requirement that activity come in bouts of at least ten minutes. Shorter bursts count. So does replacing sitting with light movement of any kind.
What this estimate cannot see
Fitness and efficiency. A trained runner moving at 10 km/h uses less energy than an untrained one at the same speed. MET tables cannot know which you are.
Terrain, gradient and surface. Walking uphill, on sand or into wind can cost dramatically more than the flat-ground value for the same speed.
Heat, cold and altitude. All three change the cost of the same work, sometimes substantially.
Afterburn. Excess post-exercise oxygen consumption is real but far smaller than fitness marketing suggests — typically a few per cent of the session’s cost for moderate work, more after genuinely hard intervals, and not a figure worth budgeting against.
Body composition. Two people of the same weight with very different lean mass do not burn identically, and weight alone cannot express that.
Treat any single estimate as carrying error of at least ten to twenty per cent even after the personal correction. It is a good instrument for comparing activities and tracking your own trend, and a poor one for balancing a ledger to the calorie.
What this page is not
It is an estimate from published MET values scaled to your own resting metabolism, and nothing on it is medical advice. If you have a heart condition, uncontrolled blood pressure, a recent injury or surgery, or you have been inactive for a long period and intend to start something strenuous, speak to a doctor about what is appropriate before using a number from a web page to plan it.
Nothing you enter leaves your browser
The activity, duration, weight, height, age and sex are used locally to compute the figures and are never transmitted, stored or logged. There is no account and no upload, and the page continues to work with the connection off once it has loaded.
Questions people ask about calories burned
Why is this number lower than my fitness watch says?
Two likely reasons. Most watches report gross rather than net, and most use the standard MET constant rather than your own resting rate. If you are heavier, older or female relative to the reference man behind that constant, the difference can be large — over fifty per cent in the examples above.
Should I eat back the calories I burned?
If you do, use the net figure and expect to eat back less than it suggests. Most people who stall on a calorie target are eating back gross exercise calories from an estimate that was already generous.
Which number do I use, net or gross?
Net for anything involving a calorie target. Gross when you want the total metabolic cost of a session, such as comparing two workouts.
How accurate is a MET-based estimate?
Good enough to compare activities and track trends; not good enough to balance a diet to the calorie. Ten to twenty per cent error is normal even with the personal correction applied, and more for interval or stop-start activity.
Why does the tool ask my height and age to work out calories burned?
So it can estimate your own resting metabolic rate rather than assume the conventional 3.5 ml/kg/min, which came from one 70 kg man of 40. That single substitution is the difference between an estimate about you and an estimate about him.
Does building muscle mean I burn much more at rest?
Less than commonly claimed. A kilogram of muscle burns roughly 13 calories a day at rest against about 4 for fat — real, worth having, and not the transformation it is often sold as. The larger benefit of training is what it does during and after the session, and to your health generally.
How much activity should I be doing each week?
150 to 300 minutes of moderate activity, or 75 to 150 minutes of vigorous, plus strengthening work on two days. The tool shows what proportion of that your session represents.
Do short sessions count?
Yes. The requirement that activity come in blocks of at least ten minutes was removed from the guidance. Accumulated short bursts count toward the weekly total.
Do you store my workout details?
No. Nothing is saved or transmitted.
Sources behind these numbers
- Byrne NM, Hills AP, Hunter GR, Weinsier RL, Schutz Y, “Metabolic equivalent: one size does not fit all”, Journal of Applied Physiology, September 2005 (PMID 15831804) — 642 women and 127 men, 18 to 74 years, 35 to 186 kg.
- Compendium of Physical Activities — the MET value for every activity listed here.
- WHO — physical activity fact sheet, updated 26 June 2024, for the weekly targets.
- CDC — physical activity basics.
- Mifflin MD, St Jeor ST et al, American Journal of Clinical Nutrition, February 1990 (PMID 2305711) — the resting-rate equation used for the personal correction.
- American College of Sports Medicine, for exercise prescription generally.
Tools people open next to this one
- Calorie Calculator — the intake side, where the net figure belongs.
- Water Intake Calculator — the fluid cost of the session you just logged.
- BMI Calculator — the screening number, with its limits stated.