Daily calorie needs chart by age, sex and activity

Your daily calorie need is the energy you spend over twenty-four hours, and at a stable weight it is also the energy you take in. It is usually called total daily energy expenditure, or TDEE. The only way to measure it properly is doubly labelled water: you drink water whose hydrogen and oxygen are isotopically tagged, and the differing rates at which the two labels leave your body over the following week or two reveal how much carbon dioxide you produced, and therefore how much energy you burned. It is accurate, it is expensive, and almost nobody reading a chart has had it done.

Everything else (this page, every calculator, the tables in national dietary guidance) is prediction. A prediction equation takes the variables that correlate with energy expenditure across a measured population, fits coefficients to them, and returns the average expenditure of people who share your inputs. The Mifflin–St Jeor equation used here was fitted to indirect calorimetry measurements in healthy adults in 1990, and it is the equation that performs best in systematic comparisons, predicting resting metabolic rate to within about 10% of the measured value for most healthy adults. Best available is not the same as exact.

So a chart of daily calorie needs describes a statistical person with your height, weight, age and sex. It does not describe you. Two people who match on all four inputs can differ by several hundred calories a day because of differences in organ and fat-free mass, thyroid function, sleep, and, above all, how much they fidget, pace and stand. That spread is the reason the band on this page is printed rather than buried.

The rows are representative ages; the columns are the five standard activity factors. Every cell is basal metabolic rate multiplied by that factor. Read across your age row to the column that honestly matches your week, then apply the ±12.5% band to whatever you land on.

Notice how flat the columns are. Moving all the way down the table, from 25 to 75, changes the moderately active figure by less than the gap between two adjacent activity columns at any single age. Which of those two decisions matters more is not a close call.

The same table for a man of 178 cm and 78 kg. The structure is identical and the level is higher, and not only because the reference man is taller and heavier. Mifflin–St Jeor adds a fixed constant for males and subtracts a larger one for females, a crude stand-in for the average difference in fat-free mass between the sexes at the same height and weight.

That constant is also why the equation drifts for anyone whose body composition is unusual for their sex: a lean, heavily muscled woman will be underestimated, and a man carrying little muscle overestimated. If you know your body fat percentage from a reliable method, the Katch–McArdle equation, which works from lean mass and ignores sex entirely, is a better starting point; our BMR calculator switches to it automatically when you supply a figure.

Bodyweight is the strongest single input in the equation: Mifflin–St Jeor charges ten kilocalories of basal metabolic rate for every kilogram of bodyweight, before any activity multiplier is applied. At a moderate activity level that works out at roughly 160 kcal a day for every ten kilograms, which is what the middle column below shows.

This is the arithmetic behind something that surprises people mid-way through losing weight: as the weight comes off, maintenance falls with it, so a deficit that was real in month one is smaller in month four even though nothing about the eating has changed. It is also the mirror-image problem for anyone trying to gain: the target keeps moving up.

The table runs across a wide range of bodyweights at a single height, so the rows at either end describe very different bodies. It is a lookup, not a suggestion that any of these weights is a goal.

The factors below are the standard set used across the field. They are brackets, not measurements: each covers a range of real behaviour, and the step between adjacent levels is larger than most people's honest week-to-week variation. The last two columns show what each level does to the reference woman and reference man at 35, so the cost of a wrong choice appears in calories rather than in the abstract.

Read the example week rather than the label. The labels flatter and the examples do not, which is deliberate: the word moderate reads as average, while in this table it describes a genuinely committed training week.

The multiplier is shorthand for four separate things, and they are worth unpacking because each responds to completely different levers.

The first is fixed on any given day; the last is the one that moves most, and the one nobody tracks. Non-exercise activity thermogenesis varies enormously between people of the same size, and it also falls quietly during a diet as the body economises on spontaneous movement. That adaptation is a large part of why a planned deficit under-delivers.

• Basal metabolic rate: the cost of being alive at complete rest, typically around 60–70% of the total. This is the part the equation on this page predicts.
• The thermic effect of food: the energy spent digesting, absorbing and storing what you eat, usually around 10% of intake and higher on a high-protein diet than a high-fat one.
• Exercise activity thermogenesis: deliberate training. For most people this is smaller than assumed: an hour of hard work is a few hundred calories, not a thousand.
• Non-exercise activity thermogenesis: walking, standing, fidgeting, gesturing, taking the stairs. In sedentary to moderately active people this is the largest source of variation between two otherwise identical bodies.

The basal equation is the well-behaved half of the calculation. The multiplier is where the estimate goes wrong, because it asks you to classify your own behaviour and people are poor at that. A single level of over-rating adds about 240 kcal a day for the reference woman and about 300 kcal for the reference man, enough to cancel a deliberate deficit outright and leave you convinced the arithmetic is broken.

Two things drive the over-rating. Intent gets confused with volume: three enthusiastic sessions a week is lightly active, not moderately active, however hard those sessions feel. And the other twenty-three hours disappear from the assessment: someone who trains four times a week and otherwise sits for ten hours a day has a very different total from someone who trains the same amount and is on their feet all day.

The international reference on this, the FAO/WHO/UNU 2004 expert consultation, does the same arithmetic under a different name. It expresses habitual activity as a physical activity level, or PAL (total expenditure divided by basal rate), and defines sedentary or light lifestyles as PAL 1.40 to 1.69, active or moderately active as 1.70 to 1.99, and vigorous as 2.00 to 2.40. Those brackets are wider than the neat single multipliers used by most calculators, ours included, and the width is the honest part.

The practical rule that follows: when you are between two levels, take the lower one. If you are wrong you will find out in three weeks and gain a little, which is easily recovered. Wrong in the other direction, you spend three weeks eating at a surplus you believed was a deficit and conclude that calories do not apply to you.

The chart gives you a hypothesis. Three to four weeks of your own measurements turn it into something worth trusting, and the procedure is not difficult.

Weigh yourself under the same conditions each morning and work with the weekly average, never the daily figure. Day-to-day movement is water, glycogen and gut contents, and a kilogram either way means nothing. Eat at the chart's maintenance figure as consistently as you can and log intake honestly, cooking oils, drinks and weekends included. After three weeks, compare the average of week three with the average of week one. If weight has held, the estimate fits you. If it moved by half a kilogram a week, your real maintenance is roughly 550 kcal a day away from the estimate in the corresponding direction, using the standard planning figure of 7,700 kcal per kilogram of body tissue.

Adjust in steps of 150 to 200 kcal rather than in one jump, then hold the new figure for another three weeks. Two rounds of this generally land closer than any equation can, because you are no longer estimating your expenditure; you are measuring it, using your own body as the calorimeter. Recalibrate after any substantial change in weight, training load or job, since all three move the target.

One caution: the procedure assumes reasonably accurate intake logging, and under-reporting is the rule rather than the exception in the research literature, often by a fifth or more, and largest in the people who most want the number to be flattering. If your measured maintenance comes out implausibly far below the chart, the logging is the first place to look, not your metabolism.

Compare this page with the estimated energy needs tables published in national dietary guidance (the USDA and HHS Dietary Guidelines for Americans are the most widely quoted example) and the figures will not match exactly. The reason is a difference in method, not a mistake by either side.

Official tables are built around a reference person. They fix one height and weight for each age and sex group, usually a value considered to sit in the healthy range, and publish the needs of that hypothetical individual. They also derive expenditure from equations fitted directly to doubly labelled water measurements of free-living people, rather than from a resting-rate equation multiplied by an activity factor, and they describe activity in walking-equivalent terms (how far you cover on foot in a day on top of the activity of daily living) rather than in gym sessions per week.

The consequences are predictable. Official tables cannot be personalised: they answer what a typical woman of a given age needs, not what you need at your actual weight and height. Equation-based charts like this one take your measurements but inherit the activity-multiplier problem described above. They are two different tools, and the sensible use of the official tables is as a population sanity check: if your calculated figure sits wildly outside the published range for your age and sex, something in your inputs deserves a second look.

Neither approach resolves the underlying uncertainty. Both are averages of measured people, presented to an individual who was never measured.

Several real factors move expenditure, and it is worth separating the large ones from the folklore.

The ageing entry deserves particular emphasis, because the belief that metabolism collapses in middle age is close to universal and does not survive measurement. The largest doubly labelled water dataset assembled to date, covering more than six thousand people from infancy to age ninety-five, found that body-size-adjusted total expenditure is essentially stable from about twenty to sixty, with a break point near sixty-three, after which it falls by roughly 0.7% a year. Weight gained in the forties far more often reflects changed behaviour, lost muscle and larger portions than a changed metabolic rate.

• Age: five kilocalories of basal rate per year in this equation, about 390 kcal a day at a moderate activity level across the fifty years from 25 to 75. Measured expenditure is flatter still until around sixty.
• Muscle mass: real but modest. Resting skeletal muscle costs on the order of 10–15 kcal per kilogram per day against a few kcal for fat tissue, so several kilograms of hard-won muscle is worth tens of calories a day at rest, not hundreds. The training that builds it costs more than the tissue does.
• Illness and fever: substantial and usually short-lived. Expenditure rises with body temperature, and serious burns, sepsis or major trauma can push it far above any chart, which is why hospital nutrition is prescribed by a dietitian rather than looked up.
• Thyroid disease: one of the few conditions that genuinely resets the baseline. Untreated hypothyroidism lowers resting expenditure and hyperthyroidism raises it, both by enough to make a prediction equation useless until treatment has stabilised.
• Pregnancy and lactation: needs rise progressively through pregnancy and further during breastfeeding, and neither is a case for a general-purpose chart. Both are managed with a midwife or dietitian, and no calculator on this site will produce a deficit for either.
• Medication: beta-blockers lower resting heart rate and expenditure slightly; corticosteroids and some antipsychotics change appetite and body composition; stimulants raise expenditure modestly. The effects are real but generally smaller than the intake changes they cause.
• Large weight loss: expenditure after substantial loss often sits below what the equation predicts for the new, lighter body: partly less tissue, partly reduced spontaneous movement. Calibrating against your own data is the only way to find the true figure.

A chart of calorie needs is often the first step towards a deficit, and that is where the harm in this category happens. Our calculation layer enforces its rails in code rather than merely in copy, and the table below reads them straight out of the module the calculators use, so this page cannot describe a limit the tools do not apply. In short: never below 80% of your own TDEE, never below 1,200 kcal for a woman or 1,500 kcal for a man, and never faster than 1% of bodyweight a week.

The reasons are practical rather than moralistic. Very low intakes make adequate protein, calcium, iron and essential fats hard to reach. Aggressive deficits cost a larger share of lean mass, which lowers maintenance further and makes the next phase harder. And the very low calorie diets used clinically are used under supervision, with monitoring, because they carry real risks including gallstones and electrolyte disturbance. A calculator that hands an unsupervised 900 kcal target to a stranger is not being helpful.

You will notice in the table that for most people the percentage rail binds before the absolute floor does. That is intended. A fixed floor treats a 50 kg person and a 110 kg person identically, which is plainly wrong; the percentage rail scales the limit to the body it applies to, and the absolute floor is the backstop underneath it.