How big a calorie deficit do I need to lose weight?

Take 10–20% off your maintenance estimate — commonly 300–600 kcal a day for an average adult. That produces a loss of roughly 0.5–1% of body weight per week. A larger deficit does not increase the speed of fat loss proportionally; it increases the share of muscle lost and the chance of quitting.

Why has my weight stalled even though I am in a deficit?

Two common reasons. Water shifts of a kilo or two from salt, carbohydrate, stress or cycle phase can mask several weeks of fat loss, so you need weekly averages over two to three weeks rather than single weigh-ins. Or the deficit is not actually there, because intake is underestimated — cooking oil, dressings, drinks and tastes while cooking are where the missing calories usually hide.

Why is the weight coming off slower than the maths predicted?

A lighter body spends less at rest and less per movement, so the deficit shrinks as you lose weight. In parallel, spontaneous daily activity drops without being noticed, and that alone can be worth several hundred kilocalories a day. Recalculate maintenance every 5–10% of body weight lost rather than setting the number once.

Why did I lose three kilos in the first week and almost nothing since?

The first drop is mostly glycogen and the water stored with it — each gram of glycogen holds about three grams of water — plus lower sodium and less food in the gut. It is real weight but not stored energy leaving the body, and it does not repeat. Fat loss settles at a steadier and much slower pace from week two or three onward.

Can I create the deficit with exercise alone?

In theory yes, in practice it is hard: an hour of brisk walking burns roughly 200–300 kcal for an average adult, less than the difference between two versions of dinner. Changing what you eat is the more reliable lever; training earns its place by protecting muscle and improving health, not by outrunning intake.

Calorie deficit: how to calculate it and how to hold it

A bowl of oats with berries and chia seeds, a plate of grilled chicken breast with avocado and roast sweet potato, a small dish of peanut butter and a bottle of water on a pale table

A calorie deficit is the one thing without which weight does not come down. The arithmetic takes two lines. The difficulty is confirming that the deficit is actually there — and the scale needs about three weeks to tell you.

What a calorie deficit is

A deficit is the gap between the energy your body spends over a day and the energy that arrives with food. When spending is larger, the body covers the difference from its own stores, and weight goes down. No single food, no meal-timing scheme, no order of courses suspends that rule: every diet that works, works through a deficit, whether or not it says so on the cover.

That makes the calculation short and the verification long. Two lines of arithmetic give you a number to start from. The next few weeks tell you whether the number was right. Most of the frustration in weight loss comes from treating the first part as the whole job.

Estimating your maintenance

Start with resting metabolic rate — what the body burns doing nothing. The Mifflin–St Jeor equation is the one most dietetic guidance uses, because it tracks measured resting expenditure more closely than the older Harris–Benedict formula in people who are not underweight:

Men: 10 × weight (kg) + 6.25 × height (cm) − 5 × age + 5
Women: 10 × weight (kg) + 6.25 × height (cm) − 5 × age − 161

Then multiply by an activity factor for everything you do on top of resting: 1.2 for desk work and little movement, 1.375 for light exercise one to three days a week, 1.55 for moderate training three to five days, 1.725 for hard training most days.

A worked example. A woman of 35, 70 kg (154 lb) and 165 cm (5 ft 5 in), lifting twice a week: 700 + 1,031 − 175 − 161 = 1,395 kcal at rest, multiplied by 1.375 gives roughly 1,920 kcal a day at maintenance. That is the number a deficit is subtracted from.

Two things about that 1,920. It is an average across days, not a target for each individual day. And it is an estimate with real error bars: prediction equations land within 10% of measured expenditure for most people, which on 1,920 kcal is a spread of nearly 400 kcal — larger than the deficit you are about to take. Two people with identical height, weight, age and sex genuinely burn different amounts, and the activity multiplier is the crudest part of the whole calculation. So the formula is not the answer. It is the hypothesis.

How big the deficit should be

Take 10–20% off maintenance. For the example above that is roughly 190–380 kcal, landing somewhere between 1,540 and 1,730 kcal a day. Expressed as a result, that pace is about 0.5–1% of body weight per week — for a 70 kg person, 350–700 g (0.8–1.5 lb).

The temptation is to take not 15% but 50%, on the reasoning that half the food should mean twice the speed. The price of that is well documented. Under a severe deficit, a larger share of the weight lost is lean tissue rather than fat; hunger, sleep quality and mood all get worse; and the probability of still being on the plan in three months falls sharply. Fast loss usually does not mean "more efficient". It means "a smaller fraction of what left was fat".

A percentage rather than a fixed number matters more the further you are from average size. Four hundred kilocalories off a 3,200 kcal maintenance is a mild trim; the same 400 off a 1,600 kcal maintenance is a quarter of the day's food, and it behaves like a much more aggressive diet because it is one.

Four signs the deficit is the right size, checked after two or three weeks: the weekly average weight is falling by roughly 0.5–1%; strength in the gym is holding; hunger is present but not the loudest thing in your head; and you could picture eating this way for six months without dreading it. If any one of those fails, the deficit is too big — and this is the case where adding calories back does more for the outcome than taking more away.

Why the arithmetic promises more than it delivers

A kilogram of fat tissue is usually costed at about 7,700 kcal (roughly 3,500 kcal per pound), which invites a tidy projection: 500 kcal a day is half a kilo a week, so twenty-six kilos a year. It never comes out that way, and the reason is not a broken metabolism.

As mass comes off, expenditure falls with it. A lighter body costs less to maintain and less to move, so the same walk burns fewer calories than it did in month one. On top of that, spontaneous movement drops without being noticed: someone in a deficit fidgets less, takes fewer incidental steps, sits longer. The difference in this non-exercise activity between two people of the same weight can run to several hundred kilocalories a day, and none of it registers as a decision.

There is also a modest reduction in resting expenditure beyond what the loss of tissue alone predicts, which shows up consistently in studies of people who have lost substantial weight. It is not large enough to stop weight loss, but it is large enough to flatten a projection made on day one.

The practical conclusion is simple: recalculate. A deficit is not a setting you choose once. Every 5–10% of body weight lost is a reasonable trigger to redo the maintenance estimate and adjust.

The first two weeks are mostly water

The scale usually drops fast at the start — two, three kilos in the first week is common — and almost none of it is fat.

The body stores carbohydrate as glycogen in muscle and liver, several hundred grams of it, and each gram of glycogen is held with roughly three grams of water. Cut carbohydrate intake, run a deficit, and glycogen stores draw down, taking that water with them. Lower sodium intake and less food in the gut push the number down further. It is real weight, but it is not stored energy leaving the body, and it does not repeat: week three will not look like week one.

This cuts both ways, and the second direction is the one that makes people quit. Refill glycogen — a large meal, a weekend, a return to normal carbohydrate intake — and the same few kilos come back within days, while the fat balance over that period may have been perfectly fine. Reading either swing as a verdict on the diet is the most common way to abandon a plan that was working.

When the scale is flat but the deficit is real

The usual point of surrender is week two or three with no movement. Here is what is normally behind that.

Water. Body mass swings by a kilo or two (2–4 lb) on salt, carbohydrate, stress, cycle phase and gut contents. Fat can be leaving the whole time while the number on the scale sits still, until the retained water goes all at once and a week's worth of progress appears overnight.
The window is too short. A 400 kcal deficit is about 400 g of fat a week. Against noise of a kilo, that signal is only visible in an average over two or three weeks.
New resistance training. In the first weeks, muscle holds more water and glycogen, which offsets fat loss on the scale almost exactly. This is a good problem and it resolves on its own.

The fix is measurement, not patience alone. Weigh yourself under the same conditions — morning, after the bathroom, before eating — and compare weekly averages rather than individual days. A single reading is a data point in a noisy series; seven of them averaged is a measurement. A tape around the waist once a fortnight catches what the scale misses, particularly early in a training programme.

If the three-week average genuinely has not moved, the deficit is probably not there. In the large majority of cases the reason is the same: more was eaten than was recorded. Cooking oil, dressings, tastes while cooking, drinks. Studies using doubly labelled water — the reference method for measuring real energy expenditure — consistently find self-reported intake falling short of measured intake by 20% or more, and the gap grows with the calorie density of the meal. It is not a question of honesty. Portions and cooking fats are simply hard to see.

What protects muscle in a deficit

Weight lost in a deficit is never purely fat. What you can influence is the ratio, and two levers do nearly all the work.

Protein. Aim for 1.6–2.2 g per kilogram of body weight (roughly 0.7–1.0 g per pound) — for the 70 kg example, about 110–155 g a day. This is well above the general population reference intake, and deliberately so: the requirement rises when energy is restricted, because the body will otherwise take amino acids from muscle. Protein is also the most satiating of the three macronutrients per calorie, which makes the deficit easier to hold as a side effect.

Resistance training. Two sessions a week is enough to signal that the muscle is in use and should be kept. Progressive load matters more than volume or novelty. Cardio has its own benefits, including for visceral fat specifically, but it does not preserve lean mass the way lifting does.

Sleep is the quiet third lever. Short sleep shifts the composition of weight lost toward lean tissue and raises appetite the following day, which is a direct tax on adherence — the mechanism is covered in more detail in the piece on sleep and weight.

One thing that does not belong on this list is chasing foods that supposedly burn more energy than they contain. Negative-calorie foods are not a real category; celery is useful in a deficit because it is bulky and cheap in calories, not because it runs the arithmetic backwards.

When these numbers do not apply

Prediction equations were built on healthy non-pregnant adults, and there are situations where subtracting anything from the result is the wrong move.

Pregnancy and breastfeeding. Energy needs rise rather than fall, and deliberate restriction is not something to design on your own. Any change to intake goes through the clinician managing the pregnancy.
Children and adolescents. A growing body has requirements that adult formulas do not model, and restriction during growth carries costs an adult diet does not. This is a paediatric conversation.
Recovery from an eating disorder, or a history of one. Calorie targets, tracking and weighing are the machinery of the illness. Do not build any of it back without the treatment team.
Very low intakes. Below roughly 1,200 kcal a day, meeting protein and micronutrient requirements from ordinary food becomes unreliable, and formal very-low-calorie diets — around 800 kcal or less — exist only as medically supervised programmes with bloodwork and monitoring, because of risks including gallstones, electrolyte disturbance and cardiac effects. That is a treatment, not a stricter version of dieting.

Medication matters too. Insulin and sulfonylureas need dose review when intake drops, or the deficit produces hypoglycaemia rather than weight loss. Thyroid disease, some psychiatric medications and corticosteroids all shift the baseline the formula assumes. If any of these apply, the maintenance estimate is a conversation to have with a doctor, not a calculation to run alone.

Where to start

Not with the formula. Start with a week of observation: eat as you normally eat, change nothing, and record all of it honestly, weighing what you can. That week gives you your actual current intake — and since your weight over that week tells you whether you were in surplus, deficit or balance, it gives you a maintenance figure grounded in your own data rather than a population average.

Subtract 10–20% from that. Hold it for three weeks. Compare weekly averages, not days. Then adjust once, based on what the scale actually did, and repeat. The formula gets you a plausible starting point in five minutes; the measurement is what makes it yours.

Count it from a photo

Frequently asked questions

How big a calorie deficit do I need to lose weight?: Take 10–20% off your maintenance estimate — commonly 300–600 kcal a day for an average adult. That produces a loss of roughly 0.5–1% of body weight per week. A larger deficit does not increase the speed of fat loss proportionally; it increases the share of muscle lost and the chance of quitting.
Why has my weight stalled even though I am in a deficit?: Two common reasons. Water shifts of a kilo or two from salt, carbohydrate, stress or cycle phase can mask several weeks of fat loss, so you need weekly averages over two to three weeks rather than single weigh-ins. Or the deficit is not actually there, because intake is underestimated — cooking oil, dressings, drinks and tastes while cooking are where the missing calories usually hide.
Why is the weight coming off slower than the maths predicted?: A lighter body spends less at rest and less per movement, so the deficit shrinks as you lose weight. In parallel, spontaneous daily activity drops without being noticed, and that alone can be worth several hundred kilocalories a day. Recalculate maintenance every 5–10% of body weight lost rather than setting the number once.
Why did I lose three kilos in the first week and almost nothing since?: The first drop is mostly glycogen and the water stored with it — each gram of glycogen holds about three grams of water — plus lower sodium and less food in the gut. It is real weight but not stored energy leaving the body, and it does not repeat. Fat loss settles at a steadier and much slower pace from week two or three onward.
Can I create the deficit with exercise alone?: In theory yes, in practice it is hard: an hour of brisk walking burns roughly 200–300 kcal for an average adult, less than the difference between two versions of dinner. Changing what you eat is the more reliable lever; training earns its place by protecting muscle and improving health, not by outrunning intake.

Settings

Articles

References

Service