How many calories are in a pound of fat?

The number is 3,500 kcal per pound, or about 7,700 per kilogram. It is one of the few pieces of nutrition arithmetic that is genuinely correct, and it has been used to justify more disappointing diets than almost anything else — because being right about the number tells you very little about what happens over twelve weeks.

Where 3,500 comes from

Body fat is not pure fat. Adipose tissue is roughly 87% lipid by weight; the rest is water, protein and cell structure. So:

• One pound is 453.6 g of tissue
• At 87% lipid, that is about 395 g of actual fat
• Fat yields 9 kcal per gram under the Atwater system
• 395 × 9 ≈ 3,550 kcal

Rounded, 3,500. Do the same in metric and a kilogram of body fat comes out around 7,700 kcal. The figure is not a guess or a diet-industry invention; it falls out of the composition of the tissue.

Which is why the next part is frustrating.

The rule that gets built on it, and why it breaks

The familiar version: eat 500 kcal a day below maintenance, that is 3,500 a week, so you lose a pound a week. Multiply out and 20 weeks gets you 20 pounds.

It does not, and there are three reasons.

Your maintenance falls as you get lighter. A smaller body costs less to run and less to move. Lose 10 kg and the same daily calories that once produced a deficit are closer to maintenance. The deficit erodes itself. This is the big one, and it is pure physics rather than metabolic misfortune — the target you calculated on day one is wrong by design by week eight.

The first weeks are mostly not fat. Cutting calories, especially carbohydrates, depletes glycogen, and glycogen holds roughly three grams of water for every gram stored. The two or three kilos that vanish in the first fortnight are largely that. It reverses just as fast when you eat normally again, which is the entire mechanism behind "I gained 2 kg over one weekend" — you did not eat 15,000 surplus calories.

Energy expenditure is not a fixed number. Non-exercise movement drops when intake drops — people fidget less, walk a little slower, take the lift. It is not dramatic, but it runs in the same direction as everything else on this list.

The practical shape of it: a linear 3,500-per-pound projection is a decent guide for the first month and increasingly optimistic after that. Real loss curves flatten. Anyone still on the original number by month four is comparing themselves against a target that stopped being true.

What the number is still good for

Not prediction. Two other things.

Sanity-checking a single day. A 700 kcal restaurant dessert is a fifth of a pound of fat, if it were surplus and if nothing else changed. That framing is genuinely useful for calibrating what a "big" day costs — and equally for seeing that one such day is not a catastrophe worth abandoning a week over.

Sizing a deficit, once. Pick a rate, convert it, and set the target: half a pound a week is about 250 kcal a day, a pound is 500, two pounds is 1,000 and is more aggressive than most people should run for long. Set it, then stop doing the arithmetic and start reading the scale.

What to do instead

Use the 3,500 figure to set a starting target, then let your own data replace it. The loop is short:

1. Set a calorie target from a TDEE estimate and a chosen rate of loss. 2. Log consistently for three to four weeks — consistency matters far more than precision here. 3. Compare your average intake against your actual weight trend, not day-to-day weight. 4. Adjust the target by 100–200 kcal in whichever direction the trend disagrees with the plan.

After one cycle of that you no longer need a formula, because you have measured your own maintenance under your own logging habits — which quietly cancels any systematic error in your log. If you consistently under-record by 10%, the adjustment absorbs it. That is why tracking weight and calories in the same place is worth more than any calculator: the calculator guesses, the trend measures.

Weigh daily if you can stand it and read the weekly average. Day-to-day weight moves several pounds on water, salt and digestion, and none of that is fat. The signal is there — it just needs more than one reading to see.

How calorie.one handles it

Both halves of the loop live on the same dashboard: the food log that produces your intake, and the weight trend that tells you whether the target was right. The trend line is smoothed, so a salty dinner does not read as a failed week. The calculator gives you the starting estimate, and after a few weeks of logging you can stop trusting it and trust your own numbers instead — which is the point. Start tracking, and let the arithmetic get corrected by evidence.

FAQ

How many calories are in a pound of body fat?

About 3,500 kcal. A pound is 453.6 g of adipose tissue, roughly 87% of it lipid — around 395 g of actual fat at 9 kcal per gram, which comes to about 3,550 and is conventionally rounded down.

How many calories are in a kilo of fat?

About 7,700 kcal, by the same arithmetic in metric. It is the figure most deficit calculators outside the US are built on.

Does a 500 calorie deficit really lose a pound a week?

For the first few weeks, roughly. Not indefinitely. Your maintenance falls as you get lighter, so the same intake becomes a smaller deficit every month, and the first fortnight of loss is largely glycogen and the water bound to it rather than fat. Real loss curves flatten; the linear projection does not.

How long does it take to lose 10 pounds?

On paper 35,000 kcal, or ten weeks at a 500 kcal daily deficit. In practice expect longer, because the deficit erodes as you shrink. Set the target from the arithmetic once, then correct it against your own weight trend after three or four weeks.

Can you lose a pound of fat in a day?

No. That would need a 3,500 kcal deficit inside 24 hours, which is more than most people spend in a day in total. Scale movements that large are water, gut contents and glycogen.