Your details
Your cut
Your fat stores can only release so much energy a day, and the number scales with how much fat you carry. Ask for more than that and the shortfall comes out of muscle. This is the only cutting calculator we know of that checks.
Two people at the same weight with different body fat need different protein. Yours is worked out from your lean mass and rises as you get leaner, which is exactly what the research recommends and almost nothing implements.
The chart draws the one percent a week limit as a line you can stay above. If your projection dips under it, you can see it happening instead of finding out three months later when your lifts have gone backwards.
Straight-line maths always overpromises, because your metabolism slows as you shrink. You get a realistic range alongside the optimistic number, and a reminder to recalculate every four weeks.
We will not hand you a target below the greater of the usual calorie guideline and thirty calories per kilogram of lean mass. If your inputs push it there, we raise the number and tell you why.
No account, no email capture, no fat burners, no coaching upsell. Everything runs in your browser and nothing you type leaves your device.
Age, height, weight and activity level. Metric or imperial, whichever you think in.
It matters more here than on most calculators, because it sets the ceiling on your deficit. Use a scan figure, three tape measurements, or let us estimate it.
One slider for weekly weight loss, plus whether you are lifting. If the pace is more than your body can support, we cap it and say so.
Daily calories, protein, carbs and fat, a projection against the too-fast line, and how many weeks to your goal.
You want to lose fat without undoing the training you have already done. The default pace here is the one the research supports rather than the fastest one you could survive.
The last few kilos are harder for a reason, and it is not discipline. You will see your deficit shrink automatically as your fat mass does, which is the correct response.
Weight has not moved in three weeks and you are not sure whether to eat less or eat more. Recalculate with your current numbers and compare against what you are actually doing.
You know roughly what you gained and want it off without losing the muscle underneath. The projection separates the two so you can judge the trade honestly.
Almost every cutting calculator takes a percentage off your maintenance calories and stops there. That misses the constraint that actually decides whether a cut works: how much energy your body can pull out of fat in a day.
That amount is roughly proportional to how much fat you are carrying, around 48 kcal per kilogram of fat mass per day as a conservative figure. Run a deficit larger than that and the shortfall does not vanish. It comes out of muscle instead.
Which produces a result people find counter-intuitive. A 90 kg man at 25% body fat carries about 22 kg of fat, so he can support a deficit of around 1,000 kcal a day. The same man at 12% carries under 11 kg, and his ceiling is roughly 520 kcal, half as much. Getting leaner does not make you better at cutting. It makes you worse at it.
This is why the last four kilos are so much harder than the first four, and it has nothing to do with willpower. It is why the calculator above will quietly shrink your deficit if you pick a fast pace while lean.
The recommendation across the literature is 0.5 to 1% of bodyweight per week, and the single most useful study on it is small but very clean. Elite athletes lifting four times a week were split into two groups: one losing 0.7% of bodyweight a week, the other 1.4%.
Both groups lost fat. The slower group lost more of it, and gained muscle on the way. Losing at double the rate did not get them there twice as fast; it just cost them the muscle. Another comparison found that dieting at 1 kg a week rather than 0.5 over four weeks came with about 5% off bench press strength and 30% more reduction in testosterone.
There is a usefully specific number here. A meta-analysis of resistance training in an energy deficit found that deficit size predicted lean mass change, and that a deficit of about 500 kcal a day was the break-even point. Below it, lifters on average gained a little lean mass while losing fat. Above it, they lost lean mass, and the losses grew with the deficit.
Two honest caveats on that figure. The analysis could not account for how much protein the subjects were eating, because too few studies reported it, and most participants were older and untrained. The real threshold is probably higher for someone eating 2.5 g/kg of protein and lifting hard, but 500 is the best anchor available, and it is far lower than the deficits most people run.
Without resistance training, roughly 20 to 30% of everything you lose is lean tissue. With progressive lifting and adequate protein, that can approach zero. Nothing else on this page comes close to that effect size. If you are cutting without lifting, you are not losing fat, you are losing weight, and some of it is the part you wanted to keep.
The recommendation is 2.3 to 3.1 g per kilogram of lean body mass, note lean mass, not total weight, which is why two people at the same weight need different amounts. The guidance is explicit that the leaner you are and the bigger your deficit, the further up that range you should sit. Protein is the last thing to compromise when calories get tight.
Two groups ate the same restricted diet for two weeks. One slept 5.5 hours a night, the other 8.5. Total weight loss was identical. But the short-sleep group lost 55% less fat and 60% more lean tissue, the same deficit, the same weight, redistributed against them. Another study found an even sharper split.
If you are on five hours a night, fixing that will do more for your body composition than any change to your macros.
Every calculator that multiplies your deficit out over sixteen weeks, including this one, produces a number that is slightly too good. Two reasons.
First, the familiar 7,700 kcal per kilogram figure is a simplification. The real energy density of the weight you lose depends on how much fat you started with, it holds up reasonably for people carrying a lot of fat and overstates the required deficit for leaner people.
Second, and more importantly, your energy expenditure falls as you lose weight, by more than the weight loss alone accounts for. That reduction persists, and it is accompanied by falls in leptin, thyroid hormones and testosterone, and rises in ghrelin and cortisol. Which is why the guidance from the researchers who study it is to use the smallest deficit that still produces visible progress.
We show a realistic band alongside the straight-line number rather than inventing a correction factor, because no reliable single coefficient for this exists. Recalculate every four weeks from your own trend and you will be more accurate than any formula.
The familiar 1,200-calorie (women) and 1,500-calorie (men) floors are rules of thumb, not physiology, they trace back to a clinical weight-loss prescription range that explicitly said to adjust for bodyweight and activity. A better-grounded floor is 30 kcal per kilogram of lean mass, below which hormonal and menstrual disruption become likely.
For a lot of people that second floor is considerably higher than the first. An 85 kg man carrying 70 kg of lean mass has a floor above 2,100 kcal, so the widely repeated advice to eat 1,500 would be a serious mistake for him. The calculator uses whichever of the two floors is higher and tells you when it has raised your target to meet it.
There is no reviewer name at the top of this page because there is no reviewer. What there is instead: every constant, its source, and a note wherever we are extrapolating rather than citing.
| Step | What we use |
|---|---|
| Metabolic rate | Mifflin-St Jeor (1990), switching to Katch-McArdle only when you say your body fat came from a real scan. A guessed body fat inside a lean-mass equation makes the answer worse, not better. |
| Rate of loss | 0.5 to 1% of bodyweight per week, per Helms et al. (2014), anchored on Garthe et al. (2011), the 0.7%-versus-1.4% trial. |
| Fat-store ceiling | 48.5 kcal per kg of fat mass per day, capped at 1,000 kcal. Derived from Alpert (2005). Worth knowing: the published figure is 290 kJ/kg/day, which is 69 kcal per kilogram or 31 kcal per pound, the "31 kcal per kg" you may have seen elsewhere is a unit error off by a factor of 2.2. We use the lower corrected value the author reportedly proposed before he died, and treat it as a guardrail rather than a law, since the paper is theoretical. |
| Muscle-loss risk | The roughly 500 kcal break-even point from Murphy & Koehler (2022), combined with rate of loss, resistance training, and sleep. Sleep effect sizes from Nedeltcheva et al. (2010). The weighting between those factors is our judgement; each one is sourced. |
| Protein | 2.3 to 3.1 g per kg of lean mass, rising as body fat falls, per Helms et al. and the ISSN protein position stand. The exact interpolation curve between those two bounds is ours, not a published formula. |
| Fat and carbs | Fat floored at the greater of 20% of calories and 0.5 g/kg, defaulting to 25%. Carbohydrate takes the remainder, with a warning below about 2 g/kg where training quality starts to suffer. |
| Calorie floor | The greater of the usual 1,200 / 1,500 kcal rule of thumb and 30 kcal per kg of lean mass, the energy-availability threshold below which hormonal and menstrual disruption become likely. |
| Projection | 7,700 kcal per kg, shown alongside a realistic band because that figure is not a constant and because expenditure falls more than weight loss predicts. We do not invent a correction coefficient, because no reliable one is published. |
One limitation to be blunt about. Restricting calories is the intervention here with the most potential to go wrong, and a web page cannot tell whether it is a sensible idea for you specifically. Under 18, pregnant or breastfeeding, on a GLP-1 or any medication that affects weight, managing a medical condition, or with a complicated relationship to eating, that is a conversation with a clinician, not a calculator.