How to Lose Fat Without Losing Muscle

By Jesse · Last updated August 20, 2026

The title is slightly dishonest, and it is worth saying so at the top. Losing weight almost always costs some lean tissue, and a diet that costs none is the exception rather than the target. The variable you actually control is the ratio — how much of each kilogram lost comes off as fat rather than muscle — and that ratio moves a long way depending on four decisions.

This is the guide that owns the composition of the weight you lose. How to set the calorie target in the first place is in how many calories to eat to lose weight, and how long the process takes in how long does it take to lose weight.

What normally happens, and what the exceptions have in common

A systematic review gathered the studies of energy-restricted, resistance-trained, relatively lean athletes and looked at what happened to fat-free mass. Across 13 groups, body fat percentage fell in every single one — dieting reliably removes fat. Fat-free mass fell in nine of the 13, by between 0.3 and 2.7 kg. Six groups gained fat-free mass, did not lose it, or lost only an amount too small to be statistically significant. Those two counts overlap, because some of the nine losses were small enough not to reach significance.

The interesting part is what those six had in common. Five of the six were either higher in body fat to begin with, in a milder deficit, or exposed to a new training stimulus. The sixth — lean, heavily restricted, and training as usual — held onto its fat-free mass by eating the highest protein intake in the entire review, 2.5 to 2.6 g/kg.

That single paragraph contains the whole strategy. The four levers below are the four ways those groups differed, in descending order of how much they are usually worth.

Lever 1: keep training, and keep the load

Muscle is expensive tissue to maintain, and a body in energy deficit will not maintain it without a reason. Resistance training is that reason. It is the only lever on this list that is close to non-negotiable: protein without training preserves much less than training without ideal protein, and every study in the review above had its participants lifting.

The practical mistake is not stopping training but changing it. Cutting the weight on the bar and adding reps to make sessions feel harder removes precisely the signal that protects lean mass. In a deficit, recovery is the constraint, so the right adjustment is to reduce volume and protect intensity: fewer hard sets at a load close to what you were using before, rather than more sets at a lighter one. Two to three sessions a week covering the major movements is enough to hold what you have.

Lever 2: scale protein to lean mass, not body weight

Protein requirements rise in a deficit, and the review above put the figure for energy-restricted trained athletes at 2.3 to 3.1 g per kilogram of fat-free mass — to be scaled up with leanness and with the severity of the restriction. The fat-free denominator is what makes that recommendation clever rather than merely high: it adjusts itself.

Every row is the same 85 kg on the scale. Only the composition differs, and yet the target moves by 39 g a day at the bottom of the range — because a leaner body is almost entirely the tissue the protein is there to defend. Expressed per kilogram of body weight, the person at 30% body fat is asked for 2.2 g at the top of the range and the person at 10% for 2.8 g. Nobody applied a leanness adjustment; it fell out of the fat-free denominator. That is also why body-weight-based targets mislead at the extremes — they ask too much of someone with a lot of fat to lose and too little of someone already lean.

To use the table you need a fat-free mass estimate, which the body fat calculator will produce from tape measurements, and the protein calculator will take whichever weight you want to base the target on. The full evidence behind protein figures generally, including where the lower everyday numbers come from, is in how much protein you really need per day.

One trial makes the size of this lever unusually clear. Two groups of young men were put in a 40% energy deficit and trained hard six days a week; the only difference was protein at 2.4 versus 1.2 g per kilogram of body weight. The higher-protein group gained 1.2 kg of lean mass and lost 4.8 kg of fat. The lower-protein group gained 0.1 kg of lean mass and lost 3.5 kg of fat. Same deficit, same training, better result on both counts.

Lever 3: lose weight slowly, in percentage terms

The best direct test of dieting speed randomised 24 athletes to lose weight at a target of 0.7% of body weight per week or 1.4% per week, all of them doing four resistance sessions a week alongside their normal training. Both groups lost a similar total amount of weight; the fast group simply got there sooner.

The slower group gained 2.1% lean body mass over the diet. The faster group ended flat. And the slower group lost more fat in relative terms despite a smaller cut in intake, because a larger share of its weight loss came from fat in the first place. There is no efficiency argument for the faster approach: it delivers the same weight in less time and a worse body composition for it.

One honest detail about that trial: the fast group never actually hit its target. Aiming for 1.4% a week, it averaged 1.0%, while the slow group landed on its 0.7% almost exactly. That is worth knowing in both directions. The composition penalty above was produced by 1.0% a week, not by 1.4% — so the cost of pushing hard shows up earlier than the headline numbers suggest. And an aggressive target is harder to hold than it looks on paper.

Percentages rather than kilograms, because the same absolute rate means very different things at different sizes. Here is what each rate implies in kilograms and in the daily deficit needed to produce it:

Notice how much room there is between the rows. At 90 kg the difference between the well-supported 0.7% and the aggressive 1.4% is 693 kcal a day — a genuinely different diet, not a minor adjustment. The calorie deficit calculator will produce these targets from your own maintenance estimate.

Lever 4: how lean you already are

This one is not a decision, but it changes how the other three should be set. Someone carrying a lot of fat has abundant stored energy to draw on, and their body has little reason to break down muscle to find more. As you get leaner, that buffer shrinks, and the same deficit starts to take a larger share of its energy from lean tissue. It is the reason the review found that the groups holding fat-free mass tended to be the ones with more of it to lose.

The adjustments follow from it. Higher body fat: a more aggressive deficit is tolerable, protein at the lower end of the fat-free-mass range is adequate, and lean mass may even increase if training is new. Already lean: keep the deficit small, push protein towards the top of the range, protect training intensity carefully, and accept a slower timeline. The leaner you are, the more the diet becomes about damage control and the less it rewards impatience.

A worked plan, start to finish

Take someone at 85 kg and 22% body fat, with maintenance estimated at 2,700 kcal.

1. Fat-free mass: 85 kg times (1 − 22%) = 66.3 kg.
2. Rate: 0.7% of 85 kg = 0.59 kg per week, which needs a deficit of about 655 kcal a day.
3. Calorie target: 2,700 − 655 = 2,045 kcal a day.
4. Protein: 66.3 kg of fat-free mass at 2.3 to 3.1 g/kg = 152 to 206 g a day. At the lower figure that is 608 kcal, or 30% of the calorie target — worth knowing, because it constrains how the rest of the day is built.
5. Training: three resistance sessions a week covering the major movements, keeping load close to pre-diet levels and trimming sets rather than weight as fatigue accumulates.
6. Review at four weeks: seven-day average weight down by roughly 2.4 kg, waist down, main lifts holding. If weight is falling faster than planned, eat more. If the lifts are sliding, eat more.

152 g of protein inside 2,045 kcal is a real constraint rather than an afterthought, and it is easier to hit with foods chosen for protein per calorie rather than protein per 100 g. Our macros by food chart ranks common foods on exactly that basis.

How to tell whether it is working

No method available at home can separate a kilogram of fat from a kilogram of muscle over a single week. Research uses DEXA or four-compartment models; skinfold calipers and tape-based estimates carry error bars wider than a month of genuine change. So the honest approach is converging evidence, reviewed monthly:

• Strength on your main lifts is the most sensitive early signal available to you. Holding or improving on a moderate deficit is a good sign. A steady decline over three or four weeks, on a deficit that is not extreme, is the clearest warning that the diet is too aggressive or protein is too low.
• Waist circumference against scale weight. Measure at the navel, same time of day, once a week. If weight is falling and the waist is falling proportionally faster, the loss is skewing towards fat. If weight is falling and the waist is not moving, be suspicious.
• Seven-day average weight against the planned rate. Single weigh-ins are noise; a fortnight of averages is signal. Loss much faster than planned is not good news in this context, it is the main risk factor for losing lean mass.
• A tape-based body fat estimate from the body fat calculator, used only as a trend across months and always by the same person with the same tape. Its absolute value is not reliable enough to act on; the direction over three months is.

What is oversold

• Meal timing and fasting windows. Getting protein spread across three or more meals is a mild positive. Which hours you eat in is not a lever on lean-mass retention once total calories and protein are set.
• Supplements. Creatine has good evidence for supporting training performance, and protein powder is a convenient way to hit a target. Nothing sold as an anti-catabolic or a fat burner competes with getting the four levers above right.
• Very low carbohydrate for fat loss specifically. Useful if it helps you adhere, and neutral otherwise. Once protein and calories are fixed, the carbohydrate-to-fat split does not change the composition of the weight lost — though very low carbohydrate intakes can reduce training quality, which matters here indirectly.
• Extreme deficits with the promise of catching up later. The lean mass lost during an aggressive diet does not come back for free; regaining it takes months of training and a surplus that brings fat with it. The slow route is shorter than the fast route plus the repair.

Common questions

These are general estimates for healthy adults, not medical advice. If you are pregnant, under 18, have a medical condition, or a history of disordered eating, talk to a doctor or registered dietitian before starting a calorie deficit. See our methodology for the sources behind these figures.