Chapter 1 · the rate of gain
How fast can muscle be gained?
Rate of lean-mass accretion, and how much of it an energy surplus buys.
The claim as you’ve heard it
"Eat big to get big. If you're not gaining, eat more. Bulk hard, cut later."
The verdict
A modest surplus buys the muscle a trained lifter can build in a month; past it, the extra is mostly fat. In every trial that set a larger surplus against a smaller one, the larger added more fat, and it added more muscle in one trial of three. How big "modest" is has not been measured: the reviews guess a tenth to a fifth above maintenance, gaining a quarter to a half of one per cent of bodyweight a week.
The picture
What the biology says
Muscle is built at a rate with a ceiling. After a session, the rate at which muscle protein is made rises with the protein in the next meal and then stops rising: 20 g of whey lifted it as far as 40 g did, and the extra from the larger dose went to oxidation and urea. A whole-body session moves the ceiling up, not away: 40 g did about a fifth more than 20 g after one.
The raised rate does not last the week. After one session it was up 112 % at three hours, 65 % a day later and 34 % at two days; in a second study, in trained arms, it was back near baseline by 36 hours. The ceiling is a rate per day, set by the training signal and the muscle it reached.
Without training, a surplus becomes mostly fat. Adults overfed by about 950 kcal a day for eight weeks stored between half and more than nine-tenths of the extra energy as fat, about 3.5 kg of it in every group; with normal or high protein they also added about 3 kg of lean mass, and with 5 % protein none. Twelve pairs of twins overfed by 1,000 kcal a day for a hundred days gained 8.1 kg, about two-thirds of it fat.
A large surplus can add body protein even without a programme, and it comes with the fat: a 40 % surplus for six weeks added 0.44 kg of body protein (3.7 %), a 10 % surplus from protein alone added none, and the protein gained rose in step with the fat gained (r = 0.82).
In trained athletes, the bigger surplus bought fat. Elite athletes counselled onto a meal plan about 600 kcal a day above their peers eating freely gained bodyweight faster (3.9 % against 1.5 %), gained fat five times faster (15 % against 3 %), and gained no more lean mass. Strength rose 6–12 % in both groups; the counselled group got slower over 40 m.
Setting a 5 % surplus, a 15 % surplus and maintenance side by side, the surpluses thickened the skinfolds and not the muscle: the sum of eight skinfolds rose 10 to 12 mm on a surplus and fell 1 mm at maintenance, quadriceps and triceps thickness changed alike in every group, and how fast bodyweight rose predicted fat gain (R² 0.49) far better than biceps growth (R² 0.24). The large surplus did lift the bench press more: 13 kg against 7 and 6.
The one trial in which the bigger surplus bought more muscle bought it dearly and measured it loosely. Bodybuilders eating 67 kcal per kg a day against 50 gained more muscle mass (2.7 % against 1.1 %) and nine times more fat (7.4 % against 0.8 %) in four weeks; but the muscle figure came from an equation of height, weight, age and sex, which grows with bodyweight by construction.
The rate has a rough line. In trained men overfed with a supplement for six weeks, a gain of about 0.55 % of bodyweight a week corresponded on average to all of it being lean mass; faster, and the surplus went increasingly to fat. The spread around the line (78 to 122 %) means an individual on that rate could be losing fat or gaining it.
Under-eating costs the gain. Across 59 randomised trials of lifting, people in an energy deficit gained less lean mass than people who were not, with strength unaffected, and the meta-regression put the deficit that cancels lean gains at about 500 kcal a day. The trials were in untrained adults, the controlled ones mostly older women; one pair studied trained lifters.
People with fat to draw on and no training history can build muscle with no surplus at all. Overweight young men, recreationally active and none of them lifting before, ate 40 % below maintenance and trained six days a week: on 2.4 g of protein per kg they added 1.2 kg of lean mass while losing 4.8 kg of fat; on 1.2 g they held lean mass and lost 3.5 kg.
How a gain splits between lean and fat depends on the fat you start with. Re-analysing four overfeeding studies, the fatter the subject at the outset, the larger the fat share of what they gained; the leaner, the larger the lean share. Forbes drew the curve in 1987; Hall's re-analysis is what carries its weight-gain side, because Forbes's own gain data leaned on anorexic patients regaining.
Positions, each at its strongest
Bulk hard, cut later
At its strongest: the deficit trials: about 500 kcal a day under maintenance cancelled lean gains, and a guaranteed surplus removes that risk; hard-gainers under-eat without noticing; fat loss is the easier problem.
Where it breaks: at the trial level, in trained lifters. The extra was fat five times over in the elite athletes and skinfolds rather than muscle in the lifters, and the cut it buys costs weeks and, past about 500 kcal a day, some of the lean mass.
Lean gain, small surplus
At its strongest: the trials: fat tracks the rate of gain and muscle barely does; the partition worsens as body fat rises; a quarter to a half of one per cent of bodyweight a week is the one figure that recurs.
Where it breaks: precision it cannot deliver. Nobody measures a surplus to a hundred calories at home, and even the trial could not hold its groups to the surplus assigned: the 5 % and 15 % groups ended up in a similar surplus.
No surplus needed
At its strongest: recomposition is real in the untrained and the fatter: 1.2 kg of lean mass on a 40 % deficit in overweight men new to lifting, and lean up while fat fell in trained women on high protein, in the trials the recomposition review collects.
Where it breaks: carried to the lean, trained lifter, whose trials all show the surplus still mattering and whose stores cannot pay for the building. The WHO trap in one move.
The mechanism in full
Muscle protein synthesis is switched on by loading and by amino acids. After a session it runs high for a day or two: 112 % above rest at three hours, 65 % at a day, 34 % at two days in untrained adults (Phillips et al. 1997), and near baseline by 36 hours in a trained arm (MacDougall et al. 1995). At any moment it saturates with the amino acids on offer: 20 g of protein raised it as far as 40 g after a leg session (Moore et al. 2009; Witard et al. 2014), 40 g did about a fifth more than 20 g after a whole-body session (Macnaughton et al. 2016), and what the machinery cannot use is oxidised. That is the ceiling: a rate, set by the training signal and the muscle it reached, not by the calories behind it.
The energy that building costs is small against a day's intake. Slater et al. (2019) put the energy stored in a kilogram of muscle at about 5,000–5,200 kJ, roughly 1,200 kcal, three-quarters of it water and a fifth protein, and the running cost of a kilogram of muscle at about 54 kJ a day. The cost of assembling it (the synthesis itself, its transcription, folding and turnover) is real and has never been quantified, which is why their textbook range of 1,500–2,000 kJ a day above maintenance is offered as a place to start conservatively, not a measurement.
What a surplus changes is not the ceiling but what is left over once it is reached, and where the leftover goes is the partition (the Why card). Two things set it. The first is the fat you carry: across four overfeeding studies, the fatter the subject at the outset, the larger the fat share of the gain (Hall 2007, on Forbes 1987). The second is how close you are to the ceiling. The untrained have room: overweight men new to lifting built 1.2 kg of lean mass in a 40 % deficit (Longland et al. 2016). The trained do not, which is why the same "eat big" produces recomposition in a beginner and a bulk-and-cut cycle in a lifter. In sedentary adults a surplus with adequate protein still adds lean mass, about 3 kg in eight weeks (Bray et al. 2012) and 2.7 kg in a hundred days (Bouchard et al. 1990, as Slater et al. report it), but lean mass is not muscle: it includes water, glycogen, organ, and the protein of the fat tissue itself, and even body protein measured by a four-compartment model rose in step with the fat (Hatamoto et al. 2024).
Below the ceiling, the gym is right: a deficit costs the gain. Across 59 trials, people lifting in a deficit gained less lean mass than people who were not, with strength unaffected, and the deficit that cancelled lean gains was about 500 kcal a day (Murphy & Koehler 2022), in untrained and mostly older adults, the only who the deficit literature offers.
Above it, the trials in trained lifters agree on fat and split on muscle. Elite athletes counselled onto about 500 kcal a day more gained fat five times faster and lean mass no faster (Garthe et al. 2013). Trained lifters on 5 % and 15 % surpluses thickened their skinfolds, not their quadriceps or triceps, though the large surplus lifted the bench press more (Helms et al. 2023). The one trial in which the larger intake bought more muscle measured muscle by an equation of height and weight (Ribeiro et al. 2019). The one attempt to find the rate put the line near half a per cent of bodyweight a week, with a spread wide enough that an individual on it could be gaining fat or losing it (Smith et al. 2021).
So the honest sentence is the verdict's: a surplus helps, a large one buys fat, and the size that buys the muscle without the fat has not been measured. The reviews' numbers, 10–20 % above maintenance and 0.25–0.5 % of bodyweight a week (Iraki et al. 2019), 1,500–2,000 kJ a day (Slater et al. 2019), are recommendations, and they are graded here as what they are.