Neither builds noticeably more muscle in the trials so far; the bigger surplus mainly adds more fat. In an 8-week trial of 17 trained lifters, muscle thickness grew about the same whether they ate at maintenance, in a small surplus or in a big one, while fat under the skin rose with both surpluses and most with the big one1.
The trial is small, so the answer comes with caveats. This article covers what "lean" and "dirty" meant in the research, what the numbers show, and where a bigger surplus did seem to help.
What the terms mean here
A surplus means eating more energy than you use. Maintenance means eating about what you use, so body weight holds steady. A "lean bulk" is a small surplus aimed at slow weight gain. A "dirty bulk" usually means eating a lot, often with little attention to what.
The research tested only the first half of that: the size of the surplus. No trial we found in trained lifters compared food quality during a surplus. So in this article, "dirty bulk" means a big surplus, not junk food.
What was studied
The key trial, published in 2023, randomised 21 trained lifters and had 17 finish (15 men, 2 women; average age about 27; at least a year of lifting)1. Everyone did the same supervised full-body training 3 days a week for 8 weeks. After a phase of eating at steady weight, each group got an energy target.
| Group | People | Planned surplus | Logged change in intake |
|---|---|---|---|
| Maintenance | 6 | none | +169 kcal/day |
| Small surplus | 6 | about 5% | +489 kcal/day |
| Big surplus | 5 | about 15% | +719 kcal/day |
All three groups had the same minimums for protein, fat and carbohydrate1. Muscle thickness was measured by ultrasound at six sites on the thigh and arms. Strength was the one-rep max, the heaviest weight lifted once, on squat and bench press. Fat was tracked as the sum of eight skinfolds, pinches of skin and the fat beneath it measured with calipers.
In plain terms: three matched groups of trained lifters, one variable changed, the size of the surplus.
What the studies found
Muscle: no group difference. The group analysis gave no support for a difference in any muscle-thickness site1. Most sites changed little in any group. The biceps grew in all three, by a group average of about 0.2 to 0.3 cm, including at maintenance.
Squat: no group difference. Group averages rose by 10.0 kg on the bar at maintenance, 11.6 kg in the small surplus and 10.7 kg in the big surplus1.
Bench press: the big surplus gained more. Group averages rose by 13.4 kg in the big surplus, against 7.3 kg in the small surplus and 5.7 kg at maintenance1. The statistics gave strong evidence for this group difference. Yet in the regression across all 17 people, body-mass gain did not predict bench gain, so the authors put more weight on the regression.
Bench press one-rep max gain over 8 weeks · kg on the bar
Fat: rose with the surplus. The sum of skinfolds rose by a group average of 12.4 mm in the big surplus and 10.0 mm in the small one, and fell by 1.4 mm at maintenance1. There was moderate evidence that the big surplus raised skinfolds more than maintenance, and weak evidence for the small surplus.
Change in the sum of eight skinfolds over 8 weeks · mm
The pattern across everyone. Pooling all 17 lifters, faster body-mass gain strongly predicted skinfold gain (R² = 0.49, meaning it explained about half the spread)1. It only weakly predicted biceps growth (R² = 0.24) and did not predict strength or other muscle sites. The authors conclude that faster weight gain, and by proxy a larger surplus, "primarily increase[s] rates of fat gain rather than augmenting" strength or muscle thickness.
Other trials line up. In 2013, 39 elite athletes in an 8- to 12-week weight-gain period were randomised to a meal plan with a planned surplus or to eating freely2. The planned-surplus group ate more (about 3,585 against 2,964 kcal a day) and gained more body weight, but fat mass rose 15% against 3% and lean body mass gain did not differ. In a 2026 trial of 23 trained lifters, the group that ate more of its planned surplus gained no more fat-free mass than the group that ate less, and more than twice the fat5.
In plain terms: across these trials, the extra food beyond a modest surplus went mostly to fat.
The nuance
The two surplus groups ended up close. Both gained about the same body mass on average, a group average of 3.3 kg each over 8 weeks, against 0.4 kg at maintenance1. The authors write that "both MOD and HIGH, on average, were in a similar energy surplus". So the trial compares maintenance with a surplus more cleanly than it compares a small surplus with a big one.
The trial was too small for muscle. Its own power analysis estimated 31 or more people per group to detect a muscle-thickness difference; COVID-19 left 5 or 6 per group1. "No group difference" here means the trial could not see one, not proof there is none.
Bench strength did improve more with the big surplus. That result is real within the trial, even if the regression did not back it. Whether a larger surplus helps strength in the upper body more than the lower body is an open question.
Why the extra food may not become muscle. A 2019 review points out that the energy surplus needed for muscle growth is unknown and has never been validated in lifters3. Textbook estimates count only the energy stored in new tissue. That is the reviewers' framing of the gap, not a measured result.
What the research groups suggest
The 2023 trial's authors recommend "conservative energy surpluses scaled to RT experience of 5–20% over maintenance" (RT is resistance training), or weight gain of 0.25 to 0.5% of body mass a week, with more advanced lifters at the lower end1. A 2019 review of off-season bodybuilding nutrition gave a similar range: a surplus of about 10 to 20% and a gain of about 0.25 to 0.5% of body weight a week for novice and intermediate lifters, smaller for advanced ones4.
These are the authors' ranges, drawn from limited data, not tested cut-offs. For how a calorie target is set in the first place, see how many calories should I eat and what TDEE is. Whether you can add muscle without any surplus is covered in can you lose fat and gain muscle at the same time.
The studies at a glance
| Study | Who was studied | Key result | Year |
|---|---|---|---|
| Garthe et al., *Eur J Sport Sci*2 | 39 elite athletes, 8 to 12 weeks, planned surplus vs eating freely | Fat mass +15% vs +3%; lean body mass gain not different | 2013 |
| Slater et al., *Front Nutr*3 | Narrative review | The surplus needed to maximise muscle growth is unknown and never validated in lifters | 2019 |
| Iraki et al., *Sports*4 | Narrative review, natural bodybuilders | Suggested a surplus of about 10–20% and slower gain for advanced lifters | 2019 |
| Helms et al., *Sports Med Open*1 | 17 trained lifters, 8 weeks, maintenance vs small vs big surplus | Muscle thickness and squat alike; bench +13.4 vs +7.3 vs +5.7 kg; skinfolds +12.4, +10.0, -1.4 mm | 2023 |
| Gavanda et al., *J Nutr*5 | 23 trained lifters, 12 weeks, surplus with or without an eating window | The group that ate more gained the same fat-free mass and more than twice the fat | 2026 |
Limits worth knowing
- Tiny groups. Five or six lifters per group in the key trial, about a fifth of what its own power analysis wanted for muscle1.
- Surpluses not reached as planned. The small and big surplus groups gained similar body mass.
- Mostly men, 8 weeks, moderate volume. Little leg or triceps growth in any group, so a surplus effect on those muscles had little room to show.
- Skinfolds, not scans. Skinfolds track fat under the skin, not total or internal fat. The paper's table labels the sum in cm; values around 80 only make sense in mm, so we show mm.
- Food quality untested. None of these trials compared "clean" and "dirty" foods at the same surplus.
- Abstract only for the 2013 athlete trial and the two 2019 review papers. We read the 2023 and 2026 trials in full.
- Funding. Two fitness companies paid the key trial's research assistants and dietitian1.
Latest research
The newest relevant trial we opened, from July 2026, was not designed as a lean-versus-big surplus test, but it points the same way5. Both groups were told to eat 300 to 500 kcal over their needs; the group that ate closer to that target gained no more fat-free mass than the group that fell short, and gained 4.38 kg of fat against 2.00 kg on average. Its authors conclude the extra energy "primarily contributed to fat storage rather than increased muscle mass". We found no newer trial comparing surplus sizes directly. When one appears, this page will be updated and its date will change.
The short version
- In the key trial, muscle thickness grew about the same at maintenance, a small surplus and a big surplus.
- Fat under the skin rose with both surpluses and most with the big one.
- The big surplus gained more bench strength, but not more squat strength.
- Across the trials, extra food beyond a modest surplus went mostly to fat.
- The evidence is thin: tiny groups, 8 to 12 weeks, mostly men.
Educational overview only — not medical advice.
