Maybe slightly, but not clearly. Across 15 trials, training only the lowering phase grew muscle by 10.0% on average and training only the lifting phase by 6.8%, yet the gap was not statistically clear3. A larger 2025 review found no clear difference either4.
Both phases build muscle. Since a normal rep already includes both, the practical gap is smaller than the question suggests.
What the terms mean
Every rep has two main phases.
- Concentric: the lifting phase. The muscle shortens while it produces force, like curling a dumbbell up.
- Eccentric: the lowering phase. The muscle lengthens while it produces force, braking the load on the way down. Gym slang calls these "negatives".
Eccentric-only training is a lab set-up: a machine or a helper does the lifting, and the person only lowers. Concentric-only is the reverse. These set-ups let researchers isolate one phase.
Eccentric overload means a heavier load, or more time, on the lowering than on the lifting5. It is possible because muscles can resist more load while lengthening than they can lift.
A meta-analysis pools the results of many studies into one estimate.
In plain terms: the question is whether the lowering half of a rep does more of the growing than the lifting half.
What was studied
Three reviews pooled the direct comparisons.
| Review | Studies | Who | Measure |
|---|---|---|---|
| Roig 2009 | 20 randomised trials | Healthy adults | Strength and muscle mass |
| Schoenfeld 2017 | 15 | Healthy adults, at least 6 weeks | Muscle size by biopsy, scans or body composition |
| da Silva 2025 | 26 (682 people) | Apparently healthy adults | Direct imaging of muscle size |
The 2017 review required that everything except the type of muscle action be kept the same, with no blood-flow cuffs and at least 6 weeks of training3. The 2025 review accepted only direct imaging (cross-sectional area, thickness or volume) and screened 15,778 records to find its 26 studies4.
The 2026 American College of Sports Medicine (ACSM) position stand then pooled 137 systematic reviews, covering more than 30,000 participants, on how training variables affect results5.
What the studies found
Eccentric edged ahead, not clearly. In the 2017 review, the average growth across studies was 10.0% with eccentric-only training and 6.8% with concentric-only3.
Average muscle growth, eccentric-only vs concentric-only training · %
As an effect size (a difference in units of how much people vary), eccentric was ahead by 0.25, with a 95% confidence interval from −0.03 to 0.52 and p = 0.0763. A confidence interval is the range the true difference most likely falls in. This one includes zero, so the result could be chance.
The larger review found no clear difference. In the 2025 review, the pooled difference was 0.285 favouring eccentric, with a 95% confidence interval from −0.131 to 0.701 and p = 0.1794. The results varied a lot between studies (I² of 84.4%, a measure of how much studies disagree). The authors rated the certainty of the evidence "very low", the lowest grade on the GRADE scale4.
Eccentric vs concentric training: pooled difference in growth · ES
Some subgroups did favour eccentric. In the 2025 review, eccentric came out ahead for upper-limb muscles, for programmes of 8 weeks or less, when muscle thickness was the measure, and with isokinetic (machine-controlled speed) actions4. Age groups (18 to 59, and 60 and over) and programmes longer than 8 weeks showed no clear difference.
The earliest review pointed to load. The 2009 review found high-intensity eccentric training was more effective for muscle mass measured as limb girth, with a trend for cross-sectional area on scans1. Its authors linked the eccentric edge to the higher loads that eccentric actions allow1.
The newest guidance leans eccentric. The 2026 ACSM position stand concluded that muscle growth was enhanced by higher weekly volume (10 or more sets per muscle group) and by eccentric-only contractions compared with concentric-only5.
In plain terms: eccentric training may have a small edge, but the best-controlled pooled estimates cross zero.
The nuance
The two phases may grow muscle in different ways. In a 10-week trial of 12 young men, thigh muscle volume grew similarly: 8% with concentric-only training and 6% with eccentric-only2. But fibre bundles lengthened more with eccentric (12% vs 5%), while the angle of the fibres (pennation) rose more with concentric (30% vs 5%)2. The eccentric group grew more near the knee; the concentric group more in the middle of the thigh2.
How the thigh muscle changed after 10 weeks, by contraction type · %
Concentric only Eccentric only
Why the reviews disagree. The 2017 review and the 2025 review both compared eccentric and concentric training with other variables equal, and neither found a clear difference. The ACSM stand is an overview of reviews, and its search ran to October 20245. It also notes that eccentric overload is often delivered with flywheel devices, so the effect of the device and of the contraction type cannot always be separated5.
Load may be the real lever. Muscles can lower more than they can lift. When eccentric training used that extra load, it tended to do better1. Not always, though: in one trial the eccentric group trained with about 1.2 times the load and still grew a similar amount2. When loads were matched, the gap was small or absent. That is a pattern across reviews, not a tested cause.
Nobody trains one phase alone. Eccentric-only and concentric-only are research tools. A normal rep with a controlled lowering already trains both. How slowly to lower is a separate question, covered in our article on whether slow reps build more muscle.
In plain terms: a controlled lowering is part of every normal rep; isolating it adds, at most, a small and uncertain edge.
What this means in practice
The trials describe what happened in groups, not a plan for one person. In them, both lowering-only and lifting-only training grew muscle, and the authors of the 2017 review concluded that a hypertrophy programme should include both actions3.
Weekly volume had a clearer effect in the ACSM overview than contraction type did5. For how many sets that means, see how many sets per muscle group per week. For how different set structures work, see set types explained. Progress still comes from adding load or reps over time; see what progressive overload is.
The studies at a glance
| Study | Who was studied | Key result | Year |
|---|---|---|---|
| Roig et al., *Br J Sports Med*1 | 20 randomised trials, healthy adults | High-intensity eccentric training grew limb girth more; linked to higher loads | 2009 |
| Franchi et al., *Acta Physiol*2 | 12 young men, 10 weeks | Similar volume growth; eccentric lengthened fibre bundles more | 2014 |
| Schoenfeld et al., *J Strength Cond Res*3 | 15 trials, healthy adults | 10.0% vs 6.8% growth; difference 0.25 (−0.03 to 0.52), not significant | 2017 |
| da Silva et al., *J Strength Cond Res*4 | 26 studies, 682 people | No statistical difference (0.285, −0.131 to 0.701); certainty very low | 2025 |
| Currier et al., *Med Sci Sports Exerc*5 | Overview of 137 systematic reviews, over 30,000 people | Eccentric-only contractions enhanced growth vs concentric-only | 2026 |
Limits worth knowing
- Abstracts only for four of five papers. We read the relevant sections of the ACSM stand in full, but not its supplementary tables.
- Small trials underneath. The Franchi trial had 6 men per group2.
- Very low certainty. The largest meta-analysis graded its own evidence very low, with high disagreement between studies4.
- Lab set-ups. Eccentric-only training needs a machine or a helper to do the lifting; it is not how people usually train.
- Different measures. Girth, thickness, cross-sectional area and volume do not always agree.
- Mostly short programmes. The subgroup that favoured eccentric was 8 weeks or less4.
Latest research
The newest source is the 2026 ACSM position stand, which updates the society's 2009 guidance5. It lists eccentric overload among the few training variables that enhanced growth, alongside higher weekly volume. It also says that most other variables, including time under tension and training to failure, did not consistently affect outcomes. Its search ended in October 2024, and it rests on reviews whose own evidence is often low in certainty.
The short version
- Lowering-only training grew muscle 10.0% vs 6.8% for lifting-only across 15 trials, but the gap was not statistically clear.
- A larger 2025 review found no statistical difference, with very low certainty.
- The 2026 ACSM stand concluded eccentric-only contractions enhanced growth.
- The two phases may grow muscle in different ways, not just by different amounts.
- A normal rep trains both phases, so the practical difference is small.
Educational overview only — not medical advice.
