How Many Sets Per Muscle Per Week? What the Dose-Response Research Shows

The short answer: more weekly sets per muscle do produce more growth, but not in a straight line. A 2017 meta-regression of 34 treatment groups found each extra weekly set added an effect size of about 0.023 — roughly 0.37% more growth per set. A 2026 meta-regression of 67 studies and 2,058 participants confirmed the upward trend and mapped clear diminishing returns, sharper for strength than for size. A network meta-analysis of 151 trials in adults over 60 found the opposite ranking, with low volume best for muscle size. Volume matters. The curve flattens. There is no single correct number.

First: what counts as a set?

The unit in this literature is sets per muscle per week — not sets per workout, and not sets per exercise. That distinction is where most disagreements start, because the same training week can be scored very differently depending on how indirect work is handled.

Pelland and colleagues (Sports Medicine, 2026) made this the explicit first step of their analysis. Before meta-analysing anything, they classified every contributing set as direct or indirect relative to the muscle measured, then tested three ways of scoring indirect sets: full credit (1), half credit (0.5), or none (0). The evidence favoured the fractional method, and that is the model their published dose-response curves are built on.

The consequence is blunt: two people who both claim 20 weekly sets for chest may be doing very different doses if one counts every dip and overhead press and the other counts only direct pressing. Pick a convention before comparing yourself to any number below.

The shape of the dose-response curve

The 2017 meta-regression that set the agenda

Schoenfeld, Ogborn and Krieger’s systematic review and meta-analysis (Journal of Sports Sciences, 2017) pooled 34 treatment groups from 15 studies. Treating weekly sets as a continuous variable, volume had a significant effect on changes in muscle size (P = 0.002), with each additional set associated with an effect-size increase of 0.023, corresponding to about 0.37% extra percentage gain. Comparing higher versus lower volume within each study, the effect-size difference was 0.241, equating to a 3.9% difference in percentage gain (P = 0.03).

Two details are usually dropped when this paper gets quoted. First, the abstract reports that slope as a point estimate with no confidence interval, so its precision is not visible. Second, when weekly sets were binned into three categories — fewer than 5, 5 to 9, and 10 or more per muscle — the effect was only a trend (P = 0.074). The top bin was open-ended at 10+, so the model had nothing to say about 25 or 40 sets.

Bigger models, and where the returns diminish

The 2026 Sports Medicine analysis by Pelland and colleagues is the largest attempt to extend that curve. Across 67 studies and 2,058 participants (79.1% male, mean age 25.2 ± 5.2 years), with models adjusted for intervention duration and training status, the posterior probability that the volume slope exceeded zero was 100% for both hypertrophy and strength. But both best-fitting models showed diminishing returns, and the authors describe the flattening for strength as considerably more pronounced.

A narrower systematic review by Baz-Valle and colleagues (Journal of Human Kinetics, 2022) restricted itself to randomised trials of at least six weeks in people with a minimum of one year of training experience. Only seven studies qualified. Comparing moderate volume (12–20 weekly sets) against high volume (more than 20), it found no difference for the quadriceps (p = 0.19) or biceps brachii (p = 0.59), but did favour higher volume for the triceps (p = 0.01). Its conclusion — 12 to 20 weekly sets for young trained men — rests on those seven studies, a thin base for such a widely repeated number.

A single trial makes the flattening visible. Enes and colleagues (Medicine & Science in Sports & Exercise, 2024) randomised 31 trained men (mean 5.1 years of experience) to constant weekly volume, or to adding four or six sets every two weeks, across 12 weeks of lower-limb training. Squat one-rep max improved most in the six-set progression group. For muscle size there were no significant between-group differences in vastus lateralis cross-sectional area (P = 0.067) or summed thigh thickness (P = 0.076), and the authors noted the 95% confidence intervals suggested results plateauing in the higher-volume conditions.

Put together, the curve looks like this:

• Going from very little to some volume is where the biggest return sits. Every prescription tested in the largest network meta-analysis beat doing nothing.
• The middle of the range is where most of the usable evidence lives — and where the trials cluster.
• Above roughly 20 weekly sets the data thin out fast, the confidence intervals widen, and group differences stop reaching significance.
• No published model identifies the point where the curve turns downward. Diminishing returns are documented; a proven ceiling is not.

Volume for size is a different question from volume for strength

Ralston and colleagues (Sports Medicine, 2017) ran the parallel analysis for strength across 61 treatment groups from nine studies. High weekly sets beat low weekly sets, but the margin was small: 0.18 (95% CI 0.06 to 0.30; p = 0.003) for combined multi-joint and isolation exercises, and 0.15 (95% CI 0.01 to 0.30; p = 0.04) for medium versus low. For exercise-specific one-rep max, high versus low was 0.14 (95% CI −0.01 to 0.29; p = 0.06) — an interval crossing zero.

An 8-week trial in 34 trained men (Schoenfeld and colleagues, MSSE, 2019) compared one, three and five sets per exercise per session, three sessions weekly. Strength and endurance improved in all three groups with no significant between-group differences — gains the authors noted were achievable in roughly three 13-minute weekly sessions — while hypertrophy favoured the higher-volume conditions at the elbow flexors, mid-thigh and lateral thigh.

The largest synthesis available, a Bayesian network meta-analysis by Currier and colleagues (British Journal of Sports Medicine, 2023), pooled 178 studies (n = 5,097) for strength and 119 studies (n = 3,364) for hypertrophy. Every prescription beat non-exercising control. Higher-load, multiset, thrice-weekly training ranked highest for strength (SMD 1.60; 95% credible interval 1.38 to 1.82 versus control); higher-load, multiset, twice-weekly ranked highest for hypertrophy (0.66; 0.47 to 0.85). All three top-ranked hypertrophy prescriptions were multiset. Crucially, that analysis coded sets only as single versus multiple: the authors state plainly that their framework could not model weekly set count continuously, and recommend dose-response methods for future work.

Individual variation dwarfs the programme differences

The effect sizes above are averages, and the spread around them is enormous. Hubal and colleagues (MSSE, 2005) trained 585 adults (342 women, 243 men) across eight centres with an identical 12-week unilateral elbow-flexor programme, measuring biceps cross-sectional area by MRI. Size changes ranged from −2% to +59%, one-rep-max strength from 0% to +250%, isometric strength from −32% to +149%. Same programme, same supervision.

Population matters too, and it can reverse the ranking. Radaelli and colleagues (Sports Medicine, 2025) ran a network meta-analysis of 151 randomised trials (n = 6,306) in adults aged 60 and over, grouping programmes into volume terciles. Low volume ranked most effective for lean body mass (SMD 0.25; 95% CI 0.10 to 0.40) and lower-body hypertrophy (0.40; 0.25 to 0.54), while moderate and high volume were best for lower-limb strength — results independent of programme duration. That is close to the inverse of the pattern in young trained men.

Why recoverable volume is not a fixed number

A set is not a standard unit of stimulus. Two variables in particular change what a set costs.

How close to failure the set is taken. Robinson and colleagues (Sports Medicine, 2024) meta-regressed proximity to failure, expressed as estimated repetitions in reserve. In every best-fit hypertrophy model the marginal slope was negative with a confidence interval excluding the null — size increased as sets were terminated closer to failure — while the strength intervals contained a null point estimate. Sets taken to the edge cost more fatigue each, which limits how many fit into a week. Our separate article on training to failure covers that trade-off.

Which exercises the sets come from. Ten weekly sets of barbell squats and ten of leg extensions are not interchangeable doses. Compound lifts distribute fatigue systemically and, under fractional counting, quietly deposit indirect volume into muscles you did not intend to train. Identical set counts can mean very different recovery demands.

Volume is also only one variable among several. How those sets are spread across the week is a training frequency question, how the load advances over time is a progressive overload question, and how fast each repetition is performed is a time under tension question. This article deliberately holds those constant.

What this evidence does not establish

• A personal maximum recoverable volume. No trial has measured one, and no meta-analysis has tested the concept.
• Where the curve turns down. Diminishing returns are modelled; an actual point of negative return is not.
• That every muscle behaves alike. Baz-Valle and colleagues found triceps responded differently from quadriceps and biceps within the same review.
• That short-term thickness changes predict long-term muscle mass. Most trials run 6 to 12 weeks; nobody has run the multi-year comparison.
• That the findings generalise. The 2026 dose-response sample was 79.1% male with a mean age of 25.
• That higher volume is worth its cost. None of these analyses measured adherence, injury rates, time burden or enjoyment.

Limitations of this literature

This is not a strong evidence base by clinical standards, and it is worth being explicit about why.

• Small samples. The underpinning trials typically randomise 20 to 40 people across three groups, leaving each arm badly underpowered for modest differences.
• Short durations. Six to twelve weeks is standard. Hypertrophy is slow, so short trials compress the signal and inflate early neural and swelling-related changes.
• Measurement heterogeneity. Ultrasound thickness, MRI cross-sectional area and DXA lean mass get pooled as though interchangeable. They are not, and each carries its own error.
• Inconsistent volume definitions. The direct-versus-indirect problem is unresolved field-wide. Currier and colleagues reported that approximating weekly volume load actively hindered their model fit.
• Risk of bias. Using Cochrane’s RoB 2 tool, Currier and colleagues found concerns regularly raised for randomisation, deviations from intended interventions, and selection of the reported result. Blinding people to their own training volume is impossible.
• Small-study and publication effects. A field built on many small trials is structurally vulnerable to them, and most of these analyses report no formal assessment.
• Open methodological disagreement. The 2017 meta-analysis drew a formal published comment in the same journal on the reliability of meta-analysing resistance training programmes, with an author reply. That dispute remains unsettled.

Declared interests. The 2026 Sports Medicine authors disclosed that all five are coaches and writers in the fitness industry, while reporting no conflicts specific to that article. The 2023 network meta-analysis was unfunded, though its senior author discloses grants and personal fees from food and supplement companies. The Ralston and Radaelli teams each declared no funding and no conflicts.

The bottom line

The relationship between weekly sets and muscle growth is real and reasonably consistent, but it is a curve with a flattening top, not a ladder. Most of the return arrives in the low-to-moderate range the trials actually studied; beyond roughly 20 weekly sets per muscle the evidence thins, the intervals widen, and the fatigue cost keeps rising whether or not the growth does. Strength flattens sooner than size, and in adults over 60 the ranking can invert. Since between-person variation exceeded every programme difference in a 585-participant trial, the research defines a plausible range rather than a personal prescription — and the only way to find your own place on the curve is to change one variable at a time and track what happens.

This article is educational, not medical or training advice. It summarises published research and is not a substitute for guidance from a qualified coach or clinician who can assess your training history, injuries and health status.

References

• Schoenfeld BJ, Ogborn D, Krieger JW. Dose-response relationship between weekly resistance training volume and increases in muscle mass: a systematic review and meta-analysis. J Sports Sci. 2017;35(11):1073-1082. PubMed
• Pelland JC, Remmert JF, Robinson ZP, Hinson SR, Zourdos MC. The resistance training dose response: meta-regressions exploring the effects of weekly volume and frequency on muscle hypertrophy and strength gains. Sports Med. 2026;56(2):481-505. PubMed
• Currier BS, McLeod JC, Banfield L, et al. Resistance training prescription for muscle strength and hypertrophy in healthy adults: a systematic review and Bayesian network meta-analysis. Br J Sports Med. 2023;57(18):1211-1220. PubMed
• Ralston GW, Kilgore L, Wyatt FB, Baker JS. The effect of weekly set volume on strength gain: a meta-analysis. Sports Med. 2017;47(12):2585-2601. PubMed
• Radaelli R, Rech A, Molinari T, et al. Effects of resistance training volume on physical function, lean body mass and lower-body muscle hypertrophy and strength in older adults: a systematic review and network meta-analysis of 151 randomised trials. Sports Med. 2025;55(1):167-192. PubMed
• Baz-Valle E, Balsalobre-Fernández C, Alix-Fages C, Santos-Concejero J. A systematic review of the effects of different resistance training volumes on muscle hypertrophy. J Hum Kinet. 2022;81:199-210. PubMed
• Schoenfeld BJ, Contreras B, Krieger J, et al. Resistance training volume enhances muscle hypertrophy but not strength in trained men. Med Sci Sports Exerc. 2019;51(1):94-103. PubMed
• Enes A, De Souza EO, Souza-Junior TP. Effects of different weekly set progressions on muscular adaptations in trained males: is there a dose-response effect? Med Sci Sports Exerc. 2024;56(3):553-563. PubMed
• Robinson ZP, Pelland JC, Remmert JF, et al. Exploring the dose-response relationship between estimated resistance training proximity to failure, strength gain, and muscle hypertrophy: a series of meta-regressions. Sports Med. 2024;54(9):2209-2231. PubMed
• Hubal MJ, Gordish-Dressman H, Thompson PD, et al. Variability in muscle size and strength gain after unilateral resistance training. Med Sci Sports Exerc. 2005;37(6):964-972. PubMed
• Arruda A, Souza D, Steele J, Fisher J, Giessing J, Gentil P. Reliability of meta-analyses to evaluate resistance training programmes. J Sports Sci. 2017;35(20):1982-1984. PubMed

For a worked example of what this looks like on one muscle group, our guide to a shoulder workout at home with no equipment follows a trial that used five sets to failure, twice a week, and measured the resulting deltoid growth.