How Many Sets Per Week for Muscle Gain? The Schoenfeld Meta-Analysis Data

Ask ten lifters how many sets per muscle group they do each week and you will get ten different answers, most of them guesses. The research here is fairly mature, anchored by a 2017 meta-analysis from Brad Schoenfeld and colleagues testing whether weekly set volume predicts hypertrophy, followed by a direct 2019 trial from the same lab group and a 2022 update that pushed the analysis into much higher volume ranges. This article walks through what those papers measured, where the dose-response relationship holds and breaks down, and how to translate the numbers into a real weekly split. Volume and protein intake are the two variables with the strongest evidence behind them for hypertrophy, so if you have not already read our breakdown of how much protein you need for muscle gain, the two articles pair well together.

The Schoenfeld 2017 meta-analysis: 15 studies, a graded dose-response curve

The full citation is 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," Journal of Sports Sciences, 2017 (PubMed record). The final analysis pooled 34 treatment groups drawn from 15 studies, all measuring some form of muscle hypertrophy outcome against a quantified weekly set volume per muscle group.

Three separate statistical models were run on the same pooled data, and all three pointed the same direction. Treating weekly sets as a continuous variable, each additional weekly set was associated with an effect size increase of 0.023, corresponding to a 0.37 percent increase in hypertrophy, significant at p equals 0.002. Splitting studies into a lower-volume versus higher-volume group within each study, the difference in effect size between the two was 0.241, equal to a 3.9 percent difference in muscle gain, significant at p equals 0.03. The third model split weekly sets into three categorical bands, fewer than 5, 5 to 9, and 10 or more sets per muscle per week, and found a trend toward a stepwise effect, at p equals 0.074, just short of significance but consistent in direction with the other two models (Schoenfeld, Ogborn, and Krieger, 2017). The paper's own conclusion is direct: the findings indicate a graded dose-response relationship whereby increases in resistance training volume produce greater gains in muscle hypertrophy, with the more-than-10-sets group outperforming the fewer-than-5-sets group across the pooled data.

It is worth being precise about what the 2017 paper could and could not establish. Because so few of the included studies tested volumes above 10 weekly sets, the authors could only model the relationship with confidence up to roughly that point. A separate 2022 umbrella review pooling 14 meta-analyses and 178 primary studies summarized this limitation, concluding that low volumes of 4 or fewer weekly sets per muscle group are enough to produce substantial hypertrophy, but that at least 10 weekly sets appears necessary to maximize gains within the range the original data could test (Bernardez-Vazquez et al., Frontiers in Sports and Active Living, 2022). This is also why the follow-up trial described next was designed the way it was: to directly test volumes meaningfully above the 10-set ceiling the 2017 meta-analysis could not fully resolve.

The 2019 follow-up: 1, 3, and 5 sets per exercise in trained men

Schoenfeld and a larger author group, including James Krieger and Jozo Grgic, ran a direct trial to test the dose-response question head-on: "Resistance Training Volume Enhances Muscle Hypertrophy but Not Strength in Trained Men," published in Medicine and Science in Sports and Exercise in 2019 (full text, PMC). Forty-five resistance-trained men were recruited, with 34 completing the 8-week protocol after 11 dropouts, split into a low-volume group doing 1 set per exercise per session (n=11), a moderate-volume group doing 3 sets per exercise per session (n=12), and a high-volume group doing 5 sets per exercise per session (n=11). Across 7 exercises and 3 weekly sessions, that worked out to roughly 6 to 9 weekly sets per muscle group in the low-volume condition, 18 to 27 sets in the moderate condition, and 30 to 45 sets in the high-volume condition, with every set taken to momentary concentric failure in the 8 to 12 rep range.

The hypertrophy results favored higher volume at every site measured. Elbow flexor, mid-thigh, and lateral thigh thickness all showed statistically significant differences between groups, with post-hoc comparisons showing the 5-set condition beat the 1-set condition at every site, most strongly for the lateral thigh, where the Bayes factor in favor of 5 sets over 1 set was 38.14, indicating strong evidence (Schoenfeld et al., MSSE, 2019). In raw terms, lateral thigh hypertrophy was roughly 5.0 percent in the low-volume group versus 13.7 percent in the high-volume group, and mid-thigh hypertrophy went from about 3.4 percent to 12.5 percent across that same comparison. Strength told a different story: squat and bench press one-rep max gains were statistically indistinguishable across all three volume conditions, with the data favoring the null model of no difference. The practical takeaway is now one of the most replicated findings in the volume literature, that muscle size follows a dose-response relationship with weekly sets in a way that strength largely does not once training is taken close to failure.

Baz-Valle 2022 and the picture above 20 sets per week

A separate 2022 meta-analysis from Eneko Baz-Valle and colleagues, published in the Journal of Human Kinetics, pushed the volume question into territory the original Schoenfeld dataset could not reach (full text PDF). The review screened down to 7 studies for the qualitative synthesis and 6 studies for the quantitative meta-analysis, deliberately restricted to trials that measured hypertrophy directly through muscle thickness or cross-sectional area rather than indirect proxies. Studies were grouped into low volume, defined as fewer than 12 weekly sets per muscle group, moderate volume at 12 to 20 weekly sets, and high volume at more than 20 weekly sets.

Comparing moderate against high volume, the quadriceps and biceps brachii showed no statistically significant additional benefit from pushing past 20 weekly sets, at p equals 0.19 and p equals 0.59 respectively. The triceps brachii was the exception, with a significant advantage above 20 weekly sets, at p equals 0.01. The authors concluded that 12 to 20 weekly sets per muscle group, trained across at least 2 sessions per week, is a reasonable standard recommendation for young, trained men, with smaller muscles like the triceps possibly continuing to respond past that point. Taken together with the original Schoenfeld data, the three papers form a consistent arc: meaningful gains from roughly 4 to 10 sets, continued but smaller gains from 10 to 20 sets, and a muscle-specific, individual picture above 20 sets, where more volume is rarely harmful but still has to be recovered from.

Junk volume, hard sets, and why not every set counts the same

None of these numbers mean anything without a shared definition of what counts as a set. The volume literature generally defines a countable set as one taken close to failure, commonly operationalized as a Reps in Reserve, or RIR, of roughly 0 to 3, since a set stopped 5 or more reps short of failure has not been shown to reliably drive the mechanical tension and motor unit recruitment that seem to underpin the dose-response relationship (Baz-Valle et al., 2022). A separate systematic review by the same lead author on total number of sets as a volume-quantification method concluded that counting sets taken to failure, or close to it, is valid specifically when the rep range stays between roughly 6 and 20-plus reps and other variables are held constant (Baz-Valle, Fontes-Villalba, and Santos-Concejero, Journal of Strength and Conditioning Research, 2021). This is the origin of the term "hard sets" used across modern programming: a hard set is one within a few reps of failure, and warm-up sets, technique sets, or sets stopped far short of fatigue generally should not count toward your weekly total.

This is also where the concept of junk volume comes in: sets that add fatigue and recovery cost without a proportional hypertrophy stimulus, typically because they were taken too far from failure, used too light a load or range of motion, or were piled on past the point a lifter's recovery capacity could absorb them. A separate meta-analysis comparing training to failure against non-failure training found no significant overall difference in hypertrophy across 15 studies, but did find a significant benefit to training to failure specifically within resistance-trained subjects (systematic review and meta-analysis, 2022). The practical reading is not that every set must reach absolute failure, but that sets need to be genuinely challenging to count toward your weekly volume target, since inflating your set count with easy, far-from-failure work will not reproduce the hypertrophy numbers reported above.

Counting sets across compound and isolation work: fractional counting

A single training session rarely trains one muscle in isolation, which creates a real counting problem. A barbell bench press is a direct set for the chest, but it also loads the front delts and triceps, just not to the same degree as a set of lateral raises loads the side delts or a set of triceps pushdowns loads the triceps. Recent methodological work addressing this classified sets as direct, fully counting toward the target muscle, fractional, counting as roughly half a set toward a secondary muscle, or indirect, not counted at all, and found the fractional method fit the dose-response data best across a pooled analysis of 67 studies and 2,058 participants (meta-regression analysis, 2026). In practice, a set of bench press should count as a full set for chest but only a fractional set toward triceps and front delt volume, while a set of triceps pushdowns counts as a full direct set for triceps. This fractional approach is the most defensible way to reconcile compound-heavy programs with the per-muscle set targets in this article, rather than double-counting a compound lift across 3 muscles or ignoring its contribution to secondary muscles entirely.

Frequency: why 2 sessions per muscle per week beats 1

Volume and frequency are related but distinct, and how to split, say, 16 weekly sets across the week has its own research. Schoenfeld, Ogborn, and Krieger's 2016 meta-analysis in Sports Medicine pooled 10 studies and found higher training frequency was associated with a significantly larger hypertrophy effect size than lower frequency, 0.49 versus 0.30, at p equals 0.002 (PubMed record). Comparing studies that held weekly volume constant while varying only frequency, training a muscle group twice per week produced superior hypertrophy to training it once per week, leading the authors to conclude major muscle groups should be trained at least twice weekly, while any benefit of a third weekly session remained unresolved.

A later, more targeted meta-analysis on this question, once more data existed to properly volume-equate the comparison, found no significant difference between higher and lower frequency once weekly volume was held constant, concluding that frequency does not meaningfully affect hypertrophy independent of volume, and that lifters can choose a frequency per muscle group based on preference and scheduling as long as total weekly volume is matched (PubMed record, 2019). The practical synthesis of both papers is that frequency mostly matters as a vehicle for volume: if you want to hit 16 to 20 sets per muscle group per week, spreading that across 2 sessions rather than cramming it into 1 is usually easier to recover from, even though frequency itself is not an independent driver of growth once total volume is equal.

Programming by training age: how many sets you actually need

Combining the dose-response data above with common findings on individual variation and recoverable volume gives a reasonable, if intentionally broad, set of starting points by training experience.

Beginners generally do well starting around 10 sets per muscle group per week, matching the point in the original Schoenfeld 2017 data where the 10-plus set category showed a clear advantage over fewer than 5 sets, while newer lifters also tend to show larger absolute hypertrophy responses per set than trained lifters, meaning the efficient, lower-volume end of the curve is enough for measurable early progress. Intermediate lifters, those with at least 6 to 12 months of consistent training, generally do best in the 12 to 20 weekly set range identified by the Baz-Valle 2022 review, split across at least 2 weekly sessions per the frequency data above. Advanced lifters face the most individualized picture: the Baz-Valle data suggests some muscles, like the triceps, may keep responding above 20 weekly sets, but the practical ceiling for any individual is set less by the muscle's theoretical capacity to grow and more by how much volume that person can recover from given sleep, stress, nutrition, and training age, which is why individual variation in optimal set count is larger at the advanced end of the spectrum than the beginner end.

Nutrition sets the ceiling on how much of that volume can actually be converted into new muscle tissue, which is the other half of the hypertrophy equation covered in our protein for muscle gain guide, and getting enough protein at each individual meal, not just across the whole day, matters too, which we cover separately in our protein per meal article. You can set both your protein target and your total calories against your own body weight and goal using the Scinergy macro calculator, which pairs naturally with a set-volume plan like the one below.

Worked example: a 4-day upper/lower split at 12 to 16 sets per muscle

Consider an intermediate lifter targeting 12 to 16 weekly sets for chest, back, quads, and hamstrings, using a 4-day upper/lower split that trains each muscle group twice per week, consistent with the frequency data above.

Upper day 1: 4 sets of barbell bench press, counted as 4 direct chest sets plus 2 fractional sets toward triceps and front delts, plus 4 sets of chest-supported rows, 4 direct back sets plus 2 fractional sets toward biceps. Upper day 2: 4 sets of overhead press, contributing fractionally to chest and triceps, plus 4 sets of lat pulldown, a direct back set with a fractional biceps contribution. Across the 2 upper sessions, chest lands around 12 to 14 chest-equivalent sets once fractional credit is included, and back lands in the 14 to 16 set range once rows and pulldowns are combined.

Lower day 1: 4 sets of back squat, counted as 3 to 4 direct quad sets plus a fractional hamstring contribution, plus 3 sets of leg curl as a direct hamstring set. Lower day 2: 4 sets of Romanian deadlift, a direct hamstring and glute set, plus 4 sets of leg press, contributing directly to quads. Combining both lower sessions gives roughly 12 to 14 sets for quads and a similar 12 to 14 for hamstrings once the squat's fractional contribution and the direct sets are added together. This lands every major muscle group in the 12 to 16 weekly set range identified by the Baz-Valle 2022 review, trained twice per week per the Schoenfeld 2016 frequency data, with every working set taken to a 0 to 3 RIR so it counts as a hard set rather than junk volume.

From there, add roughly 1 to 2 sets per muscle group every few weeks as recovery allows, watching for the point where performance across sessions stagnates, the practical signal that you have reached your current recoverable volume ceiling rather than a fixed number that applies to everyone. Pair that progression with a protein and calorie target built around your own body weight in the Scinergy macro calculator, since the volume numbers here assume the muscle protein synthesis machinery has enough amino acids and calories on hand to respond to the training stimulus you are giving it.