Ten or more weekly sets per muscle group produced roughly double the hypertrophy of one set or fewer in a 2017 dose-response meta-analysis by Schoenfeld and colleagues, and a 2022 meta-analysis by Baz-Valle and colleagues found gains continued rising beyond that mark with diminishing returns. A defensible evidence-based range for most trained lifters sits between about 10 and 20 hard weekly sets per muscle, though the upper edge of the curve is extrapolated, not measured.
24 News Click publishes information, not medical or training advice. Individual programming decisions belong with a qualified exercise professional and, where health conditions exist, a clinician.
Why are sets counted per muscle per week?
Weekly sets per muscle became the standard dose unit because it survives comparison across studies. A 2017 meta-analysis by Schoenfeld, Ogborn, and Krieger in the Journal of Sports Sciences grouped trials by weekly set counts per muscle group — one or fewer, around four to six, and ten or more — and found hypertrophy rose with each step, with the highest group showing roughly twice the effect of the lowest. Counting sets, unlike counting kilograms lifted, does not automatically change when loads change.
The count includes only hard sets: sets taken within a few repetitions of failure. Warm-up sets, technique sets stopped far short of effort, and work at unrelated muscles do not enter the denominator. That single rule explains most of the disagreement between research numbers and training-log numbers.
Compound exercises complicate the arithmetic. A chin-up set counts for the lats and, at lower effective dose, for the biceps. Reviews handle this by assigning fractional credit — a convention, not a measured biological fact, but a consistent one.
Is there a proven minimum effective dose?
Yes, and it is low. Single-set programs produced measurable hypertrophy in several trials included in the 2017 meta-analysis, and a 2017 analysis by Ralston and colleagues found multi-set programs outgrew single-set programs by a clear margin — but single sets still grew. The minimum effective dose for maintaining existing muscle is lower still: maintenance research summarised in position stands suggests roughly a third to a half of building-phase volume can hold muscle in trained people.
The practical floor for building, drawn from the grouped meta-analysis data, sits around four to six hard weekly sets per muscle. Below that, growth occurs but slowly. Above it, each added set buys less than the previous one — the classic diminishing-returns shape the Baz-Valle 2022 analysis documented across controlled trials.
Where do returns start diminishing?
Somewhere above ten weekly sets, and the evidence beyond twenty is thin. The Baz-Valle 2022 meta-analysis, which examined set dose-response across randomized trials, reported that more sets kept producing more growth within the ranges studied, with a flattening slope per set. Few controlled trials prescribe more than about twenty-plus weekly sets per muscle, largely because recovery and session time make such programs hard to run — so the curve's true ceiling is unknown.
Recovery capacity is the usual practical ceiling before any measured one. Studies in volume-tolerance research report that extreme set counts raise fatigue and dropout. The honest reading: ten to twenty hard weekly sets per muscle captures most of the measurable response for most studied lifters, and pushing higher trades small expected gains against real recovery costs.
Does frequency change the set target?
Not the total, according to the best-matched evidence. Morton and colleagues reported in 2016 that when weekly volume was equated, training a muscle once versus multiple times per week produced similar hypertrophy. Grgic and colleagues' 2018 meta-analysis on frequency reached a compatible conclusion: frequency matters mainly as a tool for distributing volume, with small possible benefits for higher totals. Four sets done four days apart and sixteen sets spread across the week can represent the same weekly dose.
Frequency earns its keep in two places: making high set totals physically possible — twenty sets in one session is impractical — and allowing quality to stay high across sessions. The total is the dose; frequency is the delivery schedule.
Do these numbers apply to everyone?
Largely to young, mostly male, resistance-trained volunteers — the population that dominates the trial base. The direction of the dose-response appears in older adults too, with the Borde 2015 dose-response work in older adults supporting progression and volume, but the specific set thresholds were not established there. Data in women is growing and, in available trials, points the same direction.
Individual response also spreads widely. In the Morton trial and similar studies, growth rates varied several-fold between participants on identical programs. The meta-analytic numbers describe group averages, and no test tells an individual where their personal curve peaks.
What counts as a hard set when adding up the week?
A hard set is one that ends within roughly zero to three repetitions of failure, the range used in most of the trials behind the meta-analyses above. Zourdos and colleagues showed in 2016 that trained lifters can estimate repetitions in reserve with usable accuracy, which is how researchers standardized effort across studies. Sets stopped with five or more reps still in the tank generally do not qualify as the stimulus being counted.
Load choice, within the ranges discussed in the load literature, does not disqualify a set. A set of six at a heavy load and a set of fifteen at a moderate load both count if each approaches failure. What breaks the count is hidden effort inflation: rest-pause clusters, partial reps, and heavily assisted reps are handled differently across studies, so the meta-analytic numbers assume conventional, full-range, unassisted repetitions.
Consistency of counting matters more than precision. A lifter who counts only conventional sets taken near failure can compare their own weeks against the evidence-based ranges; a lifter who counts everything cannot.
Weekly set targets by training phase, per the meta-analyses
| Phase | Hard weekly sets per muscle | Evidence basis |
|---|---|---|
| Maintenance | ~1/3–1/2 of building volume | Position-stand summaries of maintenance research |
| Minimum building dose | ~4–6 | Schoenfeld 2017 grouped meta-analysis |
| Robust building range | ~10–20 | Schoenfeld 2017; Baz-Valle 2022 |
| Beyond 20 | Diminishing, under-studied | Few controlled trials exist |
FAQ
Is there one ideal number of weekly sets?
No. The 2017 Schoenfeld meta-analysis found ten or more weekly sets roughly doubled growth versus one set or fewer, and the 2022 Baz-Valle analysis found continuing gains with more sets. The research describes a rising, flattening curve — not a single optimum.
Do compound exercises count for multiple muscles?
In research accounting, yes, with fractional credit: a chin-up set counts fully for the lats and partially for the biceps. This is an accounting convention used to compare studies, not a precisely measured biological split.
Does spreading sets over more days help?
Volume-equated comparisons, including the 2016 Morton trial and Grgic's 2018 frequency meta-analysis, found similar growth across frequencies. Frequency helps mainly by making higher weekly totals practical and keeping set quality high.
Can someone grow on very few sets?
Yes — single-set protocols produced measurable hypertrophy in trials included in the 2017 meta-analysis. Growth is faster with more sets up to the diminishing-returns region around ten to twenty weekly hard sets.
For more context, read Training volume or intensity: which matters more for muscle growth?.
For more context, read progressive overload.
For more context, read Rest intervals between sets: what does the evidence show?.