How Many Sets Per Week to Build Muscle

Ask how many sets you should do and you will get a number. Ten. Twenty. “As many as you can recover from.” All three answer a question that was asked wrong. Sets are not a per-session quantity. The variable that tracks with muscle growth is how many hard sets a muscle receives across a whole week — and once you count that way, most arguments about splits collapse into arithmetic. Here is what the research supports, where the landmark numbers come from, and how to build a week you can finish.

What actually counts as a set

A set counts toward a muscle’s weekly total when it was taken close enough to failure that the muscle had to work near its capacity. Warm-ups do not count. Sets you cut four reps short because the clock ran out do not count. What you are tallying is hard working sets, not the total number of times you touched a barbell.

Count per muscle rather than per exercise. A barbell row is back volume and also some biceps volume. A bench press is chest volume plus a meaningful amount of triceps and front delt. The usual convention is to count direct work fully and indirect work at roughly half. It is imprecise, and that is fine: the number’s job is to be comparable with your own number from last month. Consistency beats accuracy here.

And count across the week, not the session. Three hard sets of chest on Monday and three more on Thursday is six weekly sets. The six is the figure that means something; how it was split is a smaller question.

What the evidence actually shows

The clearest thing the literature says about volume is that it behaves like a dose. A meta-analysis of studies comparing different weekly set counts found a graded relationship: groups performing more weekly sets per muscle gained more muscle than groups performing fewer (Schoenfeld et al., J Sports Sci 2017). That dose-response finding is what almost all of the internet’s volume advice is downstream of, usually without the caveats.

The caveats matter. The same body of work shows returns shrinking as sets accumulate — each additional set buys less than the one before it. A later systematic review of how volume has been manipulated across the literature reached a similar conclusion, while noting how much individual studies differ in who was trained and how growth was measured (Baz-Valle et al., J Hum Kinet 2022).

So the honest summary is a shape rather than a number. Growth rises steeply across the first several weekly sets, rises more slowly through the middle, and eventually stops paying for itself. Where your own curve flattens depends on training age, the muscle, and how much recovery capacity the rest of your life is already spending.

Turning the curve into numbers

A curve is useful and almost impossible to train from, because it does not tell you what to write in a program. The framework most lifters use to bridge that gap is Renaissance Periodization’s volume landmarks — MV, MEV, MAV and MRV, popularised by Mike Israetel. It is a coaching model for organising training, not a trial result: a vocabulary for a shape the research supports, with starting numbers attached.

There are four terms. Maintenance volume is the least you can do and still keep what you have. Minimum effective volume is the least that reliably produces growth. Maximum adaptive volume is the range where each week returns the most. Maximum recoverable volume is the point past which you are accumulating fatigue instead of muscle. Drag through them below.

Two caveats the chart cannot carry alone. The landmarks are individual — yours differ by muscle and by training age. And they move: the same lifter has a lower ceiling during a hard diet than during a surplus.

Where to start, and when to add

Start lower than you think you should, then earn the rest. If you are new, the bottom of the productive range is plenty; untrained muscle responds to almost anything organised and repeated. If you have trained for years, the useful question is not “how many sets should I do” but “what is the smallest increase that would restart progress.”

A workable rule: pick a starting number, run it for four to six weeks while tracking loads and reps, and add volume only when progress stalls at a bodyweight and a recovery situation that have not otherwise changed. Add a set or two per muscle per week, not five, and hold it long enough to read the result. For the mechanism you are trying to keep alive underneath all of it, see progressive overload.

Spreading the sets across the week

Once you have a weekly number, frequency is only the question of how to slice it. When researchers hold weekly volume constant and change how often each muscle is trained, the frequency effect is small (Schoenfeld et al., J Sports Sci 2019). Weekly volume does most of the work.

That does not make frequency pointless. It makes it logistical. Twenty sets of quads in one session is a bad session: the last sets are performed tired, at lower loads, and cost more recovery than they buy. The same twenty across two or three days are simply better sets. Frequency’s real job is to keep your weekly volume made of sets worth counting — which is also the honest answer to how many days per week you need to train.

Which is all a training split really is — a way of distributing weekly sets:

Those rows are arithmetic, not prescriptions. They show that three very different-looking splits land in much the same weekly territory, which is why arguing about splits in the abstract wastes a good evening. Pick the one whose session length you will actually keep — full-body splits and push-pull-legs are two ends of the same sum.

Sets only count if they are hard

Volume counted without regard for effort is the fastest way to fool yourself. A systematic review with meta-analysis of proximity-to-failure studies found that training closer to failure tends to favour hypertrophy, while strength outcomes were less sensitive to it (Refalo et al., Sports Med 2023).

The practical version: a set finishing with five reps left in the tank is a warm-up wearing a working set’s clothing. Most hard sets should end within roughly one to three reps of failure. That is not an argument for grinding every set to absolute failure — a set taken to failure early in a session quietly taxes everything after it.

It is also why two lifters can both write “sixteen weekly sets” in a spreadsheet and get completely different results. One is doing sixteen hard sets. The other is doing six hard sets and ten rehearsals.

When more sets stop working

Every set is a bill recovery has to pay. The inputs deciding how much volume you can absorb are unglamorous: sleep, calories, protein, and the stress the rest of your life is charging you. Protein has the cleanest evidence — a meta-analysis and meta-regression found protein supplementation augments resistance-training-induced gains in muscle mass and strength, levelling off beyond a point (Morton et al., Br J Sports Med 2018). More in how much protein you actually need.

The signs you have crossed your own ceiling are easy to ignore: loads drifting down rather than up, joints complaining before muscles do, sessions you quietly dread, sleep getting worse in a block that is supposedly working. That is not a moral failure. It is a volume prescription that has outrun its recovery budget.

The fix is boring and fast. Cut volume for a week, keep the loads and the frequency, let the fatigue clear, then return slightly below where you broke. Recovery is the limiting factor more often than programming is.

What this looks like in a real program

The point of all of this is that you should not be counting sets in your head at 6am. A written program does the counting, distributes the volume, and tells you when to add.

With three days, Full Body (3x) reaches every muscle three times a week and gets a productive weekly total out of the fewest sessions. With four, Upper / Lower (4x) runs a middle-of-the-range volume without three-hour sessions. If you genuinely train five or six days, Push / Pull / Legs (6x) spreads a higher weekly total thinly enough that the last set of each session is still a real one. All three, plus dumbbell-only and bodyweight versions, sit in the OPTML Method library.

Bottom line

Count hard sets per muscle per week. The research supports a dose-response with diminishing returns, not a magic number, and the landmark framework everyone quotes is a coaching map laid over that shape rather than a finding in its own right. Start at a number you can recover from, hold it long enough to read the result, add slowly when progress stops, and make sure every set you counted was actually hard.

Educational content, not medical advice. Individual results vary.