Loading fundamentals 07: Weekly volume

Short answer. More weekly sets build more muscle, with diminishing returns and no clear plateau in the range studied; strength plateaus early. No set count means much until you fix how indirect sets are counted. Our position is that weekly volume is a target you build toward, because the limit most people hit is recovery.

---

What it is

Most other entries in this branch resolve into this one. Rest protects it, set endpoint protects it, frequency delivers it, load and exercise selection determine what it consists of.

Which makes weekly volume the most consequential number in a program and the one hardest to state precisely. Not because the dose-response is unclear. Because the unit is.

The proposed mechanism

Mechanical tension drives growth, and more work means more tension. That is close to the whole account, and unusually simple by the standards of this branch.

What it does not specify is the shape: linearly better, better with diminishing returns, or better up to a point and then worse. Those three differ enormously in what they recommend, and until recently the literature could not distinguish between them.

What has actually been tested

A graded relationship exists, and the slope is small per set. An early analysis using direct muscle measurements reported a 0.38% increase in hypertrophy per additional weekly set (Schoenfeld et al., 2017). The most comprehensive to date estimated 0.24% per set at an average weekly volume of 12.25 sets, with a 95% credible interval (CrI, the Bayesian range within which the true value most probably lies) of 0.15 to 0.33 (Pelland et al., 2026).

For strength, the same analysis found a marginal slope of 0.21% (95% CrI 0.16 to 0.26) with a 100% posterior probability, the model's estimated chance, of exceeding zero. The published version draws on 67 studies and 2,058 participants, 79% of them male.

Both curves bend, and strength bends harder. Both best-fitting models showed diminishing returns, considerably more pronounced for strength, which reached a functional plateau. The hypertrophy curve did not. That analysis identified no clear plateau for muscle size at all, while flagging greater uncertainty at the high end where the data thin out (Pelland et al., 2026).

The ceiling that has been located sits inside a session, not inside a week. A companion analysis from the same laboratory asked where an additional set stops giving a better than even chance of exceeding the smallest detectable effect. It found that point at around eleven fractional sets (indirect sets counted as half) per session for hypertrophy and around two direct sets per session for strength (Remmert et al., 2025). That analysis is a preprint, and it is measuring per-session volume. No comparable weekly threshold has been established, because the weekly curve did not produce one.

Strength stops rewarding additional volume almost immediately. Hypertrophy keeps rewarding it past the point anyone has measured, which is not the same as forever.

The counting convention changes the answer. The analyses had to decide how to count a set that trains a muscle indirectly: a row trains the biceps, a bench press the triceps. Three conventions exist, counting indirect sets fully, at half, or not at all. The authors ran all three and found the choice materially changes the estimate, with fractional and direct fitting best.

This means "how many sets should I do" has no answer until you say how you are counting. Ten direct sets, ten fractional sets and ten total sets describe three different training weeks. Nearly every recommendation in circulation omits this, including most of the ones citing these same papers.

On the popular volume landmarks. You will encounter a framework naming a minimum effective, a maximum adaptive and a maximum recoverable volume. It is a practitioner framework rather than a research finding, and the per-muscle numbers attached to it are not something we could trace to a primary source. The idea that returns diminish is supported above.

And the unit itself is unstable. A curve in sets per week treats one set as one set. The other entries in this branch spend much of their length explaining why that is not true.

A set to failure and a set stopped three short deliver different amounts and cost different amounts of what follows. A light set of fifteen places its stimulus elsewhere than a heavy eight. A full range and a top-half range are not the same set. A set performed sixth is performed by a muscle with far less capacity than the first.

None of that is captured by counting, and the analyses cannot adjust for what the source studies did not record. The slope is real. The x-axis is coarser than it looks.

Our reading

Volume is the one variable in this branch where more is reliably better, and the reliability is what makes it dangerous. Everything else here has a ceiling, an inversion or a null. Volume does not, within the range studied. That is why the rest of the branch points at it, and why the failure mode is predictable: volume is the easiest thing to add and the hardest to recover from.

The two curves diverge and most programs ignore it. Strength reaches a functional plateau early and hypertrophy does not appear to reach one within the range studied. Whatever the exact thresholds turn out to be, a strength block and a hypertrophy block should not carry similar volumes.

The counting problem is the whole prescription. Four sets of bench press and three of dip is seven direct chest sets and somewhere between three and seven triceps sets depending on convention. Someone following a fifteen-set recommendation without deciding this can be doing anywhere from nine to twenty-two.

So our position is that weekly volume should be a target you build toward, not a number you jump to, and that the counting convention should be fixed before the number is. The ceiling most people hit is recovery, not the dose-response curve. The curve says more works. It does not say you can absorb it, and it does not know what your sets consist of.

This is one of the reasons most OMNID programs open below their peak weekly volume and build across the block, instead of prescribing a fixed weekly figure from session one.

Where the goals diverge

Building size. The dose-response runs further than most people train, and volume is a currency of progression in its own right. (→ Loading fundamentals 05: Progressive overload)

Building strength. The curve flattens early. Volume beyond a modest amount buys very little, and the recovery it costs is recovery the heavy work needs. (→ Loading fundamentals 04: Load and repetition range)

Limited recovery capacity. Sleep, stress, age and training history bound what you can absorb, and none of it appears in a curve built from group means. (→ Sleep 01: Sleep and muscle)

Coming back from a layoff. (→ Loading fundamentals 10: Time off and coming back) Far less volume maintains muscle than builds it, and rebuilding does not require returning to your previous volume immediately.

Distributing it. Warm-up sets also draw on the same session budget. (→ Loading fundamentals 08: Warm-up) The per-session ceiling means high weekly volume cannot be delivered in one or two sessions. (→ Loading fundamentals 03: Training frequency; Loading fundamentals sub-hub)

The case against

The upper end of the curve is thinly populated. Very few trials prescribe more than twenty weekly sets per muscle, so the region past that is model extrapolation rather than observation.

The per-session ceiling comes from a preprint, and an earlier version of this entry printed the wrong numbers for it and described them as weekly. They are per-session, and they are eleven and two. We name it because it is the exact error this publication exists to flag: a threshold detaching from the condition it was measured under.

These analyses come from one laboratory network. The weekly work and its per-session companion share an author group, and the authors disclosed that several of them have previously produced commercial fitness content. The disclosure counts in their favor; the concentration still limits how far the findings travel without independent replication.

We used a unit the rest of this branch spends its time qualifying. Every conclusion here is expressed in sets, and other entries argue sets are not interchangeable. We flagged that rather than solved it, because the literature is built on the same unit.

Group means hide the thing you most want to know: where your own returns flatten.

Duration is short relative to the question. Eight to twelve weeks cannot tell you whether a high-volume approach stays productive across a year.

What is not known

Where the inverted-U begins, if it exists. Diminishing returns are demonstrated; an actual decline at very high volumes is proposed and not established.

Whether volume tolerance is trainable.

Which counting convention best reflects the stimulus. The analyses say the choice matters and that fractional and direct fit better than total. They do not say which is correct.

How recovery capacity should modify the target, which is the question most readers have.

Whether a better unit than the set exists. Effort-adjusted or tension-time-adjusted alternatives have been proposed and none has been used consistently enough across trials to build a dose-response on.

The practical warning

Tendon and connective tissue adapt more slowly than muscle and far more slowly than the dose-response curve suggests you can load them. Volume added faster than that is the standard route into an overuse injury, and the joint usually complains before the muscle does.