Evidence-Based Hypertrophy Program: Sets, Reps, Frequency

An evidence-based hypertrophy program comes down to four numbers: about 10 to 20 hard sets per muscle per week, sets taken within a few reps of failure, any load from roughly 30% to 85% of 1RM, and enough weekly sessions to spread that volume.^[1][3][10] For powerlifting strength the same base applies, plus heavier loads, more frequent practice of the competition lifts and structured periodization.^[4][10] This article ties the meta-analyses behind those numbers into a framework you can apply, with explicit caveats where the evidence thins out.

Volume: the strongest lever

The clearest effect in resistance training is also the most often overstated on social media. Schoenfeld, Ogborn, and Krieger's 2017 meta-analysis of 15 studies (34 treatment groups) found a graded dose-response relationship between weekly set volume and hypertrophy: each additional weekly set was associated with about 0.37% more muscle growth, and groups doing 10 or more sets per muscle trended highest.^[1] Pelland and colleagues' 2026 meta-regressions of 67 studies and 2,058 participants confirmed that size and strength both keep rising with volume, with diminishing returns that are much steeper for strength than for size. Counting a set that only trains a muscle indirectly (a bench press for the triceps) as half a set fitted the data best.^[10]

A hard set, for the purposes of programming, is one taken to RPE 7 or higher — roughly three reps in reserve (RIR) or fewer.^[8] Warm-up sets, back-off sets at RPE 5, and high-rep burnout sets you abandon at RPE 4 simply don't count toward the weekly total that the meta-analyses were measuring. How close to failure a set needs to run is its own question, and training to failure vs reps in reserve maps where size and strength diverge. This single accounting error explains a large fraction of programs that “look like 25 sets but feel like 8.”

How to count volume in practice

The Workout Volume Calculator sums your weekly sets per muscle group and marks each group as under 10, within 10-20, or over 20 sets. It counts every set you enter and assigns each exercise to one muscle group, so log only hard working sets and add indirect work yourself. A sensible weekly budget for an intermediate lifter targeting hypertrophy looks like this:

The “upper plausible” column is based on individual-response data that meta-analyses tend to average out. This table is a programming heuristic, not a study result: the meta-analyses report a curve with diminishing returns, not a ceiling per muscle. Some lifters tolerate more than 20 sets; others recover poorly well below it. If a muscle group isn't growing, check whether volume is inside the typical range before anything else. The dedicated breakdown of how many sets per week build muscle turns the dose-response curve into per-muscle targets.

Intensity: a band, not a point

The question “what rep range is best for hypertrophy?” is the wrong one. Schoenfeld, Grgic, and colleagues' 2017 review compared loads below 60% 1RM with loads above 60% 1RM and found no meaningful difference in hypertrophy outcomes when sets were taken close to failure.^[3]

What you can't substitute is effort close to failure: the trials that found light loads build as much muscle took every set to momentary failure.^[3] Five sets of 30 stopped at RPE 5 is not the same stimulus as three sets of 10 at RPE 9, even though it is five times the reps. For strength as distinct from size, load does matter: the same meta-analysis found larger 1RM gains with heavy loads,^[3] so you need the heavy work specifically. The full case on light vs heavy loads for muscle growth covers where light loads stop working and why heavy still wins for 1RM.

Practical load ranges

Goal              Load (% 1RM)    Typical reps    Sets close to failure
─────────────────────────────────────────────────────────────────────
Hypertrophy       30 – 85%        6 – 30          Yes (RPE 7–9)
Max strength      82 – 95%        1 – 5           Not every set
Power             40 – 65%        3 – 5 explosive No (bar speed matters)
General fitness   50 – 75%        8 – 15          Sometimes (RPE 6–8)

The RPE to Percentage Converter maps RPE back onto bar load so you can regulate intensity session-to-session instead of re-testing your 1RM every six weeks. For a main lift, a useful default is to program in RPE and verify with the percentage the bar actually landed at.

Frequency: matters less than you think, provided volume is equated

Schoenfeld, Ogborn, and Krieger's 2016 meta-analysis suggested that training a muscle twice per week beat once per week.^[2] Their larger 2019 update of 25 studies reversed the emphasis: with weekly volume equated, frequency did not significantly or meaningfully change hypertrophy.^[11] Pelland 2026 agrees for size, but found strength gains still rise with frequency, with diminishing returns.^[10]

The implication for programming is that frequency is a tool for distributing volume: a 4-day upper/lower or push/pull/legs split hits each muscle twice and lets you fit 10 to 20 sets without 10-set sessions. For powerlifters, practising the competition lifts more often is one of the few levers that improves strength without adding muscle.

Specificity and progressive overload

Progressive overload is the most frequently cited and most frequently misunderstood principle in strength training. It is not synonymous with “add weight to the bar.” The ACSM's position stand progresses training through several variables, not load alone: the load, the volume of sets and repetitions, rest periods, repetition velocity and frequency.^[7] For a lifter tracking the main lifts, four levers do most of the work: three of those variables, plus technique, which is our addition rather than an ACSM variable:

1. Load. Same reps and sets, more weight. The default for novices.
2. Volume. Same load, more sets or reps. Dominant for intermediate hypertrophy work.
3. Density. Same work in less time — shorter rest periods. Metabolic stress pathway.
4. Technique. Same external output, tighter mechanics. Frequently ignored, substantial for plateaued lifters.

For a novice on a linear-progression program, load is the primary progression and the other three should largely look after themselves. For an intermediate lifter, load increments of 2.5–5 kg on compound lifts are implausible week-over-week and you need to lean on volume progression and occasional density work. The Progressive Overload Planner projects the four-dimensional trajectory so you don't try to force load progression when volume is the more honest lever.

Periodisation: less magic than marketed

Moesgaard and colleagues' 2022 meta-analysis of 35 volume-equated studies^[4] compared non-periodised, linear and undulating programs. Periodised programs produced more 1RM strength than non-periodised ones (ES 0.31) but not more muscle (ES 0.13, not significant). Undulating periodisation beat linear for strength (ES 0.31), and in the subgroup analysis only in trained lifters (ES 0.61).

For intermediate lifters, the useful takeaways are:

• Structured variation in load and volume outperforms doing the same workout every week indefinitely.
• For hypertrophy alone, periodisation scheme barely matters; for trained lifters chasing a total, daily or weekly undulation has the edge.
• Planned deloads every 4–8 weeks are a form of periodisation and a plausible way to sustain higher average volume across a mesocycle.

Advanced techniques: useful within limits

Drop sets, rest-pause, myo-reps, and similar intensification techniques were reviewed by Krzysztofik et al. in 2019^[5]. The net finding is that advanced techniques can produce equivalent hypertrophic stimulus with less time-under-bar than straight sets, but don't systematically beat straight sets at equated volume.

The honest positioning for advanced techniques: use them when time is the constraint (finishing a session with a drop set instead of two more straight sets), not because they are a hypertrophy multiplier.

When to deploy each technique

Technique        Best-use context                              Cost
────────────────────────────────────────────────────────────────────────
Drop sets        Time-constrained accessories, final set       High fatigue
Rest-pause       Last working set when you want extra reps     Moderate
Myo-reps         Small muscles (delts, arms), volume-efficient Low-moderate
Cluster sets     Heavy strength work, preserving quality       Low
Pre-exhaust      Isolation → compound to fatigue weak point    Moderate

In a time-limited 45-minute session, the calculus shifts — a single drop set at the end of a working-set group can deliver volume you wouldn't otherwise fit. In a 75-minute session with no time pressure, more straight sets almost always beat the intensification.

Exercise selection and specificity

Specificity is the principle that training adaptations mirror the stimulus applied. For strength goals, the exercise you want to get stronger at must be trained directly or heavily. For hypertrophy, the exercise matters less than the stimulus: what counts is hard sets that load the target muscle through its range.

Three exercise-selection heuristics that hold up:

• Stability matters less than load tolerance. Free-weight purism isn't supported by the evidence. Machines, cables, and dumbbells all produce hypertrophy when loaded at RPE 7+ for adequate volume.
• Stretched-position emphasis is a trend with some support. Lengthened partials and deep-range work (training the bottom half of a leg curl, or squatting deep) appear to produce slightly more hypertrophy per set than standard range in recent small studies, but the effect is modest and not yet pooled.
• Variety within a block matters less than consistency. Rotating exercises every session tends to harm progress measurement. Stick with a core set of 6–10 movements across a 6–12 week block, then rotate.

Frequency-per-muscle tuning across sessions

A practical problem: if your weekly target for chest is 16 hard sets across 2 sessions, that's 8 sets per session. If you also want 12 sets for shoulders and 10 for triceps, a push session balloons past 30 working sets and becomes unrecoverable. Two frequency-tuning strategies:

1. Synergist volume accounting. Close-grip bench press counts as ~0.5 triceps sets even though it's primarily a chest movement; counting indirect sets as half a set is the method that best fitted Pelland 2026's data.^[10] Running separate columns for chest and triceps without discounting shared movements overstates true volume.
2. Asymmetric session split. Rather than hitting every muscle identically twice per week, split into a “heavy” session and a “volume” session per muscle. Heavy day: 3 working sets at RPE 8. Volume day: 6 working sets at RPE 7. Total 9 sets, weighted sensibly across the week.

The novice-intermediate-advanced transition

Classifying your training stage determines how much of this programming sophistication matters:

Stage         Primary progression    Programming complexity    Deload interval
────────────────────────────────────────────────────────────────────────────
Novice        Session-to-session     Minimal                    Rarely needed
Early inter.  Week-over-week         Moderate                   6–8 weeks
Late inter.   Mesocycle-over-         Periodisation enters      4–6 weeks
              mesocycle
Advanced      Annual plan            Sophisticated periodisation 3–5 weeks

Most recreational lifters who “feel stuck” are intermediates who've inherited advanced-lifter programming norms they don't yet need. A well-run intermediate block with consistent execution outperforms a sophisticated advanced program the lifter can't actually complete.

Concurrent aerobic training

Running alongside a lifting block costs less than the older literature suggested. Wilson et al.'s 2012 meta-analysis found smaller gains in size, strength and power for concurrent groups, with the penalty growing with the frequency and duration of the endurance work.^[9] The 2022 Schumann update of 43 studies found no interference with hypertrophy or maximal strength, only with explosive strength, and most when both were trained in the same session.^[12]

For hybrid athletes, a defensible stack is: morning strength, evening aerobic (or vice versa), with hard intervals kept away from heavy leg days. The full breakdown, including a 2024 finding that men lost some lower-body strength gain and women did not, is in concurrent training interference.

Tracking the four overload dimensions

A practical scoreboard to check each week:

Main lift    Load   Reps   Sets   Technique (RPE at which you hit it)
─────────────────────────────────────────────────────────────────────
Squat        +2.5   same   same   same        (load progression)
Bench        same   +1     same   tighter     (rep + technique prog.)
Deadlift     same   same   +1     same        (volume progression)
OHP          same   same   same   cleaner     (technique only)

If after a week the column is mostly blank, you didn't progress on that lift. Mostly one-check weeks are normal for intermediates. Mostly three-check weeks probably mean the prescribed loads are too light and you should raise them.

Measuring adaptation across a block

Three measurements that actually correlate with block effectiveness:

• Top-set load at a fixed RPE. Your RPE 8 top set of 5 on squat this week vs the same metric six weeks ago. If it's 5–10 kg higher at equivalent RPE, the block worked.
• Total working-set tonnage on key lifts. Sum across working sets (load × reps × sets). Increasing weekly tonnage at similar RPE is a strong adaptation signal.
• Bodyweight and body-composition trend. For hypertrophy blocks specifically, a slow, steady bodyweight rise alongside stable or improving waist measurements suggests the stimulus is being turned into tissue.

Tracking all three gives you triangulation. If load is rising but bodyweight is falling on a hypertrophy block, you're probably gaining strength without adding tissue — a sign of under-eating or insufficient volume for genuine hypertrophy.

Common programming mistakes

• Stacking too many specialization blocks back-to-back. Extra volume for one muscle has to come from recovery somewhere; run a specialization block, then bring that muscle back to maintenance volume.
• Never testing intensity. Hypertrophy is load-tolerant, but six months at 60–70% 1RM with no heavy exposure eventually costs you strength specificity.
• Changing too many variables simultaneously. New split, new exercises, new volume, new rep scheme — all at once means you can't diagnose which change helped or hurt.
• Never deloading. Accumulated fatigue silently cuts session quality without obvious single-day crashes.
• Chasing “optimal” over adherence. The theoretically ideal program you can't consistently execute is worse than the pragmatic program you actually follow.

The cost of advanced programming

A quick reality check on ambition. An advanced block-periodised program with microcycles, mesocycles, planned deloads, and day-to-day autoregulation works well — for lifters who can actually execute it. For most intermediate lifters, the program complexity exceeds their training consistency, and complexity becomes a failure mode rather than an edge.

A simple, well-run intermediate program (e.g. 4-day upper/lower with progressive overload planning and a planned deload every 5–6 weeks) is likely to beat a more elaborate program the lifter only half completes: the volume and effort that the meta-analyses reward are the sets actually done. Match complexity to what you'll actually complete for 6+ months straight.

A minimal 2026 programming checklist

• Weekly hard-set volume per muscle is inside the 10–20-set range (or you've tested outside it and have data on the outcome).
• Weekly sets are spread across sessions (twice a week per muscle is the usual way). For size the number of sessions matters little once the sets are done; for strength, more frequent practice helps.^[10][11]
• At least one load exposure per week is heavy (top set at RPE 8+ or 80%+ 1RM) for each of squat, hinge, push, pull.
• At least one overload dimension progresses weekly on the lifts you're tracking: load, reps, sets, or honest technique improvement.
• A deload or reduced-volume week is scheduled every 4–8 weeks. If you skipped the last two, you need one now.
• You're tracking, even if the tracking is rough. Memory alone underestimates volume drift.

Start from this skeleton and layer specificity as your goal demands. A powerlifter meet-prepping needs more competition-specific work, a physique competitor needs more isolation volume on lagging parts, a hybrid athlete needs fewer total hard sets to accommodate concurrent aerobic work. But in all three cases, the underlying framework is the same: volume inside the dose-response band, intensity close enough to failure, frequency enough to accumulate the volume, overload progressing on one of four dimensions.

Tools that operationalise this article: Workout Volume Calculator, 1RM Calculator, Progressive Overload Planner, and the RPE to Percentage Converter.

Population boundaries of the dose-response literature

The volume / intensity / frequency findings in this article trace primarily to the Schoenfeld meta-analyses^[1][3][11], the Pelland 2026 meta-regressions (79% male, mean age 25)^[10] and the Moesgaard periodisation meta-analysis^[4]. The sample characteristics matter:

• Male-majority samples. Most included RCTs were >70% male. Female-specific dose-response curves are inferred from male data. Recent research suggests women may tolerate and benefit from the same per-muscle set ranges, but with hormonal-phase-specific response variation that the male-dominated literature doesn't capture.
• Age 18–40 dominant. Sarcopenia-era (60+) dose-response is different; masters lifters often benefit from higher protein and similar set volume but need longer inter-session recovery.
• Drug-free framing usually assumed. The volume-plateau findings reflect natural training populations. Enhanced lifters tolerate substantially higher volumes without the plateau.
• Training-status range. Most studies recruit recreationally trained lifters (6 months to 2 years). Extrapolating the dose-response curve to elite or absolute-novice populations is inference, not measurement.
• Session duration not controlled. Studies prescribing "20 sets per muscle per week" rarely standardise session length. Actual adaptation may depend on distribution within the week as much as on the weekly aggregate.

Alternative views and unresolved debates

Two areas where the programming literature has not converged:

• Stretched-position training. Recent small trials suggest lengthened-partial and deep-range-of-motion exercises may produce somewhat more hypertrophy per set than standard-ROM work. The studies are short and few, and the size of the advantage varies by muscle, so treat it as a tiebreaker in exercise selection rather than a rule.
• Junk volume vs effective-reps frameworks. The "stimulating reps" model (only the last few reps before failure are productive) and the "a hard set is a hard set" framework disagree on how to count low-load, high-rep work. In practice both agree that a set stopped far from failure adds little, so the useful rule is the same: count sets that end within about three reps of failure.