7 claims that cleared review. Each one shows what has to be true for it to hold and how strongly it is supported — including the places the sources contradict each other.
Muscle growth increases with weekly hard-set volume per muscle, from a minimum effective dose of roughly 4 sets up through about 10-20 or more sets per week, with steadily diminishing returns as volume rises.
More weekly sets build more muscle, up to a point. Start with about four hard sets per muscle each week and add more over time, but each extra set gives a bit less back.
Only when
Volume counted as fractional weekly sets per muscle, with an indirect set weighted about 0.5 of a direct set
Sets must be hard (taken close to muscular failure) to count toward the effective volume
Diminishing returns accelerate as volume rises and evidence becomes sparse beyond about 25 fractional weekly sets, so high volumes carry more uncertainty
At equal weekly volume, training a muscle more frequently does not independently add growth (see the frequency principle)
Evidence is predominantly young adults (mean age ~25, mixed trained and untrained) over short (~10-week) interventions, with wide credible intervals
Who
General resistance trainees seeking muscle growth · Not Adults over ~70; not established beyond ~10-week interventions or very high weekly volumes
Moderate confidence
Two independent meta-analyses (Pelland, Refalo) plus one training educator (Huberman) agree weekly volume drives hypertrophy; capped at moderate because the dose-response shape and upper range are uncertain and the evidence is largely young adults over short interventions.
3 independent people · 3 sources · strongest evidence: meta analysis
Sources
The Resistance Training Dose Response: Meta-Regressions Exploring the Effects of Weekly Volume and Frequency on Muscle Hypertrophy and Strength Gains · Sports Medicine (Springer Nature) / SportRxiv preprint · 2024 · doi:10.51224/SRXIV.460
Influence of Resistance Training Proximity-to-Failure on Skeletal Muscle Hypertrophy: A Systematic Review with Meta-analysis · Sports Medicine (vol. 53, issue 3, pp. 649-665) · 2023 · doi:10.1007/s40279-022-01784-y
Free weights and machines produce no detectable difference in strength, jump performance or muscle hypertrophy when compared directly (13 studies, 1016 adults aged 18 to 60; hypertrophy standardised mean difference -0.055, 95% CI -0.397 to 0.287), while strength gains show up most in the modality -- and, on exploratory evidence, the range of motion -- that was actually trained and tested, so equipment and exercise choice is a matter of preference, goal and what will be tested rather than a lever on growth.
Machines and free weights build the same strength and size, so pick what you enjoy and can keep progressing. One catch: strength shows up best when you test it the way you trained it.
Only when
The four direct between-modality comparisons are all null with intervals crossing zero: dynamic strength 0.084 (-0.106 to 0.273), isometric strength -0.079 (-0.432 to 0.273), countermovement jump -0.209 (-0.597 to 0.179), hypertrophy -0.055 (-0.397 to 0.287)
A null is an absence of a detected difference, not proof the two modalities are identical; hypertrophy was reported by only 5 of the 13 pooled studies
The specificity result is asymmetric: free-weight-tested strength favoured free-weight training significantly (SMD -0.210, -0.391 to -0.029, p = 0.023), while machine-tested strength only tended to favour machine training (0.291, -0.017 to 0.600, p = 0.064)
The range-of-motion half of the specificity idea comes from analyses the meta-analysis's own authors label exploratory, and carries no effect size or interval anywhere
The pooled sample is roughly four-fifths male (789 men, 219 women) and bounded at ages 18 to 60; no result is reported separately by sex
The 2026 ACSM position stand reports insufficient data to determine whether hypertrophy is influenced by machine versus free-weight training, single- versus multi-joint training, varied exercise selection or range of motion -- a statement about the evidence, not a finding of no effect
None of this addresses safety, supervision, technique demand, injury history or equipment availability, which are the reasons a coach usually picks one over the other
Who
Adults aged 18 to 60 doing at least six weeks of strength training for strength, size or jump performance; pooled sample 1016 participants, 789 men and 219 women. · Not Anyone under 18 or over 60 -- the review pooled nobody outside that range. Nothing is established for athletes with a sport-specific movement requirement beyond the specificity result itself, and no source addresses beginners' technique needs.
Moderate confidence
Two voices. haugen-markus is independent (shares_voice_with lists only varvik-fredrik, a co-author) and carries the whole quantitative case: a PRISMA meta-analysis of 13 studies and 1016 participants in which every direct comparison is null. The second voice is the Wolf/Schoenfeld component (currier-bryan's ACSM position stand, and wolf-milo's exploratory range-of-motion specificity signal), which agrees but adds no independent quantity -- the position stand's contribution is a statement that the data are insufficient. Confidence is moderate rather than high because the pooled sample is four-fifths male and age-bounded, hypertrophy rests on 5 studies, and the specificity claim on the range-of-motion side is exploratory with no effect size. No conflict of interest is declared in any abstract held.
2 independent people · 3 sources · strongest evidence: meta analysis
Where sources disagree
The 2009 ACSM position stand prescribes exercise composition directly: a strength programme should include concentric, eccentric and isometric actions and both bilateral and unilateral and both single-joint and multi-joint exercises, and hypertrophy training should use similar exercise selection to strength training. These are a writing committee's consensus recommendations with no effect estimate attached, and the College replaced that stand in April 2026 with one that reports insufficient data for the very comparisons at issue. They are recorded here rather than counted as support, and must never be presented as the current ACSM recommendation. — acsm--progression-models-in-resistance-training-for-healthy-adults--2009 seg-003:c2 and seg-006:c1 -- the superseded position stand
Sources
Effect of free-weight vs. machine-based strength training on maximal strength, hypertrophy and jump performance - a systematic review and meta-analysis · BMC Sports Science, Medicine and Rehabilitation · 2023 · doi:10.1186/s13102-023-00713-4
American College of Sports Medicine Position Stand. Resistance Training Prescription for Muscle Function, Hypertrophy, and Physical Performance in Healthy Adults: An Overview of Reviews · Medicine and Science in Sports and Exercise · 2026 · doi:10.1249/MSS.0000000000003897
Partial Vs Full Range of Motion Resistance Training: A Systematic Review and Meta-Analysis · International Journal of Strength and Conditioning · 2023 · doi:10.47206/ijsc.v3i1.182
Taking every set to momentary muscular failure is not required to build muscle; training close to failure is what matters, so most sets should stop a few reps short and true failure be reserved for select sets to limit fatigue.
You don't have to push every set to total failure to build muscle. Training close to failure works just as well, and stopping a little short lets you do more quality work.
Only when
Applies to muscle hypertrophy, not to maximal strength
The small pooled hypertrophy advantage for failure is statistically fragile: it assumes a pre-post correlation of 0.75 and becomes non-significant once that correlation falls below about 0.73, and rests on low-powered sub-analyses and indirect velocity-loss proxies
Reserve momentary failure for single-joint or machine exercises, the last set of a muscle group, or muscles trained with low weekly volume or frequency
A fixed velocity-loss threshold is an unreliable individual proxy for true proximity-to-failure (e.g. at 40% velocity loss only ~56% reached failure)
Who
Healthy adults (trained and untrained, ~20-65) training for muscle growth · Not Hypertrophy only - this does not address maximal-strength training, where near-maximal practice still matters
Moderate confidence
One meta-analysis (Refalo) plus one training educator (Huberman); moderate. The pooled advantage for training to failure is trivial and statistically fragile, and the proximity evidence is low-powered and based on indirect proxies.
2 independent people · 2 sources · strongest evidence: meta analysis
Sources
Influence of Resistance Training Proximity-to-Failure on Skeletal Muscle Hypertrophy: A Systematic Review with Meta-analysis · Sports Medicine (vol. 53, issue 3, pp. 649-665) · 2023 · doi:10.1007/s40279-022-01784-y
Muscle can be built across a wide range of loads, from roughly 30% of one-rep max upward rather than only in a moderate-load zone, as long as sets are taken close to muscular failure.
You don't need heavy weights to build muscle. Loads from light-to-heavy all work, as long as you take each set close to failure. Very light weights are the exception.
Only when
Loads at or above roughly 30% of one-rep max; below about 30% 1RM hypertrophy is compromised
Low-load equivalence depends on sets being taken to or very near muscular failure
Moderate loads are often the pragmatic default: very light loads need far more reps, time, and discomfort, while very heavy loads add more joint and tendon stress and neuromuscular fatigue
Who
Adults training for muscle hypertrophy · Not Does not apply to maximal (1RM) strength, which favors heavier, near-maximal loads
Moderate confidence
A narrative review citing meta-analytic and RCT evidence (Schoenfeld) plus one training educator (Huberman); paper-plus-transcript, two people, moderate. May supersede the earlier single-person proposed principle hypertrophy--afc7.
2 independent people · 2 sources · strongest evidence: meta analysis
Sources
Loading Recommendations for Muscle Strength, Hypertrophy, and Local Endurance: A Re-Examination of the Repetition Continuum · Sports (Basel, Switzerland) · 2021 · doi:10.3390/sports9020032
Maximal strength in a lift rises with the weekly number of sets of that lift but plateaus quickly, with most of the gain reached by about 4 hard sets per week and little further benefit beyond roughly 5.
Getting stronger at a lift takes surprisingly few weekly sets. Most of the gain comes from about four hard sets a week; doing much more adds little extra strength.
Only when
Volume counted as fractional weekly sets of the specific assessed lift
Most benefit is reached by about 4 weekly sets; additional sets beyond about 5 do not consistently add strength
The very low minimum effective dose (~1 set) may partly reflect a learning effect from performing a novel strength test
May not extend to well-trained lifters, who were reported to need more direct weekly sets
Evidence is predominantly young adults (mean age ~25)
Who
General trainees seeking maximal strength in a lift · Not Adults over ~70; may not extend to well-trained lifters, who may need more direct weekly sets
Moderate confidence
One strong Bayesian meta-regression (single person, Pelland); moderate is the ceiling for a single meta-analysis, and a noted possible learning effect may inflate the very low minimum effective dose.
1 independent person · 1 source · strongest evidence: meta analysis
Sources
The Resistance Training Dose Response: Meta-Regressions Exploring the Effects of Weekly Volume and Frequency on Muscle Hypertrophy and Strength Gains · Sports Medicine (Springer Nature) / SportRxiv preprint · 2024 · doi:10.51224/SRXIV.460
Maximal one-rep-max strength is best developed with heavier, near-maximal loads, because gains in tested 1RM strength largely reflect the specificity of practicing near-maximal lifts.
If your goal is lifting the heaviest possible single rep, practice with heavy, near-maximal weights. Light weights build muscle well but transfer less to a maximal lift.
Only when
Outcome is maximal dynamic (1RM) strength; the heavy-load advantage is much smaller and statistically non-significant when strength is tested isometrically
The advantage largely reflects the specificity of practicing near-maximal lifts rather than a purely intrinsic strength adaptation
Who
Adults training for maximal dynamic (1RM) strength
Moderate confidence
Single review (Schoenfeld) citing a 14-study meta-analysis; moderate. The heavy-load edge is specific to 1RM testing and largely reflects practice specificity.
1 independent person · 1 source · strongest evidence: meta analysis
Sources
Loading Recommendations for Muscle Strength, Hypertrophy, and Local Endurance: A Re-Examination of the Repetition Continuum · Sports (Basel, Switzerland) · 2021 · doi:10.3390/sports9020032
At an equal weekly set volume, how many sessions the volume is split across has little independent effect on muscle growth, but spreading a lift's work across about two sessions per week modestly improves strength in that lift, with diminishing returns beyond that.
Training a muscle more often doesn't grow it faster if your weekly sets stay the same. But splitting a lift over two sessions a week can help you get stronger at it.
Only when
Applies only when total weekly set volume is held equal across the frequency comparison
The strength benefit refers to the specific assessed lift and is attributed partly to more frequent motor-pattern practice, not simply more sets
Diminishing returns beyond about 2 sessions per week; the frequency inference is limited by the low average weekly volume of the included studies
Evidence is predominantly young adults (mean age ~25)
Who
General resistance trainees · Not Adults over ~70
Moderate confidence
Single meta-regression (one person, Pelland); moderate. The volume-equated frequency-for-hypertrophy null is limited by study designs and the strength-frequency benefit by the low average volume of the included studies.
1 independent person · 1 source · strongest evidence: meta analysis
Sources
The Resistance Training Dose Response: Meta-Regressions Exploring the Effects of Weekly Volume and Frequency on Muscle Hypertrophy and Strength Gains · Sports Medicine (Springer Nature) / SportRxiv preprint · 2024 · doi:10.51224/SRXIV.460
This is a summary of published research, not medical advice. It cannot account for your medications, conditions or history. Talk to a clinician before acting on any of it.