Training to momentary failure is not superior to stopping short for hypertrophy
Only a trivial advantage for "set failure" under any definition, and no advantage for higher velocity-loss thresholds
Protocol
15 studies measuring muscle hypertrophy that compared (A) momentary muscular failure vs non-failure, (B) "set failure" (any definition other than momentary muscular failure) vs non-failure, or (C) different velocity-loss thresholds.
Key Findings
Set failure (any definition) vs non-failure: a trivial advantage for hypertrophy — effect size 0.19 (95% CI 0.00, 0.37), p = 0.045 — with no moderating effect of volume load (p = 0.884) or relative load (p = 0.525).
Momentary muscular failure vs non-failure: no advantage — effect size 0.12 (95% CI −0.13, 0.37), p = 0.343.
High (>25%) vs moderate (20–25%) velocity loss: no advantage — effect size 0.08 (95% CI −0.16, 0.32), p = 0.529.
The authors' conclusion: no evidence that momentary failure beats non-failure for hypertrophy; closer proximity to failure does not always give more growth — a potentially non-linear relationship.
Practical Application
Across 15 trials, sets taken to momentary muscular failure grew muscle no more than sets stopped short (ES 0.12, p = 0.343), and pushing velocity loss beyond 25% added nothing over 20–25% (ES 0.08). Hard, sub-failure sets deliver the hypertrophy stimulus; grinding every set to failure is not required.
Not from this paper; the base's own practice: run most hypertrophy sets at roughly 0–3 reps in reserve, and reserve true failure for occasional single-joint or final sets. The fatigue cost of failure training was not measured in this meta-analysis — see REF-640 for a trial where not-to-failure training won on performance.
Limitations
Failure definitions and RIR accuracy vary between studies; trained-lifter and long-term data still developing.
The pooled advantage for "set failure" is trivial and its confidence interval touches zero; the abstract reports no fatigue, recovery or performance outcomes.
How this entry was checked and changed
Why this level
regraded 2026-09-27 from B+: meta-analysis of 15 mostly small trials with heterogeneous failure definitions
Content correction 2026-09-27
"closer to failure showed a small trend toward greater hypertrophy" -> the abstract reports a trivial advantage for set failure under any definition (ES 0.19, p = 0.045) and no advantage for momentary failure (ES 0.12, p = 0.343) or for higher velocity-loss thresholds (ES 0.08); title and findings rewritten with these numbers.
"near-equivalent growth with markedly less fatigue" -> fatigue was not an outcome in this meta-analysis; claim removed.
"~0–3 reps in reserve" relabelled as the base's own practice (not in the abstract). Level B+ -> A- (meta-analysis of 15 small trials).