More sets still win. Lengthened partials, tested at scale, mostly tie

Most of what gets published about building muscle restates settled ground. Two threads genuinely moved between late 2025 and mid-2026: the shape of the volume dose-response curve, and the fate of the training method that dominated lifting discourse for three years.

Volume: still upward, and not the same curve as strength

A meta-regression published in Sports Medicine in December 2025 pooled 67 studies covering 2,058 participants (79.1% male, 20.9% female; mean age 25.16 ± 5.22 years) to model weekly set volume and training frequency as continuous variables rather than categories [s1].

Its methodological move was to classify every contributing set as direct or indirect relative to the muscle being measured, then weight indirect sets at 1 ("total"), 0.5 ("fractional") or 0 ("direct") [s1]. The fractional method fit the data best and was used for the primary models [s1]. That sounds like bookkeeping; it is the reason the curve changed. Counting a bench press as a full set for triceps growth, or not counting it at all, produces different dose-response relationships from the same underlying trials.

Under that model, the posterior probability that the marginal slope for volume exceeded zero was 100% for both hypertrophy and strength — muscle size and strength both increase as weekly sets increase [s1]. Both best-fit models showed diminishing returns, and the diminishing returns for strength were considerably more pronounced than for hypertrophy [s1].

Frequency behaved differently. The posterior probability for frequency's effect on hypertrophy was below 100%, which the authors describe as compatible with negligible effects; for strength it was 100%, again with diminishing returns [s1]. The practical reading is that spreading the same sets across more sessions appears to matter for strength and not obviously for size — a divergence that categorical meta-analyses had blurred.

This sits alongside the 2024 meta-regression on proximity to failure, which found that strength gains were similar across a wide range of estimated repetitions in reserve while hypertrophy increased as sets were terminated closer to failure [s6]. That analysis estimated repetitions in reserve from study descriptions rather than measuring it, reported only modest overall model fit, and its authors explicitly cautioned that it was exploratory [s6]. Two independent meta-regressions now point the same way: the variables that drive size and the variables that drive strength are not the same variables.

Lengthened partials: the large trial arrives

Training only the stretched portion of a repetition — "lengthened partials" — became one of the most confidently promoted ideas in strength training on the back of small, short studies. In October 2025 it got a properly powered test.

A pre-registered randomised controlled cluster trial ran across 15 sites, assigning 163 participants to lengthened-partial range of motion and 134 to full range of motion for 12 weeks, with estimated muscle cross-sectional area of the upper arm and thigh as outcomes alongside chest press, leg press and pulldown strength [s2]. The estimate for the condition-by-time interaction was −0.032 for arm cross-sectional area and 0 for thigh [s2].

The statistical detail matters. Equivalence tests against a smallest effect size of interest of [−0.1, 0.1] returned p = 0.071 for arm and p = 0.019 for thigh, against an alpha set at 0.01 and adjusted to 0.005 for multiple outcomes — so the authors could not formally reject the null that the interaction lay outside the equivalence bounds [s2]. What they report instead is that both main effects of time and any interaction effects are likely small, and that between-condition effects are small and practically equivalent [s2]. This is a trial that did not confirm a difference and also did not formally certify sameness. It is still by a wide margin the largest comparison of the two methods.

Smaller trials cluster around the same point without agreeing on the details. A within-participant study of 30 resistance-trained people, training upper body twice weekly for eight weeks, found similar muscle thickness and lat pulldown 10-repetition-maximum improvements between conditions, with Bayes factors of 0.16 to 0.3 providing moderate support for the null hypothesis of equal improvement [s3]. A within-subject study of 13 trained individuals doing unilateral preacher curls for eight weeks found similar thickness gains at 50% of humeral length but trivial-to-small superiority for the partial condition at 70% (SMD 0.10, Bayes factor 4.87), while full range showed negligible advantages in maximal voluntary contraction at 100° (SMD 0.24, Bayes factor 3.02) and one-repetition maximum (SMD 0.17, Bayes factor 1.95) [s4]. Thirteen participants is a pilot, and the authors describe the supporting evidence as weak to moderate [s4].

A four-group trial published in July 2026 added a load dimension. Forty-five participants completed eight weeks of knee extensor training as shortened partials at 80% of one-repetition maximum, lengthened partials at 55%, full range at 80%, or control [s5]. Lengthened partials and full range both produced greater vastus lateralis thickness change than shortened partials at 75% of femur length, with no differences between the lengthened-partial and full-range groups at any measurement site [s5]. Fascicle length change was greater in the lengthened-partial group than full range at 25% and 75% of femur length [s5]. The authors frame this as moderate-intensity partial-range work matching high-intensity full-range work for hypertrophy while producing superior fascicle length adaptation [s5].

Where that leaves the question

The direction of travel across 2025 and 2026 is toward smaller claimed differences, tested in larger samples. Volume remains the variable with the clearest continuous relationship to muscle growth, now with a better-specified curve and a clear separation from how strength responds [s1]. Range-of-motion manipulation, which was being discussed as a decisive lever, produced point estimates near zero in the only multi-site trial to test it [s2], moderate evidence for the null in a within-participant trial [s3], and region-specific differences too small for the studies detecting them to characterise confidently [s4][s5].

What is worth watching is whether the fractional set-counting method in the dose-response analysis gets adopted by other groups. If it does, a large share of the existing volume literature will need re-tabulating before it can be compared with anything published after it.

Sources

1. [s1] Pelland JC, Remmert JF, Robinson ZP, Hinson SR, Zourdos MC. The Resistance Training Dose Response: Meta-Regressions Exploring the Effects of Weekly Volume and Frequency on Muscle Hypertrophy and Strength Gains. Sports Medicine, published online 4 December 2025. https://doi.org/10.1007/s40279-025-02344-w
2. [s2] Gschneidner D, Carlson L, Steele J, Fisher JP. The effects of lengthened-partial range of motion resistance training of the limbs on arm and thigh muscle area: A multi-site randomised trial. Journal of Sports Sciences, published online 7 October 2025. https://doi.org/10.1080/02640414.2025.2567805
3. [s3] Wolf M, Androulakis Korakakis P, Piñero A, et al. Lengthened partial repetitions elicit similar muscular adaptations as full range of motion repetitions during resistance training in trained individuals. PeerJ, 12 February 2025. https://pmc.ncbi.nlm.nih.gov/articles/PMC11829627/
4. [s4] Havers T, Wagner N, Held S, Geisler S, Wiewelhove T. Partial Range, Full Gains? European Journal of Sport Science, December 2025. https://pmc.ncbi.nlm.nih.gov/articles/PMC12621570/
5. [s5] McMahon G, Morse C, Burden A, Winwood K, Onambele-Pearson G. Moderate Intensity Resistance Training With Partial Range-of-Motion at Long Muscle Lengths Elicits Similar Hypertrophy and Architectural Adaptations as High Intensity Resistance Training Using Full Range-of-Motion. Journal of Strength and Conditioning Research, published online 2 July 2026. https://doi.org/10.1519/jsc.0000000000005561
6. [s6] Robinson ZP, Pelland JC, Remmert JF, et al. Exploring the Dose-Response Relationship Between Estimated Resistance Training Proximity to Failure, Strength Gain, and Muscle Hypertrophy: A Series of Meta-Regressions. Sports Medicine, 6 July 2024. https://pubmed.ncbi.nlm.nih.gov/38970765/