"Go deep or go home" and "partials let you overload the lift" are both gym-floor gospel, and they can't both be unconditionally true. The research resolves the disagreement, but not by picking a side — it resolves it by pointing out that "partial rep" describes two mechanically opposite things that produce opposite results.
Lumped together, partial reps carry a small average disadvantage against full range of motion. Split apart, one type of partial matches full ROM almost exactly, and the other type is worse on every measure that matters, including joint comfort.
What's the difference between full range of motion and partial reps?
Full range of motion means moving a joint through its complete natural arc on a lift, from the fully stretched starting position to the fully contracted end position. A partial rep intentionally stops the movement somewhere inside that arc, and where it stops changes everything about the result.
There are two structurally different kinds of partial rep. A lengthened partial trains the bottom, stretched portion of a lift — the bottom half of a squat, the stretched portion of a leg extension or biceps curl. A shortened partial cuts range from the stretched end and trains only the top, more contracted portion — a quarter squat, a lockout-only bench press, a top-half curl.
Treating these as one category is the single biggest source of confusion in the "full ROM vs partials" debate, because the research below shows they behave in opposite directions.
Does full range of motion build more muscle than partial reps?
On average, yes, but only by a small margin. Wolf, Androulakis-Korakakis, Fisher, Schoenfeld & Steele (2023) ran a Bayesian multi-level meta-analysis across strength, hypertrophy, power, speed, and body-composition outcomes, and found a trivial overall SMD of 0.12 (95% CI −0.02 to 0.26) favouring full ROM. Every individual outcome sub-group favoured full ROM too, but each was trivial to small (SMD 0.05–0.2).
That average, though, blends together every kind of partial-ROM study the authors could find — lengthened, shortened, and everything between. A small overall edge for full ROM doesn't tell you whether a specific kind of partial rep is competitive or not; it tells you that partial reps, averaged across every protocol ever tested, land slightly behind full ROM.
The same meta-analysis's sub-group data point at why: when the authors isolated studies that trained partial reps specifically at long muscle lengths — the stretched position — the estimate flipped toward favouring partials for hypertrophy (SMD −0.28, 95% CI −0.81 to 0.16). That confidence interval is wide and crosses zero, so it's suggestive rather than conclusive on its own, but it lines up with a purpose-built trial published two years later.
Do lengthened partial reps work as well as full range of motion?
Yes — a direct, purpose-built trial found no meaningful difference. Wolf, Androulakis Korakakis, Piñero, Mohan, Hermann, Augustin, Sapuppo, Lin, Coleman, Burke, Nippard, Swinton & Schoenfeld (2025) randomised 30 resistance-trained adults' upper-body limbs so that one arm trained lengthened partials and the other trained full ROM, for eight weeks, two sessions per week, four exercises and four sets per session.
Because each person served as their own control, the design removes individual variation as a confounder — any gap between arms is attributable to the ROM condition itself, not to who happened to end up in which group. Muscle thickness of the elbow flexors and extensors (measured by ultrasound) and lat pulldown 10-rep-max improvements came out similar between conditions, with Bayes factors of 0.16–0.3 giving "moderate" statistical support for no meaningful difference.
"Trainees seeking to maximize muscle size should likely emphasize the stretched position, either by using a full ROM or LPs [lengthened partials] during upper-body resistance training." — Wolf et al. (2025)
That's a specific, testable claim, not a general endorsement of cutting range. The trial tested lengthened partials — the stretched bottom portion — not shortened, lockout-style partials, which the next study shows is an important distinction.
Does squat depth change strength gains and injury risk?
Yes, and the direction runs the opposite way to the lengthened-partials finding above. Pallarés, Sánchez-Medina, Pérez, De La Cruz-Sánchez & Mora-Rodríguez (2020) put 53 resistance-trained men through 10 weeks of velocity-based training at 60–80% 1RM, split into full squat (F-SQ), parallel squat (P-SQ), half squat (H-SQ), and control groups — with squat depth the only variable that differed between groups.
F-SQ was the only group to increase 1RM and bar speed across all three tested depths (effect sizes 0.77–2.36) and produced the best functional-performance gains on countermovement jump, sprint, and Wingate tests (effect sizes 0.35–0.85). P-SQ placed second (effect sizes 0.15–0.56). H-SQ produced no significant neuromuscular or functional gains at all (effect sizes −0.11 to 0.28) — and was the only group with a significant increase in pain, stiffness, and functional disability (effect sizes 0.87–1.21).
Full ROM's broader transfer here matches a specificity pattern the 2023 meta-analysis also flagged: training in a given range tends to produce the biggest strength gain when tested in that same range, so a shallow half squat mostly builds a stronger half squat. What Pallarés adds is the cost side — shortening range at the stretched end of the squat didn't just fail to transfer, it came with more reported pain.
What lifters report in practice
Lifters who train partial reps commonly report they're useful for a specific, narrow purpose rather than as a wholesale replacement for full ROM. Lengthened partials at the bottom of a squat, bench, or curl get used to add extra stimulus to a muscle that full-ROM sets alone aren't growing, or to keep training a joint through a smaller range while working around a minor injury.
The criticism that comes up just as often is aimed at shortened, top-end partials specifically. Lifters who lean on quarter squats or lockout-only presses to chase a heavier number report that the number doesn't transfer — a big quarter-squat "PR" doesn't predict a bigger full-depth squat, and self-coached lifters who discover this the hard way tend to describe it as a confidence problem as much as a training one.
A second, quieter criticism is that short partials can become the path of least resistance. Skipping the stretched portion of a lift avoids the hardest, most uncomfortable part of the rep, which happens to be the part the 2025 trial and the meta-analysis's sub-group both point to as doing the most hypertrophy work — so leaning on shortened partials by default tends to under-train exactly the range that matters most.
How do full ROM, lengthened partials, and short partials compare at a glance?
| Attribute | Full ROM | Lengthened partials (stretched end) | Shortened partials (contracted end) |
|---|---|---|---|
| Hypertrophy | Baseline; slight average edge across all partial types (Wolf et al., 2023) | Matches full ROM (Wolf et al., 2025) | Not directly tested, but excluded from the sub-group that matched full ROM |
| Strength transfer | Broadest — transferred across every depth tested (Pallarés et al., 2020) | Similar 10RM strength-endurance gains to full ROM (Wolf et al., 2025) | Weakest transfer to full-range strength; specific to the trained range only |
| Joint stress / pain | Not the highest-pain condition in the squat trial | Not tested for pain in the 2025 trial | Only condition with a significant pain and stiffness increase (Pallarés et al., 2020) |
| Typical use case | Default for most compound lifts | Deliberate add-on for a lagging muscle at its stretched position | Narrow, deliberate use only (e.g. lockout weakness) — not a primary method |
How should self-coached lifters program and log range of motion?
Train full range of motion as the default on most compound lifts. It's the condition with the broadest strength transfer in the Pallarés trial and the safe middle ground the meta-analysis's overall average favours.
Add lengthened partials deliberately, not as a replacement. If a specific muscle is lagging — say the pattern Mannarino et al. found for effective reps, where a secondary mover stalls before the prime mover fatigues — a few sets of lengthened partials at that muscle's stretched position is a targeted fix the 2025 trial supports. Reserve shortened, top-end partials for a narrow, specific purpose, such as a known lockout weakness, rather than a general training method; the Pallarés data gives no reason to expect a strength or size payoff from them, only more reported joint discomfort.
The part that's easy to lose track of is which range you actually trained on a given day. A logbook that only records weight and reps can't distinguish a full squat from a half squat, or a lengthened partial curl from a normal one — the number on the bar looks the same either way. IronLedger's per-set Notes field is built for exactly this: a short note on the set ("bottom-half only," "full depth") sits next to that set's load in your history, so a progress page trend line can be interpreted correctly months later instead of mixing two different exercises under one name.
If you're building this into a program rather than bolting it onto one, IronLedger's Starting Strength template already requires full depth on every squat rep by design, and the custom program builder lets you schedule a lengthened-partial accessory exercise on its own line, separate from — not substituting for — the full-ROM main lift. Pairing that with a sensible set count per muscle group keeps a lengthened-partial addition targeted instead of turning into uncounted extra volume.
What mistakes do lifters make with range of motion?
Most range-of-motion mistakes come from treating "partial rep" as one thing instead of two opposite things.
- Treating all partial reps as interchangeable. Lengthened partials at the stretched end and shortened partials at the contracted end produced opposite results in the research above — one matched full ROM, the other underperformed it.
- Using shortened partials to chase a bigger number. A quarter-squat or lockout-only "PR" reflects strength in that narrow range only; the Pallarés trial found half squats produced no significant transfer to full-depth strength.
- Cutting depth under load without expecting a cost. The half-squat group in the Pallarés trial was the only one with a significant increase in pain, stiffness, and functional disability — cutting range to handle more weight is not a free trade.
- Skipping the stretched position by default. The 2025 trial and the meta-analysis's sub-group both point to the stretched portion of a lift as doing meaningful hypertrophy work; avoiding it because it's the hardest part of the rep works against that.
- Not logging which ROM was used. A set logged only as weight and reps can't be told apart from the same lift done at a different depth; use a per-set note so a stalled or rising trend line on the progress page means what you think it means.
- Assuming a lengthened-partial result generalises to shortened partials. The 2025 trial tested the stretched-position case specifically — it isn't evidence for cutting range from the top of a lift instead.
Frequently asked questions
On average, yes, by a small margin (SMD 0.12, Wolf et al., 2023). But lengthened partials at the stretched position matched full ROM directly in a 2025 trial, so the answer depends on which partial you mean.
Full range of motion and lengthened partials aren't competing methods — the 2025 trial found they produce comparable size and strength-endurance gains when the lengthened partial specifically targets the stretched position. Shortened, top-end partials are the outlier: the Pallarés squat trial found them worse for strength transfer and worse for joint comfort, not just neutral. Default to full ROM, add lengthened partials on purpose where a muscle needs it, and log which range you actually used so the data can tell you whether it's working.
Sources and references
- Wolf, Androulakis-Korakakis, Fisher, Schoenfeld & Steele. Partial Vs Full Range of Motion Resistance Training: A Systematic Review and Meta-Analysis — Bayesian meta-analysis finding a trivial overall SMD of 0.12 favouring full ROM, with a sub-group estimate favouring partial reps at long muscle lengths for hypertrophy (2023).
- Wolf, Androulakis Korakakis, Piñero, Mohan, Hermann, Augustin, Sapuppo, Lin, Coleman, Burke, Nippard, Swinton & Schoenfeld. Lengthened partial repetitions elicit similar muscular adaptations as full range of motion repetitions during resistance training in trained individuals — 30-participant within-participant RCT finding lengthened partial reps and full ROM produced similar muscle thickness and 10RM strength-endurance gains over 8 weeks (2025).
- Pallarés, Sánchez-Medina, Pérez, De La Cruz-Sánchez & Mora-Rodríguez. Full squat produces greater neuromuscular and functional adaptations and lower pain than partial squats after prolonged resistance training — 53-person, 10-week RCT finding only full squats gained strength across every tested depth, while half squats gained nothing and reported significantly more pain (2020).
- IronLedger product page — verifies current per-set logging fields including Notes, RPE, RIR, and the program builder used in this article (2026).
Keep learning
