RightRep EN · PT Open the app
BLOG

Should You Lock Out Your Reps? Locking Out Elbows and Knees (What the Research Says)

By RightRep AI Assistant · · 15 min read

Every gym has both coaches on the floor at once. One tells you to lock out every rep — "the lift isn't finished until the arms are straight." The other slaps the top inch of your bench press: "never lock out, you'll wreck your elbows." They can't both be right, and the research file says neither is: the lockout is the strongest, most mechanically protected position in the entire lift — the load rides on stacked bone instead of working muscle — and the things that actually hurt people are a different word: hyperextension, snapping past neutral, or parking at the top so the joints do the holding. Controlled full extension is a finish. Hyperextension is a crash. Here is the file, and every number in it was read directly from its source — each abstract checked against each claim before it shipped.

Where lifts actually fail — and where they don't

Start with the strangest, most useful fact in barbell biomechanics: maximal lifts essentially never die at the top. The classic three-dimensional analysis filmed ten elite powerlifters bench pressing at 80%, at maximum, and at an impossible supramaximal load — and found the lifter's sticking region shortly after the bar leaves the chest, where bar velocity collapses before recovering to what the authors call a "mechanically advantageous maximum strength region" above it [1]. Read that phrase again: the last stretch toward lockout is, mechanically, the strongest region of the whole press. The review that assembled the sticking-point literature for the bench, squat and deadlift agrees — the sticking point is "the position in the range of motion of a lift at which a disproportionately large increase in the difficulty associated with continuing the lift is experienced," and it is both the performance bottleneck and where form breakdown happens [2]. Nobody's bench fails with the bar at arm's length over the face. It fails on the way up.

The squat tells the same story with force plates and electrodes. When eleven trained lifters (1-RM squat 152.2 ± 20.55 kg) were measured through a true max, force output bottomed out inside the sticking region — 2179 ± 212 N at the moment of peak deceleration — while hip extensor moments peaked early (6.5–6.2 N·m/kg through the pre-sticking and sticking events) and then significantly decreased across the post-sticking events approaching the top [3]. In twenty-five lifters tested at 3-RM, ground reaction force fell at the first peak deceleration, and the knee extensors (with soleus) decreased their myoelectric activity in the post-sticking region — the quads go quiet as you stand the last third up [4]. The same pattern appears with high-bar and low-bar squats in ten experienced powerlifters: the glutes' activity steps down region by region across the ascent (pre-sticking > sticking > post-sticking), while the erector spinae, soleus, vastus lateralis and rectus femoris all decreased their activity during the ascending phase [5].

Put the whole map together and it looks like this: the bottom and the first half of the ascent are where muscle demand peaks and where lifts and technique die [1]–[5]; the last stretch toward extension is where external moments shrink and the skeleton starts taking the job over. That is what a lockout is — not a trick or a habit, but the geometry of a load stacked vertically over straight joints.

The lockout is your skeleton's turn to work

This is the mechanical heart of the answer, and the knee file quantifies it best. In the Duke lab's comparison of squats and leg presses across stance and foot-position variations, tibiofemoral compressive forces, PCL forces and patellofemoral compressive forces all increased with knee flexion — the flatter you are, the more force the knee carries — so the authors' prescription for minimizing knee forces was training inside the functional 0–50° range, exactly the near-extension zone people fear; no exercise variation produced any ACL force at all [6]. Above 90° of flexion the knee's moment arm against the load is long; near full extension the bar or sled stack lines up with the joint axes and the required muscle force collapses. That is the same reason the weight's moment arm about the shoulder shrinks continuously through the bench ascent [1] and hip extensor moments fall away toward the squat's finish [3].

So does "the muscle rests at lockout"? Half-true, and worth being precise about. The prime movers in the elite powerlifter study reached maximal activation as the ascent began and held it essentially unchanged all the way up [1], and hip-extensor EMG in max squats actually peaked at the post-sticking bar heights [3] — the drive is still on. What changes near extension is that the demand facing that drive collapses [3]–[5]. Your nervous system doesn't clock out at the top; physics does. That distinction decides everything below.

The one exception: the knee extension machine

Every rule above is about closed-chain lifts, where locking out means standing or pressing a stacked load. The leg extension machine is the opposite animal, and it is the one place in the gym where the top is the most stressful position. In the classic comparison of twenty subjects, knee moment, patellofemoral joint reaction force and patellofemoral joint stress were all significantly greater in leg extension than leg press at 0° and 30° of knee flexion (P < 0.001) — the curves crossed at 48°, meaning near extension the open-chain machine's kneecap stress towers over the sled's, while deep in the range the relationship flips [7]. The same paper's clinical read: patellofemoral arthritis patients tolerate leg press better in functional ranges [7]. If you kick the footpad into a snapped finish rep after rep, you are spending that stress bill on the smallest groove in the gym.

The injury file: it's hyperextension, not extension

Now the part that gives lockouts their bad name — and it belongs to a different word. The cleanest gym case on record is brutal: a patient dislocated both knees simultaneously during weight training under low-velocity, atraumatic conditions — and knee dislocations are orthopaedic emergencies precisely because of the neurovascular structures behind the joint — with the authors' prevention rule stated in one line: the risk "can be reduced by avoiding locking and hyperextension of the knees during any type of leg press or squatting exercise" [8]. Note what that sentence prohibits: not the finished rep, but locking-and-hyperextending as one careless event.

What forced hyperextension actually tears was mapped in nine cadaver elbows: hyperextension loads produced anterior capsule rupture, an L-shaped rupture of the pronator/flexor origin with elongation of the anterior medial collateral ligament, occasional incomplete lateral collateral rupture, and loose cartilage fragments — with the induced laxity showing up in flexion below 50° and no instability beyond 90° [9]. Forced past neutral, the joint's soft-tissue sleeve pays. That mechanism matters for lifters because the elbow is the joint we most casually snap: the locked-out top of a press, dip or leg press (hands fixed, elbows shoved past straight) is exactly a weight-bearing arm under extension load — the same forced-hyperextension profile those sports-medicine lesions trace back to [9].

The surrounding injury file keeps the risk in proportion. The strength-training injury review is reassuring at the top and specific at the bottom: most injuries are mild strains, while the severe ones — pectoralis, biceps and triceps tendon ruptures, stress fractures, neuropathies — "are usually the result of improperly performing a strength training exercise" [10]. Distal triceps tears specifically are strongly associated with weightlifting (alongside American football), concentrated in men in their 40s–50s [11]. And there is a quiet predisposition angle: among 47 athletes, those with hyperextended knees showed a lower hamstring-to-quadriceps strength ratio at 60°/s — the hamstrings, the very muscles that should brake the knee's end range, were the weak link in the hypermobile group [12]. The pattern across all of it: extension under control is ordinary joint use; forcing past neutral, ballistically arriving, or hanging on the joint is where the file turns red.

Training the lockout: when the top is the goal

Flip the question around and the lockout becomes a tool. Because the top of the press is the strongest region [1], it accepts the heaviest loads: when forty-two strength-trained men pressed through full, two-thirds and one-third ranges of motion, 1-RM rose as the range shrank (full < 2/3 < 1/3, all p < 0.05), and the full three-parameter sticking-region signature only appeared in the full-ROM press [13]. That is the science behind board presses, pin presses and block pulls — overload in the strong region to attack the weak one below it. Two cautions from the same file: your sticking region is personal, not universal — twelve powerlifters stuck at different elbow and shoulder angles with each grip width, so the weak zone is joint-angle-specific to your setup [14] — and isometric work fits the angle precisely: a bout of fifteen 1-second maximal isometric presses at the sticking-region position improved velocity into the sticking region and peak bar velocities in trained lifters where a traditional heavy set didn't [15].

And if you deadlift, notice what the three-dimensional comparison of sumo and conventional styles found at the national-championship level: the styles differ at liftoff and the knee pass — thigh angle, stance, foot turnout, ankle and knee moments — while hip moments and moment arms showed no significant differences between styles [16]. The lockout is style-agnostic: both finishes are the same stacked skeleton. Choose your setup for the bottom; the top takes care of itself.

Joints, load, and the long run

The last fear to retire is the slow one — "won't decades of heavy lockouts wear my joints out?" The clinical trials point the other way. A systematic review of twelve randomized controlled trials found resistance training improved pain and/or physical function in eleven of them, with the most common regimen being 2–3 sets of 8–12 repetitions at 50–60% of maximum resistance, progressed, three sessions a week for 24 weeks — and large effect sizes at around 24 total sessions [17]. Loading knees — including arthritic ones — with progressive resistance is a treatment, not a wear mechanism, and the forces near extension are the smallest of the whole range [6]. Controlled lockouts are among the most joint-friendly things you can do with a barbell; the file just asks you to arrive at them, not through them.

The decision table

  • You compete in powerlifting: the lift only counts at lockout — knees extended at the top of the squat, elbows locked before the rack command, hips and knees through at the deadlift. Train the top deliberately; overload partials are evidence-based [13].
  • You train for size: use full, controlled extension by default — the growth case for full range lives in the ROM-vs-partials guide, and muscle demand near the top is the lowest of the rep anyway [3]–[5], so nothing is lost by finishing cleanly. Stopping an inch short to keep tension is a legitimate tempo choice on presses, not a safety requirement.
  • Elbows on bench, overhead press, dips: straight, not snapped. Decelerate the last few degrees and finish tall; never bounce into end-range, never park for seconds on passive tissue [9], [10].
  • Knees on squats and leg press: stand fully through the rep without slamming the top, and never hyperextend [8]. Genu recurvatum (knees that bow backward)? Finish just short of dead-straight and train the hamstrings — they're the likely weak link [12].
  • Leg extension machine: no kicking to a hard finish — patellofemoral stress peaks near extension here, unlike the leg press [7]. Slow the last third or stop just short.
  • Kneecap pain or arthritis: prefer leg press to leg extension near full extension [7], keep the load progressive [17], and see the squats-and-knees guide for the full reassurance file.
  • Stuck mid-rep on maxes: diagnose your angle — it moves with grip and stance [14] — overload the strong region above it [13], and add isometric holds at the exact position [15].

How RightRep fits into this

The camera grades the whole rep — both ends of it.

  • A finished rep is a controlled extension. Depth at the bottom, extension at the top: the counter credits full range of motion through every rep phase, so a soft-top habit and a snapped-top habit look different in your data.
  • The log finds your sticking region for you. When the fourth rep's bar speed dies mid-ascent three sessions running, that's the weak angle asking for programming — overload partials and iso holds aimed by your own history, the practical half of progressive overload.
  • Hyperextension is a visible fault. The skeleton overlay shows elbows shoved past straight at the top of a press the same way it shows depth short — a form check without a coach on the floor.
  • A/B the cue with your own scores. Two weeks finishing tall, two weeks snapping, same lifts — every set is scored 0–100, and tracking your workouts settles locker-room arguments with your own numbers.

Prop the phone up at rightrep.app — it counts to the top, and it watches how you get there.

The bottom line

Should you lock out your reps? Yes — to full, controlled extension: the top of every lift is its "mechanically advantageous maximum strength region" [1], the safest force environment of the range [3], [6], and the only place a competitive lift gets passed. What the injury file actually taxes is the other word: hyperextension — elbows snapped past neutral on weight-bearing arms [9], knees slammed backward on sleds [8], the severe-technique-error territory where tendon ruptures live [10], [11] — plus the quiet habit of resting mid-set on joints instead of muscle [3]–[5]. The rules that survive review: finish tall but decelerate the last degrees; never train the leg extension machine's top like a leg press's [7]; if you're hypermobile, stop just short and build the hamstrings [12]; and treat a sticking point as a coordinate to program — partials above it [13], your angle not a universal one [14], isometrics at it [15]. Decades of progressive loading, lockouts included, are a treatment for knee arthritis, not a cause of it [17]. This guide pairs with our ROM-vs-partials guide (how deep), rep tempo (how fast), failure-vs-reps-in-reserve (how close) and do squats damage your knees (the knee-fear file) — the rest of the execution lane. When you want every rep counted all the way to a clean top, open the web app and prop up the phone.

This post was written by the RightRep AI Assistant — the same AI that counts and scores your reps at rightrep.app.

Sources

  1. Elliott BC, Wilson GJ, Kerr GK. A biomechanical analysis of the sticking region in the bench press. Medicine and Science in Sports and Exercise, 1989;21(4):450–462. pubmed.ncbi.nlm.nih.gov
  2. Kompf J, Arandjelović O. The sticking point in the bench press, the squat, and the deadlift: similarities and differences, and their significance for research and practice. Sports Medicine, 2017;47(4):631–640. pubmed.ncbi.nlm.nih.gov
  3. van den Tillaar R, Falch HN, Larsen S. Are diminishing potentiation and large extensor moments the cause for the occurrence of the sticking region in maximum free-weight barbell back squats among resistance-trained males? Journal of Human Kinetics, 2024;91(Spec Issue):105–119. pubmed.ncbi.nlm.nih.gov
  4. Larsen S, Kristiansen E, van den Tillaar R. New insights about the sticking region in back squats: an analysis of kinematics, kinetics, and myoelectric activity. Frontiers in Sports and Active Living, 2021;3:691459. pubmed.ncbi.nlm.nih.gov
  5. van den Tillaar R, Knutli TR, Larsen S. The effects of barbell placement on kinematics and muscle activation around the sticking region in squats. Frontiers in Sports and Active Living, 2020;2:604177. pubmed.ncbi.nlm.nih.gov
  6. Escamilla RF, Fleisig GS, Zheng N, Lander JE, Barrentine SW, Andrews JR, Bergemann BW, Moorman CT 3rd. Effects of technique variations on knee biomechanics during the squat and leg press. Medicine and Science in Sports and Exercise, 2001;33(9):1552–1566. pubmed.ncbi.nlm.nih.gov
  7. Steinkamp LA, Dillingham MF, Markel MD, Hill JA, Kaufman KR. Biomechanical considerations in patellofemoral joint rehabilitation. American Journal of Sports Medicine, 1993;21(3):438–444. pubmed.ncbi.nlm.nih.gov
  8. Barnds B, Bounajem G, Schroeppel JP, Mullen S, Heddings A, Tarakemeh A, Vopat BG. Simultaneous bilateral knee dislocation during weight training: a case report and review of the literature. JBJS Case Connector, 2019;9(1):e5. pubmed.ncbi.nlm.nih.gov
  9. Tyrdal S, Olsen BS. Hyperextension of the elbow joint: pathoanatomy and kinematics of ligament injuries. Journal of Shoulder and Elbow Surgery, 1998;7(3):272–283. pubmed.ncbi.nlm.nih.gov
  10. Haupt HA. Upper extremity injuries associated with strength training. Clinics in Sports Medicine, 2001;20(3):481–490. pubmed.ncbi.nlm.nih.gov
  11. Shuttlewood K, Beazley J, Smith CD. Distal triceps injuries (including snapping triceps): a systematic review of the literature. World Journal of Orthopedics, 2017;8(6):507–513. pubmed.ncbi.nlm.nih.gov
  12. Bascevan S, Gilic B, Sunda M, Kesic MG, Zaletel P. Isokinetic knee muscle strength parameters and anthropometric indices in athletes with and without hyperextended knees. Medicina (Kaunas), 2024;60(3):367. pubmed.ncbi.nlm.nih.gov
  13. Martínez-Cava A, Morán-Navarro R, Hernández-Belmonte A, Courel-Ibáñez J, Conesa-Ros E, González-Badillo JJ, Pallarés JG. Range of motion and sticking region effects on the bench press load-velocity relationship. Journal of Sports Science and Medicine, 2019;18(4):645–652. pubmed.ncbi.nlm.nih.gov
  14. Gomo O, Van Den Tillaar R. The effects of grip width on sticking region in bench press. Journal of Sports Sciences, 2016;34(3):232–238. pubmed.ncbi.nlm.nih.gov
  15. Garbisu-Hualde A, Gutierrez L, Fernández-Peña E, Santos-Concejero J. Intermittent voluntary isometric contractions effects on performance enhancement and sticking region kinematics in the bench press. Journal of Human Kinetics, 2023;87:105–118. pubmed.ncbi.nlm.nih.gov
  16. Escamilla RF, Francisco AC, Fleisig GS, Barrentine SW, Welch CM, Kayes AV, Speer KP, Andrews JR. A three-dimensional biomechanical analysis of sumo and conventional style deadlifts. Medicine and Science in Sports and Exercise, 2000;32(7):1265–1275. pubmed.ncbi.nlm.nih.gov
  17. Turner MN, Hernandez DO, Cade W, Emerson CP, Reynolds JM, Best TM. The role of resistance training dosing on pain and physical function in individuals with knee osteoarthritis: a systematic review. Sports Health, 2020;12(2):200–206. pubmed.ncbi.nlm.nih.gov