How To Do the Front Squat
Heavy loads quietly tip you forward and hand the work to your hips. The front squat will not let that happen, because the bar falls off. Here is what the evidence supports and where I am borrowing from the back squat.
Start here
The front squat puts the bar across the front of your shoulders instead of your upper back. Elbows up, chest up, and you squat.
The short version of the technique: keep your elbows pointing forward and high, keep your torso as upright as you can, and go as deep as you can hold that position.
That upright torso is not a style choice. It is enforced. Lean forward on a back squat and the bar stays put while your hips take over. Lean forward on a front squat and the bar rolls off your shoulders. The lift audits you every rep.
Honest warning before I go further: the front squat specifically has thin research. One of the studies below tested it directly. The rest are back squat or general squat work, and I will flag each time I borrow.
Load makes you lean, unless something stops you
This is the finding that explains why the front squat exists.
A computational study modelled the body as linked segments with limits on how much torque each joint can produce, then asked what posture is mechanically possible as the barbell gets heavier. The prediction: higher loads produce systematic posture changes, specifically increased torso inclination and a shift of rotational demand away from the knee and toward the hip.
Real lifters do the same thing. Twenty-nine national and international powerlifters squatted at 70 to 90 percent of their max on force plates with 3D motion capture. As load went up, hip moments increased significantly while knee and ankle moments did not change. The relative demand shifted from knee and ankle toward the hip.
Plain English: when it gets heavy, you tip forward and your hips rescue the lift. Both a model and a room full of elite lifters agree.
The front squat is the version where you cannot do that. Both of these studies are back squat or generic squat, so I am inferring the front squat's advantage rather than quoting it. But the mechanism is the mechanism.
The one study that tested the front squat directly
Twelve resistance-trained men, averaging eight years of training, squatted at 70 percent of their max under three foot positions: flat heel, heel elevated, and forefoot elevated. Both front and back squat. EMG on the lower limb, plus motion capture and force plates.
The quadriceps result was clear. Vastus lateralis, vastus medialis and rectus femoris all showed significantly greater activation in the flat-heel and heel-elevated positions compared with forefoot elevated.
So if you are front squatting for quads, do not put your toes on a plate. Flat feet or a raised heel are the choices. A lifting shoe with a raised heel is doing something real.
Twelve trained men at a single load. Directionally useful, not a law.
Depth is set by the position you can hold, not by a number
There is no study telling you the correct depth for a front squat. What the evidence does say is that posture degrades as load rises, and that trained lifters still show measurable technique shifts under intensity and fatigue.
A systematic review looked at exactly that across the squat, bench and deadlift in experienced lifters, and found that increasing intensity and accumulating fatigue produce measurable changes in movement, with implications for both performance and injury risk, even though trained lifters are more stable than novices.
So experience buys you consistency. It does not buy you immunity.
For the front squat that translates simply. Your depth is wherever your elbows stay up and your torso stays tall. When your elbows drop, the set is over. That is not toughness, it is the bar telling you the position is gone.
Speed on the way up is a lever you can pull
Twenty-four trained collegiate athletes squatted under three eccentric strategies at matched depth: normal constant load, heavier eccentric than concentric, and normal load lowered at maximum speed.
With identical concentric load and range of motion, deliberately accelerating gave significantly higher mean concentric barbell velocity than the other two.
The practical read: intent matters. Same weight, same depth, different outcome depending on whether you drive out of the bottom or just survive it.
Back squat, 24 male athletes, acute session. Do not turn this into a program on its own.
How to actually do it
Set the bar across the front of your shoulders, not in your hands. Your fingers are a shelf, not a grip. If your wrists hurt, that is usually a sign you are trying to hold the bar with your hands.
Elbows up and pointing forward. This is the whole lift. High elbows keep the shelf, and the shelf keeps your torso upright.
Feet flat or heels slightly raised. Do not elevate your toes if you want your quads working.
Brace, then squat down between your hips with your torso tall. Go as deep as you can hold that torso.
Drive up with intent, not just enough to survive.
End the set when your elbows start dropping. That is the position leaving, and continuing past it is how the lift turns into a bad-looking back squat.
The front squat's value is that it will not let you cheat. Both a biomechanical model and a study of elite powerlifters show the same thing: as the bar gets heavier, you tip forward and your hips quietly take over. The front squat blocks that because a forward lean drops the bar. Keep the elbows high and forward, keep the feet flat or the heels slightly raised (do not elevate your toes if you want quads), squat as deep as you can hold that upright torso, and drive up with intent. When your elbows drop, the set is finished. Be aware that only one of these studies tested a front squat directly; the rest are borrowed from the back squat.
Limits of this evidence
- Only one of these studies tested the front squat directly. The load-and-posture findings come from back squat and generic squat work and are applied to the front squat by inference, not by demonstration.
- The posture-under-load prediction is a computational model, not a measurement of real lifters, so it shows what is mechanically admissible rather than what people actually do.
- Sample sizes are small: 12, 29 and 24 participants in the experimental studies.
- The foot-position study used a single load (70 percent 1RM) in trained men only, so it says nothing about beginners, women, or heavier loads.
- The elite powerlifter data used the powerlifting-style low-bar back squat under competition standards, which is close to the opposite of a front squat in torso position.
- The eccentric-speed study was a single acute session, so it shows an immediate effect on bar velocity, not a training outcome.
- No study here establishes a correct squat depth, so the depth guidance is reasoning from the posture findings rather than a measured recommendation.
- Each claim rests on a single source with no independent replication checked. Treat the piece as provisional.
Sources (5)
- Mechanobiological and neuromuscular responses to foot-position variations during front and back squat exercises. Frontiers in Physiology, 2025
- Numerical Modeling of Load-Driven Changes in Squat Technique Using a Moment-Limited Joint Framework. Bioengineering, 2026
- Differences in hip, knee, and ankle joint moments during squats across load intensities, gender classes, and performance level in elite powerlifters. Scientific Reports, 2026
- Effects of Intensity and Fatigue on the Kinetics and Kinematics of the Barbell Squat, Bench Press, and Deadlift in Experienced Lifters: A Systematic Review. Sports Medicine - Open, 2025
- Acute Effects of Accelerated Eccentrics and Accentuated Eccentric Loading on Squat Performance and Lower-Limb Biomechanics. Sports, 2025