How To Do the Back Squat
The squat is a big deal for your body. Here's the stuff that actually holds up, told straight, plus where the science is still thin.
Start here
You want to know how to squat right. Fair. It's the lift everyone argues about, and it's the one that loads your body the hardest.
So let me give you the short answer first, then walk you through why.
Good form comes down to a few things you can actually control: brace your belly before you drop, control the way you go down, and pick a foot setup that lets your knees and hips move freely. That's the spine of it. Everything else is detail.
Now the honest part. A lot of what people preach about squats is built on small studies and computer models, not big piles of proof. I'll tell you where the ground is solid-ish and where it's basically a maybe. No pretending.
Your squat puts real load through your body, and that's the point
Let's set the stakes. When you squat heavy, your joints eat a lot of force.
In a group of elite powerlifters squatting at 90 percent of their max, the peak forces were big. Around 15 times bodyweight at the hip. About 23 times at the main knee joint. Roughly 27 times behind the kneecap. About 11 times at the ankle.
Read that again. Twenty-plus times your bodyweight through the knee. That's not a reason to be scared. It's a reason to respect the lift and not treat form like a nice-to-have.
One thing to keep in mind: those numbers came from elite lifters, and they bounced around a lot person to person. So don't treat them as your exact numbers. Treat them as proof that squats are a serious load, and load likes good form.
The brace: fill your belly like a shield
Here's the first real skill. Before you go down, you take a breath into your belly and squeeze, like you're bracing to take a punch. That builds pressure inside your midsection.
Why bother? Because that inside pressure seems to take some load off your spine.
In spine models, cranking up that belly pressure lined up with less squeezing force on the spine across different efforts. Bending forward, the drop was around 18 percent. Bending sideways, around 29 percent. Twisting, around 31 percent. Extending, around 21 percent. Same idea kept showing up: more pressure, less compression on the spine.
Another model added a piece I like. That belly pressure appears to create a little push that helps hold your trunk upright, and it seems to make your lower back stiffer, meaning steadier, when you need it.
Here's the analogy. Think of a shaken, sealed soda can. Squeeze the sides and it barely dents, because the pressure inside holds the shape. Your braced belly does something like that for your trunk.
Where the can breaks down: your body isn't sealed metal, and these numbers come from computer models and small studies, not from watching it happen inside a real lifting human. So I'd call the brace a smart bet, not a settled fact. Solid but not proven.
Timing matters too, not just squeezing
Bracing isn't only about squeezing hard. It's about squeezing at the right moment.
One modeling paper pointed out that if your belly muscles fire weakly, or at the wrong time during a lift, your lower back may be left less stable and more open to injury.
And the faster stuff seemed to matter more when the lift got quicker. In the model, that pressure's steadying job appeared to carry more weight as lifting speed went up, the idea being that faster reps usually mean heavier loads.
So: brace before you move, not halfway down. Keep it tight through the rep.
Straight talk though. That injury link is an author's conclusion from one modeling source, not measured injuries in real people. Treat the timing point as a reasonable habit, not gospel.
Foot position: it changes where the work lands
People fight about heels flat versus heels raised. Here's what I can back.
Changing your foot setup, flat heel, heel raised, or forefoot raised, seemed to shift how load spread out and which muscles fired during front and back squats in trained lifters.
More specifically, flat-heel and heel-raised positions showed more quad activity (the muscles on the front of your thigh) at 70 percent of max compared to another setup.
What you do with that: if you want more quad, a raised heel is worth trying. If you want the load spread differently, a flat foot changes the picture. Neither is wrong. They just aim the work at different places.
The catch: this came from small groups of trained men, and some of it lives in the discussion section rather than hard measured results. So use it to experiment, not as a law.
Control the way down, and squat as deep as you can own
Speed and depth aren't random. They change the load story.
In elite lifters, the hip and ankle hit their peak forces down in the deepest part of the squat. The knee joints, though, stayed loaded high across a big chunk of the whole movement. So depth isn't a free trip. The bottom is where your hips and ankles get hit hardest.
Coming back up, across roughly 70 to 90 percent of max, the joint forces mostly didn't change much with heavier load, with one exception at the hip, where things did seem to differ. So the studies genuinely disagree with themselves a little here. The picture's mixed, not clean.
My take: go as deep as you can while keeping your brace and your control. Don't chase depth you can't own. The bottom is the hardest spot, so earn it.
What about training your breathing muscles?
You'll see people pushing diaphragm work, training the breathing muscle under your lungs.
Here's what showed up. After 8 weeks of diaphragm-focused core training, one group's diaphragm got noticeably thicker (around 35 percent) versus regular core work or nothing. And the authors floated that core work including diaphragm training may help with injury prevention and performance in lifts that need a stable lower back, like squats.
But cool it before you buy a course. That same 8-week program showed no clear difference in one squat measure, the peak knee flexion moment. A thicker breathing muscle is a body change, not proof your squat gets safer or stronger. Don't turn one into the other.
All of this is one small study of 37 men. Thin. Treat it as a maybe worth trying, not a must.
Why any of this is worth your time
Back pain is common and slippery. Manual lifting is thought to feed low back pain through stress on the lower spine, and in a big share of low back pain cases, over 80 percent by one account, nobody can even pin down a clear cause.
That's exactly why form is your lever. You can't control everything. You can control your brace, your speed, your depth, your feet.
And the gadgets? Belts and exoskeletons have shown limits in handling back pain. So don't outsource your squat to gear. Build the habits first.
Do these four things and you're squatting well. One, brace your belly before you drop and hold it, like taking a punch. Two, keep it tight when the bar gets heavy and the reps get faster. Three, pick a foot setup that lets your knees and hips move free, and use a raised heel if you want more quad. Four, control the way down and only go as deep as you can own, because the bottom loads your hips and ankles hardest. Skip leaning on belts to fix your form. Build the habits first. Most of this is solid-but-not-proven, so treat it as a smart bet, not a promise.
Limits of this evidence
- Most of the spine numbers come from computer models and small studies, not from measuring inside real lifters, so treat them as preliminary.
- Each finding here rests on a single source, so I can't check whether other independent studies agree.
- The belly-pressure reductions in spinal force were tied to specific loading setups and may not carry over to every effort, posture, or person.
- The powerlifter joint forces came from a small group of elite lifters and varied a lot between people, so they aren't your exact numbers.
- The foot-position findings came from small groups of trained men, and part of it lives in a discussion section rather than hard measured results.
- The diaphragm study was one small trial of 37 men; a thicker breathing muscle is a body change, not proof of a safer or stronger squat, and it showed no clear difference in one squat measure.
- The injury-risk point is an author's conclusion from a model, not measured injuries in real people.
- In one device study, the '43' figure was unclear as a percentage versus an actual pressure value, which is a reminder that even the numbers can be fuzzy.
- The bounce-back (concentric) picture disagreed with itself, mostly unchanged with load but different at the hip, so that part stays mixed.
Sources (6)
- Finite element investigation of the intrinsic stiffness contribution of intra-abdominal pressure in a transient spine and trunk model. Comput Biol Med
- Feasibility of a novel back support device to improve spine stability and muscular activity during trunk flexion: A prospective cross-sectional study with healthy controls and low back pain subjects - preliminary. Clin Biomech (Bristol)
- Biomechanical analysis of hip, knee, and ankle joint contact forces during squats in elite powerlifters. PLoS One, 2025
- Mechanobiological and neuromuscular responses to foot-position variations during front and back squat exercises. Front Physiol, 2025
- Impact of Diaphragm-Strengthening Core Training on Postural Stability in High-Intensity Squats. Life (Basel), 2024
- Intra-abdominal pressure and abdominal wall muscular function: Spinal unloading mechanism. Clin Biomech (Bristol)