Thoughts on Greg Nuckols "squat guide"?

[DannyP]

Would appreciate thoughts on a couple of things from those who've read this:

https://www.strongerbyscience.com/how-to-squat/

It's not as comprehensive as the Starting Strength book, but still pretty in-depth. Also not quite as rigid. I'm curious in general what people think of the guide; personally I've found some of it more digestible than SS and I appreciate some of his specific thoughts on different permutations of the lift. More specifically, there are two things I'm curious about, since they seem very much in opposition to the SS way:

1. Mixing and matching bar position and technique. Basically, he says there's no reason not to try low bar technique with high bar position and vice versa. In reality, is this problematic/any good reason not to do it if for some reason it feels better (other than it may not produce "optimal efficiency")?

2. Bar path. He says that under about 2x bodyweight, don't expect a completely vertical bar path since "your ass isn't weightless" and therefore the bar will invariably come forward a bit. If true, this would explain some things about my squat. But is this true/is there real merit to it?

Outside the bounds of the SS forums (where I know Greg's philosophy isn't the most popular), what are the thoughts on this guide, and its usefulness in bettering one's squat?

[KyleSchuant]

I've not read it recently, though have scanned it in the past, and 97 pages would take a while to digest, a bit much for a sunny Saturday afternoon. But in response to your points:

1. I teach low-bar to all those physically capable of doing it, but if you're squatting consistently over years, do it however you like, I'm happy. Do your novice linear progression with a front squat if you like, I don't care. It's all squatting.

2. I don't think it's because of bodyweight being a significant fraction of the movement, it's just that noobs, while overall weak, tend to have stronger quads than hamstrings and inner thighs. The body doesn't know muscles, only movement, so when you ask it to do a movement it'll accomplish it in whatever funky way it can, doesn't if it looks like ballet. It does this by using the stronger muscles and trying not to use the weaker muscles.

So the noob gets to the bottom of a squat, their hamstrings and inner thighs aren't that strong so won't find it easy to get them up with hip extension, so their body uses their quads, trying to get up with knee extension. Their knees extend without their hips extending, the bar comes forward; thus "knee slide". This will be less so with a quad-dominant squat like the front squat, and more so with a hamstring-dominant (relatively) squat like the low-bar. Thus people's insulting the low-bar squat as "the good morning squat."

By the time the person gets to a decent squat, their hamstrings and inner thighs won't be a weak link in the chain any more.

[PatrickDB]

2. Bar path. He says that under about 2x bodyweight, don't expect a completely vertical bar path since "your ass isn't weightless" and therefore the bar will invariably come forward a bit. If true, this would explain some things about my squat. But is this true/is there real merit to it?

We can debate the approximate cutoff, but isn't it obvious from basic physics (e.g. the definition of center of mass) this must true for lighter weights if this lifter+barbell system is to be in balance? It's why squatting with the empty bar feels so weird.

If you film yourself (side view) squatting with the bar and then some reasonably heavy weight, I'm sure you'll see the difference. The empty bar will be out over your toes.

[KyleSchuant]

No. By that reasoning, a chair will lean forward when you stand up from it. Loading weight on top of an object moves its center of gravity up, it doesn't move it forwards or back.

You're sloppy with the empty bar because you can be, and still lift it; you can't be sloppy with a heavy weight and still lift it.

[OCG]

The guide? From what I remember: pretty meh. It's a collection of random thoughts and shit he's been told over the years with no real strong recommendations. As much as we might rag on the StSt way, at least it's a coherent philosophy man. With strong, clear recommendations. And I strongly disagree that we want a huge (or any) degree of back angle change in the bottom half of the squat as Greg recommends.

For your points: Depends what you mean by "high bar" and "low bar" and their associated techniques. Mixing the more extreme ends is probably not a good idea. Like having the bar right on top on your traps as high as you can get it and trying to be absolutely as upright as possible with a narrow stance. Or jamming the bar as low as you can, while being as wide and as bent over as you can. Mixing aspects of those two styles isn't a great idea. But say, squatting "Rip style" with a shoulder width stance, moderately bent over, with the bar an inch higher, just above your shoulder joint? Sure.

Bar path with lighter weights will not be centred over the midfoot, however, with a bit of work, it can be pretty straight and back angle can be maintained which is a good thing to practice. Bar bath will be closer to an upside down "J", with the lifter moving the bar forwards to somewhere over the forefoot and then coming mostly straight down. The actual COG that controls forwards lean at the bottom of the squat isn't your entire body, stood up straight, it's just your torso and the bar. Technically, the combined centre of mass moves throughout the movement, as your back changes angle and as your legs fold and move simultaneously forwards and back. So it's not as simple as picking the bar or some arbitrary point high on the torso and going "this point should move in a straight line". Also, it takes way less than 2x BW to squat properly. More like just BW or a little more.

[PatrickDB]

No. By that reasoning, a chair will lean forward when you stand up from it. Loading weight on top of an object moves its center of gravity up, it doesn't move it forwards or back.

I understand your "removal" argument in the squat context as follows: in the bottom of a heavy (say 2x bodyweight) squat, the bar is over the mid-foot. So if we remove the plates, we've only changed the loading top to bottom, not front to back. Hence the barbell stays over the mid-foot. So in a balanced squat with an empty bar, the barbell must be over the mid-foot, not the forefoot.

But I don't think this is right, because the barbell is always just a hair in front of the mid-foot, even with very heavy loads.

I think I can actually prove to you, totally a priori, that your claim is wrong for at least one of the low and high bar positions. Consider a lifter in the bottom of a squat with no load. There are two places you can put the barbell: high bar or low bar. These are a few inches from each other, so at most one can be directly over the mid-foot. The other is not. So if you increase and decrease the load on this point, it moves closer and farther away from the center of mass (but never reaches it exactly).

(What should be true is that both are forward of the mid-foot, which is why the center of gravity is always a hair forward, but I don't want to dirty my hands with claims that require empirical proof...)

[KyleSchuant]

One of the things you'll notice about the center of mass is that if it moves away from the balance point horizontally, the structure falls over. This is why pyramids aren't built with the point at the bottom, it's too unstable. It's also why we don't squat en pointe like a ballerina, why rigid-soled shoes work better, and so on.

As the weight on the bar increases, the center of mass moves towards the barbell. This is the reason for different postures as the weight goes up.

As for high vs low-bar, and with front squat for that matter, what you'll find is that if you want the bar to remain over midfoot, the back angle will change. Thus the below oft-reproduced image.

Below that, find an image of what happens when you let the bar go forward of midfoot, going rear of midfoot doesn't need to be illustrated, as the person will simply fall over with any significant load. Note that this "hips rising too fast" error almost never happens with high-bar or front squat, the noob lifter's natural tendency to want to extend their knees first is exarerbated by the teaching method of hip drive.

[mgil]

Both Rip and Greg are not really correct.

The form mashup advocated is probably to accommodate various anthropometric sets. As an example, Karl Schudt is pretty darn upright during a LBBS rep. Although I haven’t seen the converse, a person in HBBS setup who is extremely horizontal. At least not with heavy weight.

Bar path will certainly deviate from midfoot, especially when the bar is light. Rotational joints moving about in two planes are certainly going to optimize the COM. But when the bar is heavy, it should converge on a straighter path since the COM is focused. The “master cue” (bar over midfoot) is a cue and to focus the lifters actions. Not necessarily what is done.

[Cody]

Balance doesn't have to be perfectly over mid-foot, the ccom has to be balanced over the "slot" between the heel and ball of the foot. We have two points of contact with the ground, not one - otherwise falling over while squatting would be as common as flipping off a Jacob's ladder at a carnival.

Also, the low bar squat leaves your hips further away from that slot, giving your ass's mass more leverage than in the high bar squat. So yes, your rear end does have weight AND it has different leverages based on form. You get more wiggle room with your form in the low bar because of that leverage, like a tightrope walker with his pole held horizontal (where it has more leverage for maintaining balance) instead of vertical (where it has infinitely less leverage).

These concepts expand to deadlifts too. One day I'll flesh out all my thoughts.

[DannyP]

Oh, and that was the other big one I meant to ask about, on the "sit back" vs. "sit down" techniques (as he describes them): On the "sit back" style (which we typically associate with SS-style LBBS), he says that your hips will break first, before the knees. That one is a big departure from the technique advocated in SS, where one is always supposed to break hips and knees simultaneously.

Is this another "different strokes for different folks" (like the mixing and matching form/bar position at the margins), is it anthropomorphic-related or is this an example of observations of a common technique error?

[Savs]

DannyP, I didn't read Nuckol's article and I can't help you with your squat. I'm just a nerd who likes to calculate stuff. Sorry!

OCG, Cody, mgil: agreed.

But I don't think this is right, because the barbell is always just a hair in front of the mid-foot, even with very heavy loads.

I think you're correct, Patrick.

Balance doesn't have to be perfectly over mid-foot, the ccom has to be balanced over the "slot" between the heel and ball of the foot. We have two points of contact with the ground, not one - otherwise falling over while squatting would be as common as flipping off a Jacob's ladder at a carnival.

Cody, fwiw, I agree. You and many others here know the following, so let's just say it's review for me. Considering a diagram of the human foot pointing to the right, I think one will fall over frontwards if the clockwise torque about the (far edge of the) toe is greater than zero. A lifter will not fall over frontwards if the combined center of mass (CCOM) is forward of the toe as long as some force (coming from tension in the calf and feet muscles) is able to create a counter-acting torque. Likewise, one will fall over backwards if the counter-clockwise torque about the (far edge of the) heel is greater than zero. It's difficult to create a counter-acting torque in this direction, and so the interval within which the CCOM must stay is around "midfoot" but extends farther toward the toe than the heel. Agreed?

(What should be true is that both are forward of the mid-foot, which is why the center of gravity is always a hair forward, but I don't want to dirty my hands with claims that require empirical proof...)

I started on the calculation a while back and posted my notes to a thread at the old place. That thread may be deleted now, and I'm too lazy to dig up those pieces of scrap paper or even look at my imgur account. Going from memory, here's what I found.

Treating the lifter's body above the waist as a single rigid body, we can write down equations for the lifter's center of mass (COM) and the bar COM in the horizontal and vertical directions (as viewed from the side (I did account for angles into the third orthogonal direction -- out of the page)). After some manipulation, we'll get four unknowns (ankle angle, knee angle, hip angle, and horizontal bar position (in space)) and three equations. The system is underdetermined.

When squatting with the empty bar, there are an infinite number of ways a lifter can arrange his/her body at each point along the movement and keep the CCOM over midfoot (even using this stricter criterion than the one stating that the CCOM must remain close to midfoot). We have experimental confirmation, that is, we know this is correct from doing it or watching it in real life. However, as soon as we impose a constraint such as the bar must remain over midfoot (or at some specific small distance from midfoot), we've cut the number of unknowns down to three, and since we have three equations, in principle the system is completely determined. That is, at every point in the movement, there is only one way the lifter's body can be arranged such that the CCOM is over midfoot. That specific arrangement depends on the lifter's anthropometry.

The bar COM will always be slightly in front of midfoot because we cannot squat with correct form and keep our body COM over midfoot through the whole range of motion. For example, we have to hold our hands forward if we have no mass in or on the bar.

I think everything I've written is correct, but I'm going from memory. Also, any disputes, corrections, and further elaboration are welcome.

[damufunman]

More physics talk!

I started on the calculation a while back and posted my notes to a thread at the old place. That thread may be deleted now, and I'm too lazy to dig up those pieces of scrap paper or even look at my imgur account. Going from memory, here's what I found.

Treating the lifter's body above the waist as a single rigid body, we can write down equations for the lifter's center of mass (COM) and the bar COM in the horizontal and vertical directions (as viewed from the side (I did account for angles into the third orthogonal direction -- out of the page)). After some manipulation, we'll get four unknowns (ankle angle, knee angle, hip angle, and horizontal bar position (in space)) and three equations. The system is underdetermined.

When squatting with the empty bar, there are an infinite number of ways a lifter can arrange his/her body at each point along the movement and keep the CCOM over midfoot (even using this stricter criterion than the one stating that the CCOM must remain close to midfoot). We have experimental confirmation, that is, we know this is correct from doing it or watching it in real life. However, as soon as we impose a constraint such as the bar must remain over midfoot (or at some specific small distance from midfoot), we've cut the number of unknowns down to three, and since we have three equations, in principle the system is completely determined. That is, at every point in the movement, there is only one way the lifter's body can be arranged such that the CCOM is over midfoot. That specific arrangement depends on the lifter's anthropometry.

The bar COM will always be slightly in front of midfoot because we cannot squat with correct form and keep our body COM over midfoot through the whole range of motion. For example, we have to hold our hands forward if we have no mass in or on the bar.

I think everything I've written is correct, but I'm going from memory. Also, any disputes, corrections, and further elaboration are welcome.

The bolded parts I think aren't entirely consistent. I have no issue with the first, but the second one isn't quite correct, we can vary the ankle, knee and hip angles to get different configurations. Take a LBBS for example, and have the knees shift from vertical to forward past the toes (and let's assume the thighs remain parallel to the ground). the ankle and knee angles both close, and the hip angle opens (back angle changes) to keep the bar over midfoot.
This was an issue I had (in my head, not while squatting. Well, actually while squatting too, but thinking-wise) with the SS model. The knee position wasn't defined and they were somewhat arbitrarily saying more or less horizontal back angle. Knees move to position by 1/23 way down then stay there. But where is that?

[Savs]

The bolded parts I think aren't entirely consistent. I have no issue with the first, but the second one isn't quite correct, we can vary the ankle, knee and hip angles to get different configurations.

I believe those other configurations will not put the CCOM directly over midfoot, assuming the bar is very close to a position over midfoot. There is only one configuration that will do that, the math doesn't lie. The catch is, I think, we don't need to be exactly over midfoot, as Cody and mgil pointed out.

Also, to add to my earlier post now that I'm rereading it, maybe it wasn't clear. Patrick, I meant that the body COM is always at least a little behind midfoot, so the bar has to be in front. I think that's correct.

[damufunman]

I believe those other configurations will not put the CCOM directly over midfoot, assuming the bar is very close to a position over midfoot. There is only one configuration that will do that, the math doesn't lie. The catch is, I think, we don't need to be exactly over midfoot, as Cody and mgil pointed out.

Also, to add to my earlier post now that I'm rereading it, maybe it wasn't clear. Patrick, I meant that the body COM is always at least a little behind midfoot, so the bar has to be in front. I think that's correct.

Oops, missed combined COM part. I stand corrected. Do we want the body's COM (exactly) over midfoot? This then turns into an unweighted squat with COM over midfoot, which usually feels odd to me.

[Savs]

Do we want the body's COM (exactly) over midfoot? This then turns into an unweighted squat with COM over midfoot, which usually feels odd to me.

I have to stay out of making recommendations about what we should do to lift more weight, or whatever the goal is. I defer to experiencd coaches when it comes to correcting people's technique. However, I don't think we can put the body's COM directly over midfoot throughout the whole loaded squat movement. There just aren't enough body segments with enough mass in front of midfoot to balance those behind midfoot. I think that's true.

While I'm rereading my earlier post, I think the following statement is wrong.

A lifter will not fall over frontwards if the combined center of mass (CCOM) is forward of the toe as long as some force (coming from tension in the calf and feet muscles) is able to create a counter-acting torque.

I think once the CCOM gets past the front of the foot or the back of the heel, it's over. Done. No saving it.

[CoconutChris]

Both Rip and Greg are not really correct.

The form mashup advocated is probably to accommodate various anthropometric sets. As an example, Karl Schudt is pretty darn upright during a LBBS rep. Although I haven’t seen the converse, a person in HBBS setup who is extremely horizontal. At least not with heavy weight.

For those who don't know, Karl is also kyphotic. Most kyphotic lifters I see end up squatting with an upright position, at least when inferred at the lumbar spine, likely because the bar is more vulnerable to rolling than with a lifter with normal spinal alignment. Jordan is an exception, but his kyphosis is not as bad, and note how when his squat form breaks down, it begins with loss of thoracic extension and having the bar roll forward.

[OCG]

Oh, and that was the other big one I meant to ask about, on the "sit back" vs. "sit down" techniques (as he describes them): On the "sit back" style (which we typically associate with SS-style LBBS), he says that your hips will break first, before the knees. That one is a big departure from the technique advocated in SS, where one is always supposed to break hips and knees simultaneously.

That would be the bog standard typical, uhh, powerlitingscience? View of it. Fuck, what's a good equivalent to broscience? But yeah, nothing special here, it's just yet another thing he was told by some other lifter and is repeating without much thought.

Is this another "different strokes for different folks" (like the mixing and matching form/bar position at the margins), is it anthropomorphic-related or is this an example of observations of a common technique error?

So. this is another "cues vs. form" thing were Greg is kinda unclear so we have to make some assumptions. "Sit back" is simply a cue, to emphasise hip involvement and possibly bending over more or getting the bar back, depending on how you do it. Now, one can but does not have to do this by moving the hips back at the start without bending the knees, but you'll have to bend them quickly and it's kind of a weird way to start a squat. The better way is probably just to emphasis moving the hips back without over thinking what your knees are doing.

[Murelli]

I've not read it recently, though have scanned it in the past, and 97 pages would take a while to digest, a bit much for a sunny Saturday afternoon. But in response to your points:

1. I teach low-bar to all those physically capable of doing it, but if you're squatting consistently over years, do it however you like, I'm happy. Do your novice linear progression with a front squat if you like, I don't care. It's all squatting.

2. I don't think it's because of bodyweight being a significant fraction of the movement, it's just that noobs, while overall weak, tend to have stronger quads than hamstrings and inner thighs. The body doesn't know muscles, only movement, so when you ask it to do a movement it'll accomplish it in whatever funky way it can, doesn't if it looks like ballet. It does this by using the stronger muscles and trying not to use the weaker muscles.

So the noob gets to the bottom of a squat, their hamstrings and inner thighs aren't that strong so won't find it easy to get them up with hip extension, so their body uses their quads, trying to get up with knee extension. Their knees extend without their hips extending, the bar comes forward; thus "knee slide". This will be less so with a quad-dominant squat like the front squat, and more so with a hamstring-dominant (relatively) squat like the low-bar. Thus people's insulting the low-bar squat as "the good morning squat."

By the time the person gets to a decent squat, their hamstrings and inner thighs won't be a weak link in the chain any more.

1) I disagree, squats are not all equivalent. If you only front squat you have to change deadlift volume accordingly if you want "general strength" (whatever that is). StStLP is built around the low bar squat, and only gets away (for some people, for some time) with the absurdly low pulling volume because of the LBBS carryover into the pulls (again, for some people, for some time).

2) Some unathletic (in the general sense - hand-eye coordination, motor coordination, balance, proprioception and explosiveness are features of "athleticism") people have a lot of trouble with bodyweight squats but can squat better with weight on their backs. This is where that combined COM things get handy. In laymen terms, the bar enforces your balance. The quads vs. posterior chain thing is not because of a natural tendency, since we humans use hip extension a lot for most natural movements, it's because of poor lower back control - a thing that is worsened by a lifetime of sitting.

Finally, knee slide. You got the term wrong. Unresisted knee extension = a GM squat, that part is right. Knee slide is when your knee shifts forward in the bottom of the squat, and has a lot of probable causes that have to be diagnosed by the lifter's symptoms (hip flexor pain, knee pain, loss of balance, etc.) and by watching their form. Knee slide is often caused by squishiness in the bottom, not enough knees out, incorrect model (trying to high-bar a lowbar), or a failure of the calves (hey, @Hanley).

Failure of posterior chain and adductors may lead to knee cave and knee slide, but a failure of the quads is what often leads to GMing your squat. This part Carni got right.

[KyleSchuant]

I don't know of anyone who only front squatted for years. I've known people to emphasise it, but they were all weightlifters, so they did plenty of pulls from the floor. Enormous numbers of people have only high-bar squatted, and nothing awful happened. This mythical muscle imbalance from the "wrong" squat in practice doesn't happen, ever, because nobody only squats. Plenty only bench, but nobody only squats, except for a few months because of some injury.

People lift better with some weight than none, simply because they have to. You can snatch or squat the empty bar sloppily, it's extremely hard to snatch 60kg sloppily, or squat 140kg. Having a nonzero weight makes you do things properly. This is why so many Crossfit gyms go wrong in trying to teach the lifts with a stick - lots of people can't even do a front rack position with a stick, but have no trouble with the barbell. They try to teach them with the stick, it doesn't work well, they blame muscle imbalances, prescribe corrective exercises, six weeks later try agina with the empty bar, voila, it works, magic! And the trainer gets to pretend to themselves they're a physiotherapist.

Knee slide leads to good morning squats, too. If they slide too far forward then they slide back without the hips extending, so the bar comes forward and it becomes a good morning.

Yes, knee slide has many causes. But a good morning squat is a common effect of knee slide. And knee slide in a noob is usually, as I said, their body trying to use their stronger muscles and avoid using their weaker muscles. Knee cave often accompanies it, as does thoracic rounding, etc.

[perman]

People lift better with some weight than none, simply because they have to. You can snatch or squat the empty bar sloppily, it's extremely hard to snatch 60kg sloppily, or squat 140kg. Having a nonzero weight makes you do things properly.

Having read your and Murelli's post, it seems like you're not even acknowledging his point. Without a bar, you basically have to force half your bodyweight in front of your COM and half behind it, while with a bar with weight if you start balanced and squat with a straight bar path, you stay in balance. Meaning that squatting without weight actually requires more coordination in that sense.

You should listen to Murelli, he gets less mechanics cred than he should since Savs hogs all that particular glory.