Article: Common Squat Errors (Community Project)

[PatrickDB]

Introduction.

The purpose of this thread is to communally draft an article about squat technique.

It was @mgil's hope when this forum was started that it could also host various long-form articles by its members, and after talking with him, we decided that a community project like this would be a good way to revive that mission.

The Problem.

People are confused about how to low bar back squat.

If you check the technique sections on the SS forum, the squat accounts for by far the largest volume of threads and concerns. Even among posters who have read the book in detail, a number of common and easily diagnosable errors remain. Further, the book is too verbose to serve as a convenient technique reference when analyzing video. Ideally, we could include subtleties and perspectives not present in the book and break with any conventions we find counterproductive.

The Solution.

We write a checklist of common squat errors that can be used to diagnose a busted squat.

A person thinking of posting or responding to a form check should be able to look at the video, run down the checklist, and be able to identify and fix the most common problems.

This isn't a totally new idea. Scott Hambrick put a rudimentary version on reddit here, and Austin Baraki and Alan Thrall have "3 Common Errors" YouTube video. But neither of these are near complete. We can do better by being more detailed and more comprehensive.

I envision our article as composed of two parts: a checklist of common errors (including how to identify them and fix them), and a more theoretical section of frequency asked questions.

Here are some examples to kick things off. I will update this as people make suggestions.

Errors:
Knee slide
Excessive or insufficient depth
Losing thoracic extension
Grip and elbow position
Funky bar path
Bar position
Valgus/Varus Knee

Frequently Asked Questions:
What is butt wink, and what can I do about it?
What's the difference between hip drive and a "good morning squat"?
Do I need shoes and a belt?
Why are people always yelling at me to get tight, what does this mean, and how should I do it?

What now?

Post ideas and talking points for the article in this thread.

Once there is sufficient momentum, I will draft a LaTeX document compiling everything we've talked about. I will post it on a communal platform like ShareLaTeX and you can PM me for access if you want to edit it. However, I'm happy to do all the writing myself if necessary. I am not (yet) a very proficient lifter and so will not be of much help on the technique side, but I can take care of any editorial tasks.

If this goes well, I'd like to start a similar project for bench technique, since there are a lot of useful cues and ideas that aren't in the book. The deadlift is already pretty well covered by the two Baraki/Thrall videos and the SSOC videos, and SS 2nd edition takes care of the strict press (since it's pretty hard to screw up). I have no interest in the "Press 2.0," but maybe someone else does.

[damufunman]

Can we also have snatch and clean & jerk articles?


Mostly kidding.


I think emphasis on stance, the bar position, and bar over midfoot are good starting points, as they set the foundation for the LB squat. Hip and knee position/travel are important but variable based on anthropometry, and aside from the bar position and possibly stance are, kinematically, what differentiate HB from Lab squats. The checkpoints at the bottom most position are also good: knees in line with feet, knee some amount forward/behind toes, hip crease below top of knee, and back angle (which, again, is variable but likelihood more horizontal than many beginners expect) probably gets you to about 90% good squat if you can do with your body what you're trying to do. From there coaching is hugely helpful.

[PatrickDB]

The checkpoints at the bottom most position are also good: knees in line with feet, knee some amount forward/behind toes, hip crease below top of knee, and back angle (which, again, is variable but likelihood more horizontal than many beginners expect) probably gets you to about 90% good squat if you can do with your body what you're trying to do. From there coaching is hugely helpful.

These are all fairly nuanced topics, though! For example, one of the three common errors Baraki talks about is knee position relative to the toes, and here you have to avoid being too far forward/"into the toes" and also holding the shins too vertical. I think he mentions the knees being about an inch in front of the toes as a general guideline, with variation depending on anthropometry.

Depth is also an interesting issue. If I recall correctly, the book advocates using a crease in clothing as a depth marker, which seems ... suboptimal? My depth shouldn't depend on the clothing I'm wearing, it should be a purely anatomical standard. I think the IPF rules are better here.

There was a big thread about this:
https://startingstrength.com/resources/ ... depth.html

Coaching is always hugely helpful, but if you look at most of the form checks on the SS forums I get the distinct impression that someone going through a short checklist could "fix" a lot of them. (There are rarer and more complex questions, for example about certain medical issues or injuries, but we're not trying to deal with those here.)

edit: I just noticed @KDW was laying down some knowledge in that thread. Tagging him in case he wants to discuss further here.

[cwd]

If there's a checklist for video form-checks, make it clear the priority of things.

I.e. as a non-coach who's obsessively studied the books, I spent some time responding to form check threads on StSt, only to realize I wasn't being all that helpful.

It's one thing to spot errors on video, it's another entirely to spot which error is the first in the cascade and should be addressed first.

I.e. if someone's wrists are bent, wearing the wrong shoes, rounding their back, collapsing their knees, and going up on their toes, what do you start with?

[chrisd]

I'd go with Everett' definition of full depth, it still applies to the low bar squat. Full depth is the lowest the lifter can manage WITHOUT SPINAL FLEXION and without surgical alteration. Once your hip movement is maxed out, your only way of getting lower is to

a) allow spinal flexion, thus allowing the pelvis to tilt

or

b) Lower the pelvis, keeping the spine rigid and thus move the bar backwards and fall over

I get bugged by the actions of a certain someone on a certain Facebook page who tells people to cut their depth, despite no visible spinal flexion being present. If flexion is present, keeping a rigid spine will limit the depth, that's how it works.

[mgil]

From coaching a few people in person, “knees out” is the only cue from the book that works consistently, for me.

From what I can recall, a ton of people on SS were breaking at the hips first and then the knees. Knees out works there too. If they think of moving knees out they will usually move them “first” and the hips break at the same time.

I don’t think TUBOW is terribly useful. Almost every case of knee slide I’ve seen is cleaned up with the lifter staying tight at the bottom and driving out of the hole with no change in back angle.

[damufunman]

I'd go with Everett' definition of full depth, it still applies to the low bar squat. Full depth is the lowest the lifter can manage WITHOUT SPINAL FLEXION and without surgical alteration. Once your hip movement is maxed out, your only way of getting lower is to

a) allow spinal flexion, thus allowing the pelvis to tilt

or

b) Lower the pelvis, keeping the spine rigid and thus move the bar backwards and fall over

I get bugged by the actions of a certain someone on a certain Facebook page who tells people to cut their depth, despite no visible spinal flexion being present. If flexion is present, keeping a rigid spine will limit the depth, that's how it works.

I like this. One might ask how do you know you're just above where you get spinal flexion?
Although you could let your knees slide, back angle get more vertical, and drop your hips a bit more like I used to do, without the bar moving down any.

ETA: @ mgil I found TUBOW to be useful, it helped as a cute to stay tight.

[mgil]

@damufunman it works, but to me it’s a better solution to get the lifter used to staying tight and clearing up the problem before introducing the external feedback, since removal of that feedback will need a training period as well.

TUBOW won’t cause cancer, but I don’t think “terribly useful” is correct. “Sometimes Helpful In Training Block Of Wood” is more fitting.

[PatrickDB]

If there's a checklist for video form-checks, make it clear the priority of things.

I think this is a very good suggestion. Will probably have to wait until a later draft, but definitely noted.

From coaching a few people in person, “knees out” is the only cue from the book that works consistently, for me.

Is this useful for people whose knees already track over the toes? Jordan sometimes talks about not liking this for everyone because it can lead to the knee being outside the toes.

I'd go with Everett' definition of full depth, it still applies to the low bar squat. Full depth is the lowest the lifter can manage WITHOUT SPINAL FLEXION and without surgical alteration.

How much spinal flexion is permissible? Zero? I don't think anyone can get near parallel maintaining the same exact spinal position they have while standing, especially if they're like most people and have some anterior pelvic tilt. If you watch Jordan squat, he goes into neutral even cutting depth aggressively.

On the other hand, there's a huge grey area around neutral.

[mgil]

Getting knees outside toes is tough, specifically for dudes.

[damufunman]

@damufunman it works, but to me it’s a better solution to get the lifter used to staying tight and clearing up the problem before introducing the external feedback, since removal of that feedback will need a training period as well.

TUBOW won’t cause cancer, but I don’t think “terribly useful” is correct. “Sometimes Helpful In Training Block Of Wood” is more fitting.

Yeah that's fair. I think it did help to some extent, but at high intensity, high RPE things start to revert to more knees forward, so I get ether you're coming from. BTW, have you couched many people? I get the feeling you'd make a good couch.

[PatrickDB]

Yeah that's fair. I think it did help to some extent, but at high intensity, high RPE things start to revert to more knees forward, so I get ether you're coming from.

I'm a little skeptical about being super strict about knee slide in someone who isn't experiencing pain or shifting balance onto the toes. (Or, to use @mgil's framing, when it doesn't come from another issue upstream.)

Jordan displays a small but noticeable slide on all the squats I've seen of his and is pretty blasé about it in his IG "how to squat" slideshow. Austin and the lifters currently on his IG story also have their knees moving the entire time. I could probably dig up other SSCs who do, too. You're transmitting a ton of force through them, of course they won't be motionless.

[augeleven]

“Sometimes Helpful In Training Block Of Wood” is more fitting.

so many lols.


sometimes helpful in training piece of straight timber?

[OCG]

I don’t think TUBOW is terribly useful. Almost every case of knee slide I’ve seen is cleaned up with the lifter staying tight at the bottom and driving out of the hole with no change in back angle.

The TUBOW is useful for the kind of people that have no idea what they're doing with their knees or ass. But I agree, if they're too loose, divebombing, have a poor rack etc. you fix those things first. But if after that, you've still got significant knee slide then the TUBOW becomes actually useful as a tool for letting people become more aware of their own body.

As far as knee movement goes, the knee is fine to shift a bit forwards at the bottom but it shouldn't bounce or be on a constant trajectory forwards. It should go, top half the knees come into position. Bottom half, the knees mostly stay in place and maybe shift slightly/change shape. By like an inch. In theory, there isn't any reason why the knees could constantly move forwards throughout the squat. In practice, this requires a constantly changing back angle and tends to fuck with bar path. The only person I've ever seen do it well is Mike T.

"Knees should be fixed in the bottom half of the squat" is hyperbole. Like a lot of things. Yes, there will be a bit of shift there, but it should be quite small compared to the rest of the movement. Jordan's squat above is a great example. First half: knees move say 6". Bottom half: Knees move maybe 1" at most.

[mgil]

“Sometimes Helpful In Training Block Of Wood” is more fitting.

so many lols.


sometimes helpful in training piece of straight timber?

This works well too!

@damufunman I’ve coached a few people in person and I used to comment on a lot of form checks back on SS. Having moved into my garage has reduced my time available to continue with in person stuff, which is okay since my kids keep me pretty busy along with work.

[mgil]

I don’t think TUBOW is terribly useful. Almost every case of knee slide I’ve seen is cleaned up with the lifter staying tight at the bottom and driving out of the hole with no change in back angle.

The TUBOW is useful for the kind of people that have no idea what they're doing with their knees or ass. But I agree, if they're too loose, divebombing, have a poor rack etc. you fix those things first. But if after that, you've still got significant knee slide then the TUBOW becomes actually useful as a tool for letting people become more aware of their own body.

As far as knee movement goes, the knee is fine to shift a bit forwards at the bottom but it shouldn't bounce or be on a constant trajectory forwards. It should go, top half the knees come into position. Bottom half, the knees mostly stay in place and maybe shift slightly/change shape. By like an inch. In theory, there isn't any reason why the knees could constantly move forwards throughout the squat. In practice, this requires a constantly changing back angle and tends to fuck with bar path. The only person I've ever seen do it well is Mike T.

"Knees should be fixed in the bottom half of the squat" is hyperbole. Like a lot of things. Yes, there will be a bit of shift there, but it should be quite small compared to the rest of the movement. Jordan's squat above is a great example. First half: knees move say 6". Bottom half: Knees move maybe 1" at most.

I agree with this stuff.

The root cause issues that @PatrickDB mentions are indeed where things go awry. A good coach (and dudes like Rip are usually good at this) can pinpoint the main problem and fix it. On the video Alan Thrall just posted, Rip is doing a nice job of getting good squats out of people on those video clips from the old dvd series.

Which brings me to a recent pet peeve of people being told to look down, or even worse “point their nipples at the floor” when they have a soft thoracic spine. That’s led to some silly spectacles on social media. Since people aren’t normally conscious of the orthogonal vector oriented about the moment of their aereolae, that cue seems to have people fold too much.

The old tennis ball under the chin seemed to be better at getting a neutral cervical spine as well in comparison to the “look down” stuff that seems to be overemphasized.

To spread the love, Starett cues don’t make a lot of sense either and lead to some wonky squats as well. People squatting high because the foot forward stuff leads to impingement and folks rolling over onto the sides of their feet. I’ve seen some awful squats come out of that camp as well.

The recurring theme is forced anatomical positions that don’t work. I think a lot of the good coaches out there look for simple things like absence of knee valgus and control of the spine while squatting. Fixing those two things, however they may be cued, gets most people close enough to a proper squat that they can train safely.

[chrisd]

If there's a checklist for video form-checks, make it clear the priority of things.

I think this is a very good suggestion. Will probably have to wait until a later draft, but definitely noted.

From coaching a few people in person, “knees out” is the only cue from the book that works consistently, for me.

Is this useful for people whose knees already track over the toes? Jordan sometimes talks about not liking this for everyone because it can lead to the knee being outside the toes.

I'd go with Everett' definition of full depth, it still applies to the low bar squat. Full depth is the lowest the lifter can manage WITHOUT SPINAL FLEXION and without surgical alteration.

How much spinal flexion is permissible? Zero? I don't think anyone can get near parallel maintaining the same exact spinal position they have while standing, especially if they're like most people and have some anterior pelvic tilt. If you watch Jordan squat, he goes into neutral even cutting depth aggressively.

On the other hand, there's a huge grey area around neutral.

Twerk practice at the start, marginal depth, twerk at the bottom (ooh err!), GM at the top. that guy should try high bar and take some of the moment arm off his lower back.

[chrisd]

I like this. One might ask how do you know you're just above where you get spinal flexion?
Although you could let your knees slide, back angle get more vertical, and drop your hips a bit more like I used to do, without the bar moving down any.

1) Your spine doesn't flex.

2) Everyone has forward knee movement at the bottom of the squat. I asked about this on the coached Q&A and was roundly ignored. If you watch the demos by any of the SSCs, check knee position against the background and you will see knees travel forward a small amount as the hip passes through parallel and back again as they rise. That's part of the squat, not a part that's talked about, but it's there.

The cue for the TUBOW is "You can rock it, but not knock it over" or something like that.

[PatrickDB]

Here are some things I don't fully understand and are relevant to some examples in my initial post.

Question about bar path

Is it the case that bar path should be vertical?

When I unrack the bar and stand up, the bar is on my shoulders and clearly behind my mid-foot. So the path is roughly angled down and forwards on the descent, then almost straight up in the first part of the ascent, then back to the standing position.

I wonder if it would be more efficient to begin slightly bent over (like Jordan), with the bar over the mid-foot, so the it can move straight up and down.

Question about hip drive

What exactly differentiates hip drive from a "good morning squat"? Obviously if the bar gets forward of the mid-foot, something has gone wrong, but otherwise how do we distinguish a slight normal change in back angle from a dangerous or inefficient one?

For example, in the video of Jordan squatting, I see no problem. (The bar position is straight over the mid-foot and everything else is dictated by segment lengths and where the hamstrings are in the strongest position to push.) But Chris does.

Question about butt wink

I think we all agree we want the lumbar as close to normal anatomical position as possible in the bottom of the squat. Can this actually be improved, beyond cutting depth to the legal minimum and using a heel?

Mobility work is often recommended, but (in people who have been squatting for a while) I've never seen this actually fix butt wink. My current take is that butt wink is mostly a structural/anatomical issue and there's basically nothing that can be done about it except modifying the lifter's mechanics with a heel raise.

Subquestion about heel height: The standard cheap beginner shoe recommendation is the Adidas Powerlifts, with a .6" effective heel raise. I have used these for a while, but will be switching to Adipowers (which will actually fit now, hurrah) with a .75" raise. Is there any benefit to going higher for people of average height? Tom Campitelli floated the idea of average height, long femured people using larger heels to improve the bottom position, and while this seems logical, I've never seen a powerlifter use a heel this large.

[OCG]

1. At the top with a heavy weight, your hips and knees should be locked (ish) and you should bend at the ankles to bring the bar forwards to where it needs to be. With a lighter weight, in particular a weight comparable to your bodyweight or less depending on how strong you are, you'll see what you describe, a reverse "J" bar path. This is because at the top, your balance point isn't close enough to the bar, so in order to be balanced the bar will end up behind midfoot, then as you bend over and your ass comes back, your balance moves back and the bar comes forwards. Then straight down.

Something I don't see discussed enough is, when you use a nice straight bar path, the bar sits "dead" on your back. It doesn't want to roll and fuck up your elbows. Also not having any forwards or backwards momentum, it tends not to push you forwards. It's not just "efficient" is prevents the bar trying to roll or push you around.

2. "Hip drive" can also be cued as "leg drive" or "using your quads". It's simply a cue intended to counteract the chest/bar drive and keep the torso angle fixed out of the bottom. I feel like a proper "chest drive" is severely underrated and under discussed around certain parts, but that's another topic. As to what degree of back angle loss is acceptable, that's a bit more of a squishy thing I think. Perhaps we can take bar speed or smoothness of the lift? If you start learning forwards enough the bar slows, it's bad?