Chad Wesley Smith (from jtsstrength.com) visited Kelly Starrett and made a little interview with him.
Recently Kelly has been receiving a lot of flak about the “Knees Out” cue for squats.
In the video he clarifies what he means and makes a lot of good points.
Just like you shouldn’t let your elbows flare in the bench press or let your back round in the deadlift, your knees coming in is not optimal.
Knees out is just a cue. Think Knees Not In.
For a more in depth discussion check out Kelly’s 3 part series about the topic here.
Let’s use the mobilizations as skill transfer exercises to do the most important thing.
Update 23.02.2014: A discussion between Kelly Starrett, Lon Kilgore, Quinn Henoch and Jacob Tsypkin on the The “Knees Out” Cue.
Everett says
The knees out cue is misleading, as is the idea, “Just
like you shouldn’t let your elbows flare in the bench press or let your back
round in the deadlift, your knees coming in is not optimal.” We know that
many of the best lifters disobey all of these commands.
1. Rounded back: Some pull quite well with a rounded back, but I’d agree with
most that this is the exception to the rule.
2. Elbows out or in? This is more just a matter of personal preference. Some
bench quite well with elbows out like a bodybuilding. (In fact, the elbows-in cue
is probably partly the result of a lot of material written when shirted
benching was on the rise.) I like too keep my elbows in, but look at a lifter
like James Henderson–one of the best benchers of all time, and he flares his
elbows way, way out–almost at a right angle to his body.
3. Knees out: what most coaches probably mean by this is not to let the knees
buckle inwards. One’s knees buckling is bad. There are, however, many squatters
who lift quite well without any exaggerated pushing out of the knees. The
problem with this cue is it is the sort of thing that random people will yell
in gyms across the US during a squat rep. They heard it once, but they don’t
understand it. To be honest, when I got serious about my squat, I secretly
decided instead of thinking “knees out, sit back” to think “knees in, sit
forward.” I took my squat from around 195 kgs to 282.5 kgs doing this, so I
think it worked. Jesse Norris squats very similar to the way I do—neither of us
have optimal leverages for squat, so traditional cues may not be so useful. I’ve
attached a video of my squat so you can see the technique in action. For
beginners, you probably want to emphasize “knees out” because their knees are
prone to buckling. Once a lifter is more experienced, however, this cue is not
as universally applicable.
http://www.youtube.com/watch?v=BxtE79T-Wvo
JJ says
all good points.. Kstarr uses the term athletes to be encouraging (IMHO) to most individuals in the sense that “everyone in some way is an athlete.” But the Athlete’s we see here on ATG, are a breed of their own. Making up less that a small percent of the general population, they can sort of “do what they want, if it works for them.” I agree with many of Kstarrs teachings and also with Everett statements… it really comes down to efficiencies and preference’s.
Generally speaking, internal rotation is the death of a joint, James Henderson does a fabulous job slacking that “theory” in the face. Most science is usually generated from a cumulation of research… but like django, in the movie… there is 1/10,000 of an exception.
Great rebuttal to the current topic under fire…
-JJ M.S in Human Movement
Jake W says
re #3, where you say, “what most coaches probably mean by this is not to let the knees buckle inwards,” he adresses this quite directly; that the only way to cue this, if a lifter’s knees are in fact buckling, is not to say, “knees not in” or “knees neutral;” the cue (just a cue!!) is “knees out.”
i take exception with his blanket advice to keep your toes forward, even if you’re flexible enough to pull this off.
Chris Theoharis says
yep the toes forward thing is absolutely the issue. toes forward and pushing knees out seems to be the primary cause of the patellofemoral pain cited by Takano’s chiro.
Everett says
This is exactly my point. I haven’t read Starrett’s book, and I don’t particularly plan on it, so I’m not really disagreeing with him–just saying that cue isn’t always helpful. I’m sure he’s likely careful to use it in appropriate contexts. The problem, however, is that when you go to any CFit, or strength-oriented gym in the U.S., you’ll hear “knees out!” being yelled during every squat.
As a rule, if you squat with your toes pointing forward, you should keep your knees in more. Toes pointing out will help you keep your knees out. I’m sure there are studies that cite this, but that’s not really the point–anyone who has squatted for a considerable amount of time knows that angling the knees out while keeping the toes forward is a recipe for knee pain.
Everett says
It should read: “I’m sure there are studies that support this..”
Dan says
The nature of weightlifting discussion on the internet, leads to examination of minutia in movement. Details ARE IMPORTANT. However, details without direction lead to misunderstanding. The book is especially confusing, as it is written for a certain audience, imo, someone not particularly experienced with interest in Crossfit movements. So, hip external rotation is important. The problem is people don’t look at the body as a system. (with individual variances at times, but that is for someone with more experience and knowledge of biomechanics, and evidence of variance to determine.)
I’ve said this in reddit discussions, but the “arguments” over right or wrong in Kelly’s teaching are really cyclical and inane. Kelly never says ‘external rotation of the hip’ means MAXIMUM external rotation of the hip. It may be maximum ER for someone with very little ER, but for all athletes, it does not mean flat out max ER. Varus is not better than Valgus and vice versa. Both “parties” are saying seemingly conflictin things, but I think they don’t realize that they share intent. Semantics and diction are the issue. To me, the squat is most easily analyzed kinematically. This can only return answers at a very basic level, however, so biomechanical knowledge of the body is necessary to correct individual problems, imbalances, etc.
As a coach one must tell athlete’s what to do, in a way that they understand. I have full belief that coaches like Kelly understand why, but they never explain it in that way, likely because the general beginning athlete will not understand increased ACL forces during flexion at certain angles, and how a collapsed arch leads to that.
Chris Theoharis says
pg. 85 or 86 of supple leopard, the hips are clearly externally rotated to the point where the knees are no longer over the plane of the foot. hip external rotation is emphasized. KStar is an advocate of toes forward to create torque during the squat but for many body types this seems to lead to PF pain. this is the main issue. Kelly is still a great coach, resource, DPT.
Dan says
Agree totally with your statement. I was going to mention the discrepancies in the photos but chalked it off to just a bad photo. Not sure why they went with such a case in a book that would have been read by the general population, likely to want to copy it.
Dan says
whoops, forgot to add in another thought I had. From the 2 athletes I work with that have experienced PF pain in some form from these types of cues, it is mostly due to hip rotation coming from trying to push the knees out, exclusively. Once you learn how to internally and externally rotate your hips without that cue, things have seemed to clear up.
I think you mentioned down there that lack of hip IR is part of it. I think this can be a major issue if someone has been brought up on the internet, reading external external external. Everyone beats inward knees like a dead horse, but no one points to the studies showing that in healthy athletes, internal rotation of the femur happens during knee extension and external during knee flexion.
chris says
with all its flaws, these discussions would be a bit easier for everyone if all had read their good ol “starting strength”. from knowledge about biomechanics to simple distinctions between “cue” and “actual desirable form” (see the “arched back” example by rippetoe).
to the point: for the GENERAL population without any pecualiarities or high level special variations, the most sensible way is to have the femurs aligned with the foot. if that means you have to push your knees out, the cue is “knees out”, SO THAT IT PRODUCES ALIGNED, NOT MAXIMALLY EXTERNALLY ROTATED FEMURS (you would fall anyway if u achieved that). if u happen to have your knees to far out, the cue would be “knees in”! same btw with “arched back”: u dont wanna arch your back maximally, but you want to fight the (contrarian) tendency of a kyphosis (“rounding”) of the lower back. your goal is to have your natural lordosis as that creates the least pressure on the discs.
however, the recommendaten to point feet forward when the femurs are not is indeed debatable, as then there would be no alignment of these structures creating unwanted forces (you guys already mentioned patella problems).
in every sport, cues often exaggerate things by stating the opposite of the current state to reach the desired goal.
ps sorry for the shouting, i didnt know how to implement italics 🙂
Gregor says
Thanks for the comment. Italics are set with the
html tag.
More allowed tags here http://help.disqus.com/customer/portal/articles/466253-what-html-tags-are-allowed-within-comments
Leon says
I think you hit the nail on the head and Rippetoe argues very convincingly in Starting Strength. . K. Starrett argues for toes pointing directly forward while having the knees far out. So far indeed that they are not anymore above the feet. Personally, I tried this approach two times over the course of several weeks. And it hurts. Everytime I get knee problems (area around the patella). It feels like the knees get twisted. And it doesn’t feel like a strong position. Quite the opposite actually.
A second point is that he argues the knees should not go forward (at least until the very bottom position of the squat): I tried to follow that, but again it leads to trouble. What happens is that I fight to keep the shin vertical and at the very bottom the knees shoot forward. So the squat becomes two movements. Very awkward indeed. Various other sources – among them C: Poliquin – argue that the knees
should move naturally. That would keep them healthy and smooth, also for daily activities. And it’s true. In almost every sport movement except the box squats knees move forward all the time.
In short: When I squat “naturally” my knees feel great. If I try to follow Starrett’s advice I get problems.
Chris Theoharis says
Starrett has a few different exercises in which he demonstrates the strength of the feet forward, externally rotated position (Diane Fu at the end of 3-part series). I personally know many folks who squat with that style and have had no PF pain for years. I also know folks (including myself) who have reported PF pain or tendonitis as a potential result. In my case, I squatted for 6+ months with feet forward without having any pain feedback.
Part of the argument has to do with whether Starrett’s points about creating torque through the joint truly lead to better, stronger squats. There is definitely a stylistic difference and Starrett has a lot of experiential evidence as well as a good biomechanical argument. Takano even admits at the end of his series on “knees out” that there needs to be more research before we can abandon traditional squat gospel of knees tracking over plane of the foot, but he does allow for that possibility.
This is a more complex topic than thumbing through Rippetoe’s analysis of the squat, nor is he a biomechanical giant among men.
Also, what is “natural”? To the point about knees forward, I don’t think you’re following Starrett’s cues as he would coach them. All coaches agree that you want to load the hips, not the knees, and attempting to keep the shins vertical is one strategy to ensure your athletes are just doing that. It doesn’t mean the knees don’t travel forward, it means the hips need to be loaded first.
chris says
hm, i dont understand your hip vs knee statement and what the alignment
in one plane has to do with the other (out/in vs vertical/tilted
forward)? maybe i misunderstood starett´s remarks then. im fearing a big misunderstanding here ;). im not a native english speaker.
i was talking about the “in/out”/ femur vs foot alignment. not about verticality of the shins/kness tracking beyond toes. my understanding is that he wants to create an artificial torque by pointing the feet forward when the femurs as a whole are pointed outwards (and are externally rotated by pushing out/knee not in) to create a stable base. atm im more convinced that this artificial torque might provide more stability, but also a higher risk of injury or chronic stress on the structures.
the forward tilt of the shins depends on various factors, including squat variation (high bar/low bar) and body proportions. so dependent on goals, individuality and preferences, they tilt forward in different degrees and hip torque vs knee torque differs subsequently (with all their disadvantages and advantages). easily illustrated with the typical (geared) powerlifter vs oly lifter type of squat.
i agree that as long as performance is concerned, there are variations on how to squat efficiently. see bret contreras´ recent blog post about elite level squats http://bretcontreras.com/the-keys-to-stronger-squats/ .
i also agree that rippetoe is not the last answer. (however, his analysis of the squat would provide many with far more insight than they currently have). starrett and rippetoe (and his (former) cos kilgore and pendlay) as most coaches in the strength training business have equally little or no formal education in biomechanics nor do they conduct lab testing and scientific research in general but present their theories on logical reasoning and anecdotal evidence. again i agree that not until some reasearchers – like mcgill did with back biomechanics – get some grants on “squatting variations and torque/joint health/performance” can we have more solid data and evidence based answers on this.
Leon says
I read the book and he clearly says and shows that the knees do not travel for the most part of the movement. The knees come only forward at the very bottom. And this against what many coaches say and also how kids squat (as an example of “natural”).
Actually he even states that this point alone is worth the whole book. So he does emphasize it. And apparently his squatting works for him very well as he has very good numbers.
However, I find Rippetoe’s argument more convincing. I think in the end one has to have a compromise between maximal stability in the feet or a more easy to assume knee position.
Another thing that confused me is his guide to asssume a correct torso position. This actually leads to a posterior pelvic tilt. A good position for gymnastics, but not for lifting weights off the ground. When I try I either get back pain or simply cannot lift much at all.
Tyler Satnick says
To me this entire situation is hysterical. All Kelly has been trying to say the whole time in terms if squatting is, that he has experienced, I’n general, feet are turned out MORE then they should be for any given stance and that shins are not vertical ENOUGH, and that hips are opened ENOUGH and that these things are work abounds for peoples lack of mobility and proper stabilization strategies. People can’t seem to see that these cues are directions toward a neutral style of squatting throughout the whole hip system. If your stance is narrow you should be turned out 5-12degrees, if it’s shoulder with it should be 12-15 and for sumo stances up to 30 degrees, turning the lower leg out more than the hip can match is generally untenable, because the hip system does gain stability from rotation. In terms of tracking, the knee should not track the foot (the arch of your foot will collapse in this position which is an obvious tell). The knees out cue is a hedge cue designed to take a person from buckled to neutral, period. People experiencing pain from this and straighter feet are most likely in a wide-ish stance who are not keeping the shins vertical enough, going too deep, or have feet too straight, in a acceptable stance with or going overboard by letting the foot roll to the outside and driving the knee out way beyond neutral. I swear this whole topic reminds me so much of the “arch” concept that used to be (and sadly still is) controversial. Any elite athlete or lifter that has variances in their squat technique is NOT evidence or reason to adopt it as the basic human model of squatting.
Tyler Satnick says
To me this entire situation is hysterical. All Kelly has been trying to say the whole time in terms if squatting is, that he has experienced, I’n general, feet are turned out MORE then they should be for any given stance and that shins are not vertical ENOUGH, and that hips are opened ENOUGH and that these things are work abounds for peoples lack of mobility and proper stabilization strategies. People can’t seem to see that these cues are directions toward a neutral style of squatting throughout the whole hip system. If your stance is narrow you should be turned out 5-12degrees, if it’s shoulder with it should be 12-15 and for sumo stances up to 30 degrees, turning the lower leg out more than the hip can match is generally untenable, because the hip system does gain stability from rotation. In terms of tracking, the knee should not track the foot (the arch of your foot will collapse in this position which is an obvious tell). The knees out cue is a hedge cue designed to take a person from buckled to neutral, period. People experiencing pain from this and straighter feet are most likely in a wide-ish stance who are not keeping the shins vertical enough, going too deep, or have feet too straight, OR are in a acceptable stance and going overboard by letting the foot roll to the outside and driving the knee out way beyond neutral. I swear this whole topic reminds me so much of the “arch” concept that used to be (and sadly still is) controversial. Any elite athlete or lifter that has variances in their squat technique is NOT evidence or reason to adopt it as the basic human model of squatting.
Afferbeck says
A lot of the problem here lies with people taking cues for specific issues and blowing them out into general rules for all to follow. Like Kelly said in the video, the ‘knees out’ advice is usually a good one for beginners, as knees collapsing while squatting is a common fault due to weakness, instability, and just plain doing a new movement. This advice then gets watered down through the internet until it turns into some blog article called ‘5 Tips For Learning How To Squat’ where it simply says ‘1. Force the knees out.’ Then beginners find this advice and apply it, thinking they are doing the right thing. Then a couple of years on wonder why their knees are hurting, ‘but I’ve been forcing my knees out!’.
Not all lifters are the same; while collapsing knees is a common fault, it also goes the other way, with people like myself having plenty of external rotation and zero internal rotation. So the ‘knees out’ advice was welcome because that was very natural for me to do. But due to lack of hip IR, which is needed in the bottom of a deep squat (unless the feet are under or inside the hips), the twisting force is sent to the tibia and knee instead. And then forcing the knees out EVEN MORE results in further twisting further down. Cut to a couple years later and hello knee pain.
And then we have people who have come from more specific sport backgrounds with coaches and who have not been learning everything from websites aimed at beginners. They hear ‘knees out’ is some kind of golden rule which they’ve never known about. And when they give some advice to the contrary, that knees neutral/in can be good, the internet goes nuclear because The Gospel has been questioned. And here we are.
Chris Theoharis says
^^this. your second paragraph basically describes the situation i found myself in.
Struan says
When did squatting get so complicated? Just put your feet at shoulder width, keep your lumbar tight, and sit on/between your feet, then stand up.
william says
Thanks for more information about knee pain . http://www.paincenterny.com/knee-pain/
buttermancan says
I have personally followed some of Kelly’s methods. He is great for a lot of things but he does seem to subscribe an excessive knees out technique. If you look at his photos in his book….with feet pointing forwards the knees are crazily out over the feet. I tried this for some time and it lead to knee disorders. I love loads if his stuff and I’m no expert but I just don’t buy it and think it an extreme exaggeration