What you’re getting yourself into:
-2,200 words. 8-10 minute read time. If you’re in a hurry, you can skip on down to the “Practical Takeaways” and save the dense stuff for later.
1) The origins and insertions of the hamstrings and rectus femoris allow them to extend the hip and knee simultaneously, even though their actions oppose each other.
2) Two joint muscles allow force from single joint muscles to be transmitted to joints they wouldn’t otherwise be able to effect. For example, the rectus femoris allows the glutes to help extend the knee.
3) You can put these principles to work for you by learning the best way to grind through your sticking point on a squat
4) Because two joint muscles transmit force throughout all of your hip and thigh musculature, squats aren’t truly knee or hip dominant, regardless of how they look or what the external torques at the joints are.
Before we get into this post, I want to let you know about our giant How to Squat guide. It covers everything you need to know about every aspect of the squat – from biomechanics to correcting weaknesses to technique. Click here to open it in a new tab so you can check it out after you’ve finished reading this article.
If you guys haven’t figured this out yet, I think about squatting a lot. It was the one lift that took a long time to “click” with me. Usually I can pick up a new movement pretty quickly, but squats were always uncomfortable and awkward for about half a decade once I started doing them. Because of the amount of work it took to master them, they now have a special place in my heart. I don’t think there’s a subject I’ve written more about, and I don’t expect that to change any time soon.
As some of you know, I started writing a book about squatting last summer. I got more than 90 pages into it (before adding pictures/formatting/etc.) before eventually scrapping the project. I scrapped it, mainly because I knew it wouldn’t be overly useful for anyone. 80 of those 90 pages were solely about technique – hand placement, foot placement, bar position, etc. and for almost all of them, the final verdict was “it doesn’t really matter – just do what’s comfortable and makes the most sense of your personal anatomy, as long as it’s in line with your training goals.” Obviously, each point was fleshed out significantly more (i.e. recommendations for who should squat with their elbows down versus back, who may benefit more from high or low bar positions, how to determine your best stance width, etc.), but I eventually recognized that the major message could be summed up in a single statement of ambivalence, and that the whole project was venturing into major neckbeard territory, which isn’t something I’m interested in.
However, it wasn’t all for naught. I brushed up the piece of it that I assumed the most people would be the most interested in – high bar vs. low bar squatting – and the response to it has been overwhelmingly positive. If you haven’t read it yet, I’d suggest you do so, because this article picks up where that one left off, going a bit further down the biomechanical rabbit hole.
Assuming you’ve read my last article and you have a basic understanding of torques, applied to the squat, it’s time to talk about Lombard’s Paradox.
W.P. Lombard was a biomechanist in the early 1900s, and he investigated an interesting phenomenon. When you walk or run or rise from a chair, the quadriceps and the hamstrings contract at the same time to cause movement. Your quads and hammies, however, are antagonists, producing opposing movements (hip extension and knee flexion for the hamstrings, and hip flexion and knee extension for the quads). “Why,” Lombard wondered, “don’t these antagonistic contractions cancel each other out?” Why don’t they just lock your hips and knees in place? Opposing muscles firing together seems like a great way to accomplish absolutely nothing, NOT sprint at high speeds or jump or move heavy loads.
The answer, it turns out, has to do with the different distances from the hip and knee joint that the quads (specifically rectus femoris) and hamstrings originate and insert.
The hamstrings originate on the ischial tuberosity, and insert just below the knee joint, on the back of the tibia. The rectus femoris originates just above on hip on the anterior inferior iliac spine and the rim of the hip socket, and inserts a couple of inches below the knee joint on the tibial tuberosity, via the patellar tendon.
What that means is that when the hamstrings contract, the amount of hip extension torque they produce is considerably greater than the amount of knee flexion torque they produce. The opposite is true for the rectus femoris: It produces much more knee extension torque than hip flexion torque. Those basic facts “solve” Lombard’s paradox. Since the origins and insertions of these muscles mean the hamstrings are much more efficient at producing torque at the hips, and the rectus femoris is much more efficient at producing torque at the knee, when they contract together, you get both knee and hip extension.
Okay, cool. Co-contraction of the quads and hammies doesn’t give you rigor mortise. But what can we actually do with that information?
Here’s where things get cool:
Let me point you in the direction of a paper titled From Rotation to Translation: Constraints on multi-joint movements and the unique action of bi-articular muscles by G.J. van Ingen Schenau (1989).
It was a study trying to figure out how, exactly, humans coordinate muscle contractions to move and transfer force so efficiently. The study itself examined speed skating, jumping, and cycling; however, the authors found the same principles to be in play for all three, and proposed that their findings are generalizable to all movements involving two-joint muscles (such as the hamstrings and rectus femoris) based on some basic physics and geometry. The model they propose makes a lot of otherwise-confusing facts about squatting make a lot more sense. (No, I’m not proposing this captures the full complexity of the situation. Click here for further reading on the usefulness and drawbacks of models).
In their investigation of jumping, they examined the role of the quads in plantar flexion – pushing off the balls of your feet as you leave the ground for a jump. Of course, the quads don’t cross the ankle joint. If you cut all the other muscles away from the human body and contracted the quads, nothing’s going to happen at the ankle. However, force from the quads is transferred to the ankle by the gastrocnemius, which originates just above the knee and inserts on the heel. As the quads extend the knee, the gastroc would stretch slightly if it were relaxed. However, if it remains the same length or contracts, the lengthening that would otherwise occur due to knee extension is instead transferred to the ankle where it aids in forceful plantar flexion.
In fact, previous studies had estimated that only 25% of the plantar flexion force that occurs when jumping is actually attributable to the contraction of the calf muscles. 25% actually comes from the quads, and is transmitted to the ankle via the gastroc, while the other 50% comes from the elastic properties of the Achilles tendon.
From this, we can draw some surprising conclusions, and make some practical recommendations. I’ll start with the surprising conclusions.
1) The glutes are knee extensors.
2) The quads are hip extensors.
Just let that sink in for a while.
Here’s the same model from this study, illustrating how the glutes can cause knee extension via the rectus femoris:
Likewise, here’s an illustration of how the quads (vastus lateralis, intermedius, and medialis) can cause hip extension via the hamstrings:
This concept also makes an interesting piece of a study I’ve discussed before make a lot more sense. In the Bryanton study I discussed in this article, at face value it would appear that squats are quite challenging to your calf muscles, with relative muscular effort (RME) hovering between 65%-80% for a hefty portion of the movement. However, RME was normalized based on the strength of the calf muscles alone; when you throw the assistance from a forceful quad contraction into the mix, it makes sense that your calves aren’t as wrecked after a hard squat workout as your quads and hips are.
1) If you’re prone to turn your squat into a good morning, these findings throw another potential culprit into the mix – your glutes. Since they can contribute to knee extension via the rectus femoris, if you’re unable to produce enough knee extension torque to come up out of the bottom of a squat without your back angle relative to the ground decreasing (hips rising faster than the bar), they could potentially be to blame.
2) This provides even more evidence for the idea that it’s probably NOT your hamstrings that are limiting you. Actually, they are probably the last thing that could potentially fail you. When weights get too heavy and you wind up having knee extension without concomitant hip extension, what that’s actually doing is lengthening your hamstrings so they can increase their independent contribution to hip extension torque, rather than allowing them to function primarily to transfer force between segments.
In my previous article looking at the effects of wearing a belt on squatting performance, one of the things that jumped out at me was the effects of fatigue on forward lean in the squat. As you get more and more tired, you start leaning farther and farther forward. Based on this model, the most reasonable explanation is that the prime movers (the quads and glutes) start fatiguing as the set wears on, so the hamstrings, which are still relatively fresh, are put in a position where they can independently do more to aid in hip extension. Getting your hamstrings more involved in the movement should not be a priority – their increasing involvement means your quads and glutes are failing you.
3) Lombard’s paradox explains beautifully the proper way to grind a squat.
Below is a video of my friend John Phung, who may have this technique more dialed in than anyone I know. Watch as he comes out of the hole, and bar speed slows almost to zero. In that position, co-contraction of the hamstrings and rectus femoris can reposition your body under the bar so you can finish the lift. When you fail a squat above parallel, the culprit is pretty straightforward: The amount of hip flexion torque the bar is exerting on you exceeds the amount of hip extension torque you’re capable of producing.
By driving your traps back into the bar (to provoke a stronger contraction from your hamstrings, similar to a good morning), and squeezing your glutes to drive your knees forward and out (transferred down your rectus femoris, resulting in increased knee extension torque to keep your knees from flexing as more stress is shifted to your quads), you shift your hips forward slightly. That shortens the moment arm they’re working again, and voila! Same weight + shorter moment arm = less hip extension torque necessary to get through your sticking point and finish the movement.
Before he knew this trick, this is a lift that would have pinned him. The key to grinding a lift is to keep yourself from getting stuck with your butt a mile behind the bar, and to fight to get your hips forward, even if the bar remains motionless for a moment – Lombard’s paradox lets you do that.
4) “Knee dominant” and “hip dominant” squats are misnomers
This concept also helps us make sense of a study I’ve linked in the past that has been met with disbelief (and, admittedly, I didn’t really understand it either). Muscle activation in the back squat and front squat is just about identical. Is there a longer moment arm for the hips in a back squat? Of course. However, although a squat may look more “hip dominant,” the quads are still aiding in hip extension via the hamstrings. Conversely, although a squat may look more “knee dominant,” the glutes are still aiding in knee extension via the rectus femoris. If the external torque is less at one joint, the prime movers at that joint are “freed up” to help overcome the increased external torque at the other joint.
Now, Westside-style box squats may truly be hip dominant, and something like this is probably knee dominant in a really substantial way:
But just about any other form of squat is going to train your hip and thigh musculature in pretty much the exact same way. (A quick caveat – a wider stance may train your glutes a little harder, but other than that, the only major variable that changes how hard your muscles are working is depth.)
So what do squats train? Is it the quads, or the hamstrings, or the glutes? All of the above to a point, but none of the above specifically. The squat trains the squat. For training purposes, just find where the bar is most comfortable, squat as deep as you safely can, and repeat. The whole system of muscles that extends your hips and knees is tied together by your hamstrings and recti femori, which make sure the work is distributed over all the muscles involved.