The Squat: Should Your Knees Travel Past the Toes?

Should the knees migrate past the toes when performing a squat? I posted this question on downloadsocial media, and the immediate response by most was “No!”. I expected this answer from most everyone, from novice to advanced lifters. To you, I happily say, you’re wrong! The debate on proper squat mechanics will never die, but I am going to steal a line from Randy B., an athletic training, and performance enhancement peer, who answered my question: “Absolutely, [the knees] should [go past the toes]. Don’t believe urban legends or follow sports med sacred cows!” I couldn’t have said this any better! Randy is spot on. This urban legend could lead to injury. The purpose of this blog is to shed some light on the debate and provide the rationale for proper squat technique.

Early studies demonstrated that squatting with external loads caused undue stress and damage to soft tissue at the knee joint (1). It was demonstrated that powerlifters (lifting 2.5 times their body weight) reached maximum values of 8,000 N (Newtons) at a deep squat (1). This research caused experts to change squat mechanics. Maintaining a vertical shin angle prevents excessive knee flexion, thus limiting the stress placed on the knee joint. This, in turn, would possibly prevent damage to integral knee structures image002such as the menisci, articular cartilage, and ligaments. There you have it, the fear mongers began screaming: “Don’t let your knees go past your toes!”

Proper squat mechanics requires optimal movement at the ankle, knees, hips, and spine during the descent of the squat. When these joints move in unison, forces will be disturbed equally throughout the kinetic chain. The body is an interconnected chain, and restricted movement at one joint will lead to compensatory movement and dysfunction in other joints (6). Restricting knee flexion elicits compensatory and excessive forward lean at the spine. Restricting the knees significantly increases the load placed on the low back and hips. This is NOT a good thing!

Low-back pain and degeneration is one of the most debilitating forms of musculoskeletal injury seen in the adult population, affecting nearly 80% of all adults (3). The annual costs attributable to low-back pain in the United States has been estimated to be greater than $26 billion (4). In addition, 6% to 15% of athletes experience low-back pain in a given year (5, 6).

Squatting with a flexed lumbar spine decreases the moment arm (leverage), increases compressive load posteriorly, and increases the risk of disc herniation! (2) In addition, shear forces significantly increase as lumbar flexion increases from the neutral position.  It is advisable to maintain a neutral spine throughout the squat and avoid excessive flexion or extension and because lumbar forces are increased with lean, it is better to remain as upright as possible at all times (9). This is done by allowing the knees to move forward.

To keep it practical and simple, I created a mock scenario using a generalized equation and a subject squatting 135 lbs. Torque is a measure of rotational force about an axis of rotation.  There are many forms of and equations for torque, but in its most general form, torque (T) equals the product of the vertical force (F), the length of the lever arm connecting the axis to the point of force application (r), and the angle (Sin θ); thus the equation (T=rF Sin θ). Take a look at my simplified version below using the two images and notice the torque values at the knee and low back when the knees are restricted versus unrestricted.

Vertical shin (restricted knee) Parallel lines (unrestricted knees)
Squatting to allow toes go beyond the knees
Base measurements F = 135lbs (600.5 Newtons)
αlb= 78°; αk= 102°
A to B = 2.75 inches (0.07 meters)
B to C= 19 inches (.48 meters)
A to C = 16.25 inches (0.41meters)
F = 135lbs (600.5 Newtons)
αlb= 90°; αk= 90°
A to B = 9.5 inches (0.24 meters)
B to C= 19 inches (0.48 meters)
A to C = 9.5 inches (0.24 meters)
Torque at the knee Τk = r F Sin θ
Τk = 0.07m (600.5N) x Sin 102°
Τk = 43.01 N·m
Τk = r F Sin θ
Τk = 0.24m (600.5N) x Sin 90°
Τk = 144.12 N·m
Torque at the low back Τlb = r F Sin θ
Τlb = 0.41m (600.5N) x Sin 78°
Τlb = 240.82 N·m
Τlb = r F Sin θ
Τlb = 0.24m (600.5N) x Sin 90°
Τlb = 144.12 N·m

To take this further, a study by Fry et al. in 2003 examined the joint kinetics that occur when forward displacement of the knees is restricted versus when such movement is not restricted. This study demonstrated that while restricting forward movement of the knees may minimize stress on the knees, it is likely that forces are inappropriately transferred to the hips and low-back region (see the following table). Thus, appropriate joint loading during the squat may require the knees to move slightly past the toes while the center of gravity remains directly above the midfoot (8).

Restricted knee movement Unrestricted knee movement
Knee torque 117.3 N·m 150.1 N·m
Hip and low back 302.7 N·m 28.2 N·m

The restricted squats also produced more anterior lean of the trunk and shank and a greater internal angle at the knee and ankles. The illustration below provides a more detailed description of the angular changes produced at joints during both variations of barbell squats. These subtle changes in the joint angles significantly alter loads upon those joints.


It makes sense to protect your knees from unnecessary forces, however, your patellar tendon and PCL are more than strong enough to handle the forces. In Olympic powerlifters peak forces at the patellar tendon (8,000 N) and the PCL (2,500 N). These loads are much lower than their max capacity  of  patellar tendon (15,000 N) and the PCL (5,000 N) (9). These structures can handle the forces.

Some take home points:

  1. DO NOT purposely restrict movement of any joint unless there is an injury or condition that requires you to do so.  Allow the body to follow a normal path.
  2. If you have a pre-existing knee injury, such as a PCL tear or patellofemoral pain, then restricting knee flexion is conducive, but do so with caution!
  3. Squat while maintaining a neutral spine and allow the knee to move beyond the toes. When you do this, you distribute forces evenly and reduce the risk of low back failure and maximize muscle coupling.
  4. Quadriceps development is maximized by squatting to parallel, with no additional activity seen at higher flexion angles (10.)
  5. A full-depth squat while allowing forward knee migration, increases muscle activity of the glutes and the posterior chain (9).
  6. The front squat produces significantly lower knee compression and lumbar stress in comparison with back squats, making it a viable alternative for those suffering from various knee and back ailments (8).

Lift safely!



  1. Nagura, T, Dyrby, CO, Alexander, EJ, and Andriacchi, TP. Mechanical loads at the knee joint during deep flexion. J Orthop Res. 20: 881–886, 2002.
  2. Matsumoto, H, Suda, Y, Otani, T, Niki, Y, Seedhom, BB, and Fujikawa, K. Roles of the anterior cruciate ligament and the medial collateral ligament in preventing valgus instability. J Orthop Sci. 6: 28–32, 2001.
  3. Walker BF, Muller R, Grant WD. Low back pain in Australian adults: prevalence and associated disability. J Manipulative Physiol Ther 2004;27:238–244.
  4. Luo X, Pietrobon R, Sun SX, Liu GG, Hey L. Estimates and patterns of direct health care expenditures among individuals with back pain in the United States. Spine 2004;29:79–86.
  5. Nadler SF, Malanga GA, DePrince M, Stitik TP, Feinberg JH. The relationship between lower extremity injury, low back pain, and hip muscle strength in male and female collegiate athletes.Clin J Sport Med 2000;10:89–97.
  6. Nadler SF, Malanga GA, Feinberg JH, Rubanni M, Moley P, Foye P. Functional performance deficits in athletes with previous lower extremity injury. Clin J Sport Med 2002;12:73–78.
  7. Powers CM. The influence of altered lower-extremity kinematics on patellofemoral joint dysfunction: a theoretical perspective. J Orthop Sports Phys Ther 2003;33(11):639–646.
  8. Fry, A.C., J.C. Smith, and B.K. Schilling. Effect of knee position on hip and knee torques during the barbell squat. J. Strength Cond. Res. 17(4):629–633. 2003.
  9. Schoenfeld, BJ. Squatting kinematics and kinetics and their application to exercise  performance. J Strength Cond Res. 24(12): 3497–3506, 2010.
  10. Watkins, J. Structure and Function of the Musculoskeletal System. Champaign, IL: Human Kinetics Publishers, 1999.


17 thoughts on “The Squat: Should Your Knees Travel Past the Toes?

  1. Kim

    Thank you!! I am a personal trainer and I see this statement everywhere! Knees past toes is a functional movement. Try to get up from a chair or walk up stairs without knees passing toes. But… Like icing injuries… No one will take heed.

  2. Joshua Ong

    Hi Josh, thank you for your post. I have loved reading your articles. What is your opinion on box squats with regards to a relatively straight knee?

    1. Joshua Stone Post author

      I wouldn’t do any squat with limited knee flexion if I didn’t have to. If I wanted to target my glutes and hip hinge, sure, but why if it might lead to injury?

  3. Dr. Chad Nowlin

    Great post! No matter how much time goes by I think to many people consider the squat as simple as going up and down with weight but there is so much more.
    Another thing to consider is at what point the knees cross the toes. If the knees are the first things to move in a squat then the PF joint is going to have much more load through it throughout the entire squat range. If the hips are loaded first then it greatly reduces the forces throughout the PF joint. If someone has adequate/full mobility in the hips and ankles then they should be able to perform a full or deeps squat without the knees getting past the toes or minimally so because the ankle allows plenty of external hip rotation. We have to remember the squat does not just happen in the saggital plane. It also has frontal and transverse components that require different aspects of mobility throughout the kinetic chain to be able to achieve a safe squat. If you only look at a squat from the side then it only seems right for the knees to have to move in front of the toes to achieve a certain depth. But if you throw in there some mobility requirements for the hips and ankles in the transverse plane of a squat then you can suddenly have a really good looking squat with knees never crossing the toes. A good example: Kelly Starrett. Even still… if the knees push out over the toes at the bottom of a squat it is of no consequence since full knee flexion is actually the most stable knee position in a squat. I try to teach all my patients how to squat in a way that follows the arthrokinematics of the knee: a posterior roll with anterior glide. If the knees go forwards first it is more of a joint sheer then posterior roll <— very stressful to the PF joint. Knees over toes is not really bad unless it happens early in the squat.

  4. Bridget

    I think lesser – qualified personal trainers can often recognize that their clients are doing something wrong in their squats, but they don’t know what exactly it is. They say “don’t let your knees go over your toes” when I think they mean “keep a neutral spine” or “engage your glutes.” But, I don’t think most people understand how to self correct with that kind of advice. What mistakes do you see most often and how do you get your clients to correct poor technique? Is it a matter of first fixing their disfunction and then good squat technique will naturally come?

    1. Joshua Stone Post author

      I see lots compensation patterns. A the top of the squat I see early hip movement, or early knee movement, instead of simultaneous movement. Mid-squat I will start to see knee adduction and at the bottom of the squat I will see excessive hip hinge, and excessive lordosis. It really depends on the degree of dysfunction. That said, I can see those patterns at any stage during the movement.

      If it is something that the client can self-correct during the movement I will simply use verbal cues or light tough to get them properly aligned. Sometimes this works, and other times it does not. If a client is unable to perform the squat correctly, we will regress. I may take them off the squat until they can demonstrate a proper body weight squat. As load, speed, and volume increase the degree of compensation will also increase. For example, if a compensation exists during a body weight squat then I must do isolation exercises to get the proper muscles engaged. If the client demonstrates poor technique during a body weight squat, that poor technique will become even more pronounced during a loaded squat. A client of mine will not progress to a body weight or loaded squat unless they demonstrate the functional ability to do so. If a client demonstrates a clean body weight squat there is a potential that when we add weight to the squat or speed to the movement the compensation may present itself again. So we progress slowly from easy to hard and at each stage we observe for compensation to occur. If no compensation occurs, we keep going. Ultimately, you want the client to jump, cut, run, lift heavy weight without compensation. Sorry, that is probably a much longer-winded answer than you had expected.

  5. Mike

    I like your concept but and agree with you somewhat. But I believe the proper squat would be knees behind toes. Look at good front squats people have no problem make than happen if they have the mobility. I would say in both pictures you used neither are a good squat. and hip position is different in both and is basically not measurable (in that formula) yet putting more stress on disc. The biggest question I think is we shouldn’t perform this movement without proper stability and/or mobility. This squat is pretty good and from what I can see his knee are almost aligned with his toes.

  6. jon

    Nice article but I am not sure I agree with your beliefs. I have competed in powerlifting for 18 years, grew up in gyms and have been fortunate to train with incredible lifters, bodybuilders, OLY, strongman.
    The article is written by scientist going in many facts and charts to show superior intelligence to persuade the readers but in MOST scenarios, knees over toes will hurt your knees, I know about the Ukrainian lifter with incredible flexibility who goes against this but I have seen loads of athletes (mostly bodybuilders doing hack squats on their toes and others) get knocked out of the game with major patella problems, hinge joints should not be stressed in excessive ROM.
    I just saw an article the other day about how fruits and vegetables are bad for you, so guess you need to come out with a major contradiction to knowledge to get viewers to your site.

    1. Soph

      The difference is that his summary was not “beliefs” they were research findings. Opinions are like …. well we all have one but that doesn’t count as evidence. So when you have vetted, peer reviewed research to support your opinion it becomes evidence, when you have multiple papers to support now it is a preponderance of the evidence and we can say it is supported. Many people don’t understand the difference, this is ultimately the issue and why so many fallacies prevail. Along with the fact that even if you are taught something wrong, as long as it is taught first it is much harder to get someone to change their “belief”. I suppose the moral there is get in early with propaganda, which clearly has already happened on “knees and toes during the squat”. Great article supported by evidence!

  7. Herb

    Thank you very much for this post! Many years ago I videoed MANY people performing the squat and the only subjects who kept their knees behind their toes had very short femurs!!!!

  8. Pingback: Don't Let Your Knees Go Past Your Toes When You Squat

Leave a Reply

Your email address will not be published. Required fields are marked *