90 Degree & Parallel vs. ATG Squats
- Part 1 -

By Joel Seedman, PhD



Comparing ass-to-grass (ATG) and 90 degree or parallel squats is one of the more heavily debated issues in the strength and conditioning industry.  Having employed both techniques in my own training and that of my clients as well as researching the neurophysiology and biomechanics of each, I can tell you that squats at 90 degrees or parallel beat ATG squats in nearly all circumstances no matter the individual differences or training goals (read more here).  Although I’m likely to receive life threatening emails and hate comments from that statement alone, the following points will hopefully extinguish some of this animosity so long as the reader is level-headed enough to absorb the information.

Note:  When examining joint angles in relationship to each other, a parallel squat is actually slightly below 90 degrees (approximately 110 degrees) and a 90 degree squat ends up being slightly above parallel. See Figure 1 of this study here for a visual illustration.  With that said, a proper squat is somewhere between 90 degrees and parallel. This represents the squat depth parameters that will be referenced in this article.  

“Go Deep or Go Home” Mentality

When I first began training I would have said the most common technique error I witnessed in gym settings was squatting too shallow and failing to reach proper depth.  Fast-forward over a decade later and I would still agree that improper squat depth is the most common issue.  However, it’s now on the opposite end of the spectrum – too deep. I’m by no means advocating half squats or partial movements. Rather I’m suggesting the industry standards for what is considered proper depth has gone from one extreme to the other. In one regard this makes perfect sense.  Before the surge in online instruction and YouTube videos, half squats and partial reps were quickly becoming commonplace.  Rather than providing the appropriate level of instruction, the quick fix for the industry was to suggest rock bottom or “ass-to-grass” squats.  Soon, anyone performing shallow squats would be called-out by your typical half-educated lifter letting him know their squats were nothing short of iron game heresy. Even worse, posting a YouTube video with anything but “ass to-grass” squat technique would be met with a flurry of hate comments and dislikes with enough animosity to warrant excommunication from the strength industry altogether. 

As a result many lifters began going to any lengths necessary to increase their range of motion, flexibility, and mobility, with the end goal of reaching as deep a squat as possible even if it meant sacrificing proper mechanics.  Although you’re still likely to run into your average newbie performing half squats and quarter squats, the majority of dedicated lifters now squat with excessive depth by allowing their bodies to collapse at the bottom of the movement. 

Just Because You Can, Doesn’t Mean You Should

A key factor that’s led to the promotion of faulty squat mechanics (i.e. excessive squat depth) is mobility assessments demonstrating differences in human anthropometry particularly anatomical variations of the hip joint.  These individual differences have led to the assumption that each person has his or her entirely unique set of protocols when it comes to ideal squat depth and mechanics.  Unfortunately this notion is completely flawed and inaccurate.

After spending well over a decade coaching hundreds of athletes of all different shapes, heights, ages, and sizes the one thing I can tell you is that while maximal range of motion and mobility boundaries vary greatly from person to person, proper squat depth, mechanics, and ideal range of motion are very similar from individual to individual.  In fact with proper training and coaching a 5-foot female gymnast and a 7-foot male basketball player will have remarkably similar squats. In essence individual differences in anthropometrics only indicates maximal range of motion, not ideal or proper range of motion. Individual differences manifest themselves primarily when the squat is performed incorrectly as there are endless variations of what faulty squat mechanics can look like. In contrast, proper mechanics on any movement including squats, presses, pulls, hinges, lunges, etc. produce a very similar biomechanical outcome from person to person. Read more about the topic of individual differences and squat form here.

On a similar note, just because an individual can squat to extreme depth with no apparent aberration in technique or spinal alignment (i.e. butt wink) does not mean this is their ideal squatting depth.  This only indicates what their maximum depth is.   In fact these same individuals typically exhibit significant laxity in their hips and hypermobility throughout their body both of which can be highly problematic.

Learning from Mistakes

It’s important to highlight the fact that for well over five years I had many clients squatting with “ass-to-grass” depth.  These were done under strict control using technique that most would have considered textbook.  However after seeing the negative ramifications, joint issues, constant tightness, continual soreness, heightened inflammation, and altered movement mechanics (including gait deterioration) I began to realize this was not optimal.  

In fact I soon understood why foam rolling and soft tissue work were quickly becoming so popular. If the lifter was going to incorporate excessive range of motion then soft tissue work was essential.  However proper mechanics warrant no such treatment as the movements themselves provide the very benefits that others seek to gain from soft tissue modalities.

It was these very issues that made me realize that while squatting to such extreme depths was visually appealing to the eyes and ego (for ATG fanatics), the benefits did not outweigh the physical ramifications. 

Don’t Squat Like a Baby

Many in the industry suggest the squatting technique of babies is the gold standard for squat mechanics.  In reality this assumption could not be more flawed for several reasons.  First, babies' mechanics represent the complete opposite of proper motor control and ideal muscle function as their movement and physiques could not be more crude and underdeveloped.  These miniature humans barely have the stability, motor control, or strength to balance on two legs yet we’re looking to them as the epitome of optimal muscle function and mobility.  This could not represent more illogical reasoning and backwards thinking.  

In fact the amount of core stabilization, intramuscular coordination, motor control, muscle activation, spinal integrity, and overall stability of a baby’s physique is practically non-existent when compared to that of a high functioning adult. If a baby’s movement represents the epitome of optimal muscle function then perhaps having the ability to suck on your toes should be the new standard for measuring hip mobility.

It should also be noted that babies actually have more bones in their bodies than adults, as many of their bones simply haven’t fused.  This also leads to exaggerated flexibility and excessive range of motion throughout their body which adults should by no means be attempting to replicate.  Renowned strength coach Nick Tumminello has also pointed this out in his writings emphasizing how ludicrous it is for the fitness industry to use a baby’s squat as the gold standard for movement.

Passive Squat vs. Active Squat

Many who assert the ideology of using a baby’s squat to represent the ideal squat pattern are quick to point to the “third world squat” often performed in other countries particularly throughout Asia.  In a “third world squat” the individual simply relaxes into the deepest squat position where their butt is nearly touching the floor.  Because this is almost identical to what babies do it’s assumed that society and technology are responsible for ruining people’s squat mechanics.  Now here’s where things get interesting.

The third-world squat should in fact be a position that most humans are capable of performing.  And yes, various factors in our society have contributed to the elimination of this ability in many adults.  Here’s the catch; the third-world squat and a strength-training squat are two entirely different movements requiring completely different recruitment patterns.  In fact the third-world squat is a passive squat where little if any muscle activation is evident as the individual simply hangs out on their joints, tendons, ligaments, and connective tissue.   In other words, the individual collapses allowing gravity to pull them down into the bottom position where they can simply use their body’s structure as a chair of sorts.

In contrast, a strength-training squat is an active squat. In this case the lifter should be firing their muscles aggressively in order to maintain stability, motor control, force, and muscle stiffness all of which are essential for taking strain off the joints and using the muscles as shock absorbers.  When a passive squat is incorporated into strength training scenarios with heavy loads, the muscles are in an overly relaxed and overly lengthened state particularly on the eccentric motion thereby stressing the joints and connective tissue rather than the muscles.  Instead of using the reciprocal muscle groups to pull the lifter into the proper position via high levels of co-contraction, the individual relaxes/collapses to varying degrees and relies on both gravity and the external load to pull them into the rock bottom position. 

In such a scenario the lifter is exhibiting low levels of proprioception and muscle activation as muscle spindle recruitment is predicated on increased muscle stiffness and co-contraction, both of which are absent during the passive squat.  This would suggest that not only is a significant amount of tension being transferred from the muscles to the joints, but the ability to fine-tune movement via proprioceptive-related feedback is limited from lack of muscle spindle recruitment.  As a result the squat pattern resembles a very sloppy and uncontrolled movement rather than a tight and crisp motion.

Gain Stability First

Performing mobility work to become more mobile seems logical however this can be the very issue that limits mobility.   In fact, overdoing it on mobility exercises, stretching, and soft tissue work can desensitize muscle spindles allowing the lifter to perform movements such as squats with excessive ROM. This leads to localized chronic inflammation which over time is the very thing that limits mobility and range of motion.

The single biggest problem with a majority of lifter and their squat pattern is not mobility but lack of stability, tightness, and motor control.  As the lifter gains stability their body naturally begins to perform the movement pattern with the ideal range of motion.   In other words, gain stability first and optimal mobility natural follows, not the other way around.  The last thing you want to do is gain ROM that you cannot stabilize.

Finding the Perfect Balance

The popular ideology of gaining more and more mobility in an indefinite and unlimited fashion is completely contrary to the laws of human movement.  Instead the goal should be to find optimal mobility and maintain it.  

Furthermore all movements have both a maximal ROM and an optimal ROM.  Rarely do the two coincide.  The same is true of any athletic skill or basic movement such as punching, sprinting, throwing, kicking, hitting, etc. Each has an optimal range of motion and the goal is to find the perfect balance between overly compact motion and excessive ROM. Read more about finding your squat form with eccentric isometrics here.

Science-Based Conceptual Congruency

When determining proper arthrokinematics and positioning for any movement it’s critical to examine the scientific principles foundational to neurophysiology, skeletal muscle physiology, motor learning, and biomechanics. If the mechanics for a movement such as a squat are correct, then these concepts will not only be practically displayed within the muscular action itself but these principles will be congruent with each other showing no signs of contradiction.  The following sections will examine some of these principles to illustrate how they align and affirm how a squat parameter of 90 degrees to parallel represent the optimal squat depth. To read more about each of these topics as well as the hundreds to studies to support this check out my book MOVEMENT REDEFINED.

The Stretch Reflex Redefined

Many lifters are quick to justify their excessive squat depth by suggesting they’re taking advantage of the stretch reflex.  However this argument is inherently flawed.  In fact what many consider to be effective utilization of the stretch reflex is actually not the stretch reflex at all.  Rather it’s a rebound effect that’s a byproduct of using their tendons, ligaments and connective tissue as flimsy and fragile springboards to bounce off of. This has little to do with the stretch reflex and is in fact diametrically opposed to how you would ideally want to activate the stretch reflex mechanism. 

To optimally employ the stretch reflex, a heightened level of structural tightness and musculoskeletal stiffness must be present as this is fundamental to how muscle spindles operate. And as we know, muscle spindles are the key players when it comes to activation of the stretch reflex mechanism. When muscles demonstrate minimal stiffness qualities such as that commonly seen with excessively deep squats, this disengages or inhibits muscle spindle activation therefor minimizing the involvement of the stretch reflex mechanism. 

Furthermore, reduced levels of muscular stiffness have been shown to decrease proprioception and motor control leading to further degradation of mechanics and reduced kinesthetic awareness.  Such a scenario is commonly responsible for muscle spindle desensitization in which case the body has learned to disengage its proprioceptive mechanisms when presented with heavy loads and high tension.  Rather than producing increased sensory feedback, the muscle spindles gradually become inhibited and blunted from excessive stretch and exaggerated ROM. 

In essence the individual has regrettably achieved the ability to override their body’s natural protective barriers and force-production mechanisms that would normally resist an exaggerated stretch. As a result the body’s ability to create optimal joint angles, fine-tune position, and make subtle adjustments to technique and mechanics becomes greatly compromised as the proprioceptive mechanisms are not functioning as they should.  Unfortunately this becomes the individual’s natural state of kinesthesia not only during squats but also during all related lower body movements and physical activities.

The Transfer Effect

It would be one thing if these effects were only isolated to the strength training session itself. However, based on principles of motor learning we know that movement transfers and impacts other related motions.   Therefore performing movements or in this case squats with excessive ROM not only negatively impacts the body during the actual training session but the effects are long lasting with detrimental carryover to other activities including running, jumping, kicking, lunging, hinging, and even walking. In essence the trainee has improperly re-programmed the neuromuscular system to consistently operate under conditions of structurally compromised positions both from a force production and injury standpoint.  This unwanted-transfer effect is applicable to the following points as well.

Length-Tension Relationship

Optimizing muscle length is a crucial component when discussing parameters of muscle function. The force-length or length-tension relationship is a physiological principle that points out how different positions or lengths of sarcomeres/muscle fibers will produce different amounts of muscular tension and force.  If a muscle contracts with too much myofilament overlap (overly shortened position) or not enough overlap (overly stretched position), force generating capabilities will be impaired as cross-bridge cycling will be compromised. 

Besides hindering muscle activation and reducing motor unit recruitment there is also diminished intramuscular tension, which ultimately curtails the strength and hypertrophy-inducing stimulus of the activity.  Researchers have concluded that a muscle will produce the most tension and force when sarcomeres are in the moderately stretched position.  This would indicate that some stretch is good but too much can lead to sub-maximal results in terms of muscular tension and cross-bridging.  For the squat, this represents an ideal range that is somewhere between 90 degrees and parallel.

Anatomical Levers

A key concept in biomechanics is utilizing angles and positions that maximize leverage and torque production.  This involves 90-degree angles, perpendicular positions, and parallel joint segments.  The ATG squat could not represent a more biomechanically dysfunctional position when it comes to maximizing anatomical lever arms.  In contrast the 90 degree or parallel squat takes full advantage of these constructs by optimizing moment arms and leverage. 

Elastic Energy

These theories are not only supported by neurophysiological principles but also by biomechanical fundamentals dealing with elastic energy.  Simply, if an object or muscle is too elastic then there will be too much deformation.  When the rate of deformation is too high not only is the muscle stretched beyond its natural length but too much energy is used to restore the muscle to its original position.  As a result much of the energy goes back to re-formation of the structure thereby compromising force production.

The biomechanical concepts relating to muscle stiffness, dictates that there must be a compromise as to how much rate deformation will occur when it comes to maximizing performance as well as safety.  This compromise favors increased levels of stiffness as most biomechanists and physicists concur that individuals should create the stiffest usable condition of an object/muscle for optimal strain energy.  This strain energy will ultimately produce the greatest benefits for force production, power, and movement as well as safety and technique. 

Collapsing and bouncing out of the bottom of a squat not only wreaks havoc on joints but it reinforces the action of minimizing torque, power production, and movement efficiency.  At best this dangerous maneuver will bounce the lifter back to a position they would have squatted to in the first place (depth-wise) however because of the reduced muscular activation there will be an obvious sticking-point with significantly diminished torque throughout the concentric movement.  In contrast, the parallel or 90 degree squat represents the epitome for maximizing muscular stiffness, proprioceptive feedback, and elastic energy, as the muscles are stretched to their maximum range while maintaining optimal stiffness qualities.

Performance Training

These concepts and physiological principles correspond to other performance enhancement techniques including plyometric training and dynamic movement.  For example, it would never be advisable to have to have an athlete perform plyometrics by collapsing on impact or using excessive ROM.  Instead the goal is to maximize muscular tightness and stick the landing by employing the biomechanical and physiological principles described above. In turn these factors optimize the stretch reflex as it allows the muscles to function like coiled springs rather than wet noodles.   

squat Like a Human

As highlighted by the previous sections, the parallel or 90 degree squat represents a movement pattern where key principles foundational to neurophysiology, skeletal muscle physiology, motor learning, and biomechanics are congruent with each other.

With this in mind it’s no coincidence that parallel or 90 degree squats are the ideal squat method for humans as the concepts involved are not based on man-made rules but instead predicated on scientific principles that remain constant from human to human.  Unless you’re from another planet there’s no reason to think your body functions differently.

What About Olympic Weightlifters?

Accomplished Olympic weightlifters represent a very unique and rare bread of athletes.  Besides being incredibly strong and explosive, Olympic weightlifters have some of the strongest tendons, ligaments, and connective tissues of any sub-population.  This allows them to drop excessively deep under heavy loads with few immediate injuries relatively speaking.

In reality many Olympic weightlifters have flawed squat mechanics oftentimes displaying a significant valgus knee collapse with excessive eternally rotated foot/ankle complex - both of which are common by-products of reaching excessive squat depth.  Rather than seeing the squat as a therapeutic movement these athletes often have minimal concern for squat mechanics recognizing the sole purpose of the squat for their sport is to reach the most extreme depth they can handle with the ultimate goal of cleaning or snatching the heaviest possible weight.

In fact many of these extreme depths would be impossible without the assistance of weightlifting shoes as their bodies are able to rely on this unnatural elevation and lateral support to assist in further degradation of natural body mechanics.

It should be noted that not all Olympic weightlifters demonstrate flawed squatting mechanics.  In fact some of the all-time greats including Pyrros Dimas were known for avoiding excessive depth particularly during training as extreme positions were often saved for the most dire circumstances such as max attempts in competition. Similar themes are shared amongst some of the strongest Olympians as the most proficient lifters will stick the catch phase rather than letting the weight drag their bodies towards the floor once they receive it.

What About Powerlifting Rules?

It’s important to understand that powerlifting rules and guidelines are not based on optimal human mechanics but instead specific criteria for allowing visually subjective judging to be streamlined for competitive circumstances. Although many powerlifting organizations require depth that represents excessive range of motion (usually by a few inches) some organization hold criteria that are actually quite close to ideal squat depth with respect to optimal human mechanics.

Ironically some of the strongest powerlifters in the world utilize squat mechanics that just so happen to represent ideal or close to ideal squat depth.   They accomplish this by employing extremely controlled eccentric movements, stopping approximately at parallel, maintaining maximal tightness, and avoiding any type of collapse at the bottom position.  In fact the type of records that are broken would be impossible to perform safely with ATG squats. 

One has simply to watch various YouTube videos of 1000-pound plus squat attempts to notice a common theme.  Nearly all the lifts involve questionable depths as most are at the parallel mark with occasional lifts just above it.  In reality few if any 1000-pound squats would ever occur if the lifter went significantly below the parallel mark.  The amount of force they would be forfeiting would be substantial not to mention the strain it would place on their spine and surrounding joints.  In such scenarios the lifter is more concerned with stopping the eccentric phase at the point their body tells them to rather than breaching their natural depth mechanics to satisfy a subjective man-made rule and regulation.  The position where this occurs is almost always at approximately parallel.

How to Compete as a Powerlifter

On a side not I’ve actually trained a fair number of powerlifters and reaching proper depth for competition is never an issue for 4 main reasons. 

  1. First and as previously mentioned the ideal parallel squat depth is quite similar to the criteria listed for many organizations. That means a difference of one to several inches at most. Simply taking a few weeks before the meet to slightly adjust form and get comfortable with the extra depth will do the job.

  2. Secondly, if you perform pause squats or eccentric isometrics during training (which you should) and eliminate this pause during competition, you’ll inevitably gain a slight increase in depth due to the elasticity of the muscles.

  3. Third, the closer the load is relative to one’s max, the greater the level of compressive forces, which tends to increase ROM by a few centimeters.

  4. Fourth and most importantly is the exponentially greater training stimulus that parallel or 90 degree squats have on the body compared to ass to grass squats when it comes to strength and hypertrophy. As a result the weight improvements accrued during training from parallel or 90 degree squats will be so far superior, even if you have to reduce the load slightly to reach depth at competition, your numbers will still outweigh what you would have achieved from relying on the subpar stimulus of training with ass to grass squats.


But the Research Says……

Currently, a majority of research studies suggest deep squats are more effective and healthier on the joints than partial squats or parallel squats.  When comparing deep squats to partial squats these results are to be expected.  However the fact that deep squats have been shown to be safer and more effective than parallel or 90 degree squats can be attributed to one factor: faulty research and flawed application of practical training methods. 

Unfortunately, most research studies involving heavy strength training are carried out by lab rats who have no clue how to properly squat and demonstrate even greater incompetence when it comes to coaching these basic movements.  I’ve actually had the opportunity to witness many kinesiology investigations at university settings and to suggest there is a lack of proper coaching and cuing is an understatement.   The fact that ATG squats appear superior to parallel or 90 degree squats during these investigations can be traced back to improper execution of the squat, namely lack of posterior chain activation due to faulty hip hinge mechanics.

Because most individuals do not hinge adequately during the squat (unless properly instructed), the glutes and hamstrings are nearly dormant until excessive depth is reached.  In these circumstances excessive depth is necessary to activate the knee stabilizers and posterior chain all of which would have been fully activated throughout the entire motion if in fact proper 90 degree or parallel squat mechanics were employed. 

In essence activating the posterior chain during squats is critical for protecting the knees and surrounding joints.  This can be done either by properly hinging throughout the entire motion of a biomechanically sound parallel squat (which requires adequate coaching) or by employing excessive depth, which promotes inflammation, spasticity, dysfunction, and faulty mechanics. 

To summarize, the biggest factor that dictates posterior chain activation during squats (as well as other lower body movements) is not depth but instead is ample hip hinge mechanics. 

The Flipside of Research

Although the above discussion highlighting numerous flaws in current kinesiology research methods may help relay the aforementioned points to many individuals, some skeptics will need further scientific support for the notion that 90 degree squats are in fact ideal. With that said there is in fact substantial research supporting this although it’s likely these studies were not without their own set of flaws.

Squat Depth Analysis and Optimal ROM

Recent studies examining squat depth further support the concept of optimal range of motion and 90 degree joint angle mechanics. Strength coaches have long held the belief that larger ranges of motion, significantly greater than 90-degree joint angles, such as ass-to-grass squats, are ideal for building strength, size, and power output in athletic populations, mainly because of the difficulty of the task and the soreness associated with it. However, a recent study comparing the effect of training at different squat depths on joint angle specific strength, as well as transfer to sprint and jump performance, found that both partial squats (slightly less than 90 degrees) and parallel squats (slightly greater than 90 degrees) significantly improved vertical jump performance, with slightly greater improvement observed in the partial squat training group, while far less transfer was found from the deep squat training protocol (significantly greater than 90 degrees) to sprint or vertical jump performance [8]. In other words, deep or ATG squat training improved individuals’ ability to perform ATG squats but did not appear to enhance other sports related performance attributes. In contrast, the groups that trained at squat joint angles closer to 90 degrees produced superior results with significant improvements in jump and sprint performance. Perhaps the best results would have occurred had the researchers used a group that employed exactly 90 degree joint angles rather than slightly above or below. Obviously additional research is warranted.

90-Degree Joint Angles and Muscle Activation

Many strength coaches and practitioners will still argue that performing movements with greater ROM, such as ATG squats or squats well in excess of 90 degree joint angles, produce more muscle activation and ultimately greater long-term benefits in terms of strength and hypertrophy. Even if this were true (which it is not), the gains in strength and hypertrophy would not outweigh the negative ramifications associated with the disruption of optimal body mechanics, or the structural damage and inflammation of the surrounding joints. The notion that deeper squats or a greater range of motion on any movement produces more muscle activation is quite inaccurate, as shown in a number of research studies that not only invalidate this myth but, in fact, suggest quite the opposite.

Studies have shown that not only is excessive squat depth unnecessary, 90 degree joint angle mechanics are ideal, both biomechanically and structurally, as well as neuromuscularly, in terms of muscle activation and motor unit recruitment. In fact, contrary to what has incessantly been preached in the strength conditioning industry, a recent study that examined the effects of squat depth on muscle activation, showed that moving significantly past 90 degree joint angles or parallel positions did not produce greater muscle activation [12].

Yet another study of squat depth and its effect on muscle activation, described even more profound results. In this particular study the researchers examined 3 different squat depths: significantly above 90 degrees (20 degrees of knee flexion), exactly at 90 degrees, and significantly deeper than 90 degrees (approximately 140 degrees of knee flexion) [13]. While most practitioners would have predicted that the deepest squats (140 degree joint angle) would produce the greatest muscle activation in the quadriceps and gluteal muscles due to the greatest degree of stretch, the results indicate the exact opposite. More specifically, 90-degree joint angle squats appeared to produce the greatest muscle activation in the thighs and glutes, followed by the short or partial squat group (20 degrees of knee flexion), with the deep squat group (140 degrees of knee flexion) producing the least activity in the lower body musculature. It should also be noted that glute activity was unusually low in the deep squat group (140 degree) relative to the other groups, further contradicting the common, yet false belief, that deeper squats are ideal for glute development. In reality, they’re quite inferior when compared to proper squats at approximately 90 degree joint angles.

A similar study showed that incorporating partial squats with a range of motion of approximately 90 degrees of knee flexion in maximal strength training, produced superior results in terms of dynamic and isometric measures of maximal strength, as compared to performing only full ROM squats with a larger range of motion (i.e. 120 degrees) [14]. Ironically, the group that performed partial squats not only improved their ability to produce force at 90-degree angles but also at larger 120 angles. In other words, it appears that using optimal 90 degree joint angle mechanics may increase strength and force production at larger joint angles, such as 120 degrees, even more so than training exclusively at these larger joint angles. This is likely due to the increased motor unit recruitment and improved body mechanics associated with approximately 90 degree joint angles, which increases strength and muscularity to a far greater degree than does collapsing and using excessive range of motion.

Simply put, the results of these studies, as well as others highlighted in prior sections, indicate that 90 degree joint angles represent the optimal biomechanical positions not only in terms of producing and absorbing force, and protecting the joints, but also in terms of producing the highest levels of muscular recruitment. In other words, due to the greater levels of motor unit recruitment and leverage, the muscles are not only in the ideal position to produce optimal force and torque, the 90-degree joint angles are also the safest on the joints due to the fact that the muscles are firing at maximal levels (a key component of shock absorption) thereby taking the greatest amount of stress off the joints and connective tissue. Additionally, these results suggest that from a functional strength and hypertrophy perspective, 90-degree joint angles are ideal for maximizing size and force production due to the improved ability to recruit more muscle fibers, a prerequisite for optimizing muscle growth.

Lastly, it should be noted that in many of the previously mentioned squat studies a consistent trend becomes apparent when comparing joint angles greater and less than 90 degrees. For instance, in nearly every case it appears that while 90 degree joint angles are optimal, going significantly beyond 90 degrees (deep squat) seems to produce far inferior results compared to stopping short (partial squats) by nearly all measures, including muscle activation, force production, performance, jump height, and power output. This is likely indicative of some deeper and more profound physiological response such as neurological inhibition and autogenic inhibition. Simply put, stopping short of 90 degree joint angles may not fully maximize muscle activation by simply limiting the degree of motor unit recruitment. However, going significantly beyond 90 degree joint angles appears to breach our body’s optimal range of motion, producing varying degrees of inhibitory signals, neurological shutdown, proprioceptive distortion, and sensory interference. These results suggest that stopping short of 90 degrees is far superior than going significantly beyond it.

Fatigued Squat vs. Power Squat

Increased ROM beyond that which is considered optimal is never ideal for any movement and can often times indicate dysfunction or aberrations in movement patterns.  Recent research has shown that large ROM is associated with fatigue, reduced proprioceptive feedback, and as a result, increased risk of injury.  In other words just because you’re producing exaggerated ROM with ever-increasing levels of mobility doesn’t indicate that you’re performing productive movement. In fact, these very characteristics as demonstrated by research are associated with sloppy, fatigue-related movements involving decreased motor control, compromised muscle function, increased risk of injury, reduced muscle activation, neuromuscular inhibition, and decreased proprioception, all of which are far from advantageous.  Another way to think of this is that quality movement involves controlled, crisp, and concise movements with moderate/natural ROM whereas dysfunctional movements involve the exact opposite, namely excessive ROM, and exaggerated motions.  This concept is applicable not only for squats but for all movements.

Appropriate Blanket Statements

Many coaches and trainers are quick to point out that there is no such thing as a one-size fits all approach as doing so is simply making use of blanket statements.  However when it comes to coaching advice anything that we accept as a standard norm for any movement and performance concept is a blanket statement. In fact most of what we see in the industry involves such blanket statements. For example the recommendation of avoiding valgus knee and ankle collapse during lower body movements, avoiding elbow flare on bench press, minimizing cervical flexion and hyper extension on axial loading movements, coaching optimal shoulder and scapular packing on pressing exercises, or advocating the need to avoid lumbar flexion during deadlifting, these are all blanket statements which don't change from person to person as they are universally sound principles predicated on scientific concepts that remain constant no matter who you are.

The same is true for the squat. There is an optimal way to squat just as for any movement or skill there are optimal biomechanics. In fact this is how expert biomechanists make a living by determining the exact angles and positioning to perform a movement that maximizes performance, training, and overall osteokinematics. There are rarely “numerous optimal” ways to perform a particular movement.  Often it boils down to one specific approach and precise manner of execution that maximizes physiological performance.  The squat is no different.

Research and Real World Experience

As previously mentioned a recent study performed at the A.T. University in Arizona [8] compared the effects of squat depth on sprint and jump performance.  Participants were assigned either to a partial squat training protocol, parallel squat protocol, or full deep squat (ATG) protocol.  Results showed that partial squats and parallel squats both had a significant impact on improving vertical jump performance (with slightly greater results in the partial squat group) while the deep squat protocol elicited essentially no improvements in sprint or jump performance.  In other words deep or ATG squats appeared to have no favorable impact on performance related attributes other than improving the ability of the individual to perform ATG squats.

This represents the same phenomena I’ve witnessed when training my own athletes.  For instance, I often work with athletes who before seeking my services were previously squatting with ATG technique (often times with form that would be considered very solid even amongst ATG enthusiasts).   When we asses them we notice a direction correlation: those who were incorporating ATG squats display significant flaws in jumping technique, running form, and overall movement mechanics as they’ve disrupted their natural biomechanics and length-tension relationship of their muscles.  Once we re-train them to squat to 90 degrees with proper mechanics not only do they stop complaining of pulled muscles and continuous tweaks they were once accruing during their time of using ATG squats, but their vertical jump height, sprint speed, and overall form on these various activities markedly improves within weeks.

Eventually It Breaks

Although some lifters can temporarily “get away” with excessive squat depth, eventually it will produce negative ramifications.  I’ve seen this numerous times in lifters I’ve consulted with as they explain how they’ve been squatting to extreme depth for years with no pain or discomfort and eventually it hit’s like a ton of bricks.  Suddenly, out of nowhere there’s seems to be a tipping point where a once seemingly healthy lifter begins to have incredible pain in the hips, knees, and low back, as well as other physical symptoms.  For some this may occur after months while for others it may take years.  Regardless of the time frame, overall strength and performance not to mention muscle function and joint health would have improved to a much greater degree had proper mechanics and depth been employed in the first place.

Re-Thinking Eccentric Movement

Most lifters completely forget the function of eccentric movement.  In fact many trainees and coaches think of eccentric motion as an arbitrary portion of movement that just happens to stimulate muscle growth through various forms of musculoskeletal lengthening.  In reality eccentric movements serve 2 primary functional purposes for the body and are usually a means to an end when it comes to movement. 

First eccentric motions are useful for absorbing force during movements involving high impact or heavy resistance. The other purpose of eccentric movement is to set the individual up for the most powerful and efficient concentric motion.  If the eccentric motion is too large or too short these functions will be compromised.  In other words fulfilling some pre-determined path or satisfying a certain depth criteria for successfully completing a lift is an unnatural, counter-intuitive, and non-functional approach to eccentric motions. 

Instead the lifter should approach each eccentric phase with the goal of achieving ideal mechanics for maximizing force production and absorption not only for that eccentric phase itself but also for the subsequent concentric movement.  

The Truth About Shear and Compressive Forces

Although various forms of research including a study by Escimalla showed that squatting movements performed at joint angles in excess of 90 degrees place undue shear and compressive forces on the knee joint, some individuals will point to other literature suggesting that 90-degree joint positions produce greater shear and compressive forces on the surrounding joints than other joint angles. Unfortunately, this has led to many erroneous conclusions and misinterpretations of what constitutes proper and improper movement. High levels of shear and compressive forces on a joint (a debatable issue in and of itself) do not necessarily result in significant trauma to that joint, or indicate that the joint is in a potentially hazardous position.

In fact, most 90-degree joint angles create significant shear and compressive forces on the corresponding joints, however, the actual impact and trauma on those joints is minimal due to the fact that the surrounding musculature is in the ideal position to produce and absorb force. Additionally, many therapeutic positions and movements such as sprinting and jumping as well as various lifts such as Romanian deadlifts involve high levels of shear and compressive forces, suggesting that narrowly focusing on shear and compressive force issues alone, rather than as part of a larger, more complex system, is very uninformative, and actually quite misleading.

A helpful illustration is provided by comparing the widely popular Romanian deadlift (RDL) or barbell hip hinge and the deadlift. We know, from the perspective of force vector physics, that the bent over torso position produces very high levels of sheer and compressive forces on the spine, yet as long as spinal alignment, hip hinge mechanics, and proper form are employed, not only is the RDL quite safe on the spine, it’s extremely healthy and therapeutic. On the other hand, from a shear and compressive force standpoint while a very upright deadlift position, technically speaking, would incur very little stress on the spine, if the spine is flexed, the amount of strain and risk to the spine would be exponentially greater compared to performing an RDL or hip hinge with pristine 90 degree joint angle mechanics.

However, if we compared the two based solely on the principles of shear and compressive forces, we would conclude that the upright deadlift position with a flexed spine is far safer than the biomechanically efficient RDL with neutral spinal mechanics, yet we know the opposite to be the case. In other words, as mentioned above, shear and compressive forces are two of many factors that, when considered in isolation from other factors, can be quite misleading and even produce false assumptions about what constitutes proper mechanics. Factors dealing with movement mechanics such as, spinal integrity, joint arthrokinematics, joint positioning, proprioception, neurophysiology, and muscle activation are much more critical in the long run.

It is critical, therefore, that we examine this biomechanical component in the context of other factors including neurophysiological and structural components such as proprioception, intrafusal muscle fiber innervation, agonist antagonist co-contraction, reciprocal inhibition, muscle stiffness properties, elastic energy, lever arms, the length-tension relationship of muscle fibers, and somatosensory feedback to name a few. All of these factors, when looked at closely, and as previously discussed, show that 90-degree joint positions are ideal for the human body not just some of the time, but nearly all of the time. These concepts are scientific constructs that remain constant from person to person, the exception being individuals born with some extreme musculoskeletal abnormality or deformity.

Finally, how we view the human body is extremely important when considering this and all other biomechanical and movement related topics. If we were to isolate the bony structures of the body and view it simply as a skeleton or set of robotic segments, with no connection to the neuromuscular system, then shear force and compressive forces might provide great insight. However, when we examine the body in its totality, as one large, complex system governed by anatomical, neurophysiological, biomechanical, and physical principles, it completely alters the dynamics of what constitutes therapeutic movements and contra-therapeutic movements.

Parallel for Beasts, ATG for Weaklings

Many lifters will claim that ATG squats are superior to parallel squats or 90 degree squats simply because they’re more difficult, intense, and strenuous on the body.  In reality the opposite is true.  Parallel or 90 degree squats require markedly greater activation, muscular tension, concentration, mental fortitude, intensity and overall strength. ATG squats are the antithesis of this involving reduced muscle activation, and neuromuscular relaxation.

Supporters of ATG squats suggest that the increased load the lifter is capable of handling on parallel squats is simply a form of cheating as they’re making the exercise easier.  Again this notion could not be more flawed.  When performing ATG squats it is true that the lifter won’t be capable of handling as much weight however this is byproduct of decreased motor unit recruitment and reduced muscle activation necessary for achieving such a collapsed position.  As a result the lifter is punished with compromised contraction strength and reduced force-producing capabilities. 

The sensation of the lighter load feeling heavier is simply a byproduct of faulty mechanics giving the lifter the illusion that he or she is doing more work.  In reality they’re doing less work as they’re reinforcing neuromuscular inefficiency making the movement feel unnecessarily taxing and physiologically exhausting. 

On the other hand a proper 90 degree squat rewards the lifter with greater force-producing capabilities as every muscle fiber is firing maximally turning the muscles into coiled springs.  This requires the highest levels of concentration, and focus, as well as mental and physical intensity as there can be no weak links or energy leaks.  Essentially this maximizes load, power, speed, motor unit recruitment, intramuscular tension, and hypertrophy, all of which are compromised during ATG squats. 

In essence, with 90 degree squats the lifter is controlling the load – with ATG squats the load is controlling the lifter.

Read more about proper squat form here.

*** Stay tuned for Part 2 of this article series in which Dr. Seedman will address practical application of proper parallel squats in terms of exercises cues, instructional pointers, movement selection, programming, and further analysis of optimal squat mechanics. ***


  1. Blackburn, T., The Relationship Between Muscle Stiffness and Muscle Spindle Sensitivity in the Triceps Surae. National Athletic Trainers Association Research and Education Foundation, 2004.

  2. Dutto, D.J. and G.A. Smith, Changes in spring-mass characteristics during treadmill running to exhaustion. Med Sci Sports Exerc, 2002. 34(8): p. 1324-31.

  3. Kistemaker, D., et al., Control of position and movement is simplified by combined muscle spindle and Golgi tendon organ feedback. Journal of Neurophysiology, 2012. 109: 1126–1139, 2013.

  4. Zhang, L.Q. and W.Z. Rymer, Reflex and intrinsic changes induced by fatigue of human elbow extensor muscles. J Neurophysiol, 2001. 86(3): p. 1086-94.

  5. Edman, K.A., T. Radzyukevich, and B. Kronborg, Contractile properties of isolated muscle spindles of the frog. J Physiol, 2002. 541(Pt 3): p. 905-16.

  6. Pedersen, J., et al., Localized muscle fatigue decreases the acuity of the movement sense in the human shoulder. Med Sci Sports Exerc, 1999. 31(7): p. 1047-52.

  7. Changela, P. and R. Selvamani, A Study to Evaluate the effect of Fatigue on Knee Joint Proprioception and Balance in Healthy Individuals. International Journal of Scientific and Research Publications, 2012. 2(3).

  8. Matthew R. Rhea, J.G.K., Mark D. Peterson, Drew Masse, Roberto Simao, Pedro J. Martin, Mike Favero, Diogo Cardozo, Darren Krein, Joint-Angle Specific Strength Adaptations Influence Improvements in Power in Highly Trained Athletes. Journel of Human Movement, 2016. 17 (1), 43-49.

  9. McLester, J. and P.S. Pierre, Applied Biomechanics: Concepts and Connections. 2008: Thompson Wadsworth.

  10. Lieber, R. and J. Boakes, Sarcomere Length and Joint Kinematics During Torque Production in Frog Hindlimb. Division of Orthopedics and Rehabilitation, Department of Surgery, Veterans Administration Medical Center and University of California, 1988.

  11. Escamilla, R.F., Knee biomechanics of the dynamic squat exercise. Med Sci Sports Exerc, 2001. 33(1): p. 127-41.

  12. Contreras, B., et al., A Comparison of Gluteus Maximus, Biceps Femoris, and Vastus Lateralis Electromyography Amplitude in the Parallel, Full, and Front Squat Variations in Resistance-Trained Females. J Appl Biomech, 2016. 32(1): p. 16-22.

  13. Marchetti, P.H., et al., Muscle Activation Differs between Three Different Knee Joint-Angle Positions during a Maximal Isometric Back Squat Exercise. J Sports Med (Hindawi Publ Corp), 2016. 2016: p. 3846123.

  14. Bazyler, C.D., et al., The efficacy of incorporating partial squats in maximal strength training. J Strength Cond Res, 2014. 28(11): p. 3024-32.