Why the Deep Squat matters
A deep squat looks simple, but getting all the way down and back up requires several parts of the body to work together. The ankles, knees and hips move through large ranges, while the pelvis and trunk adjust above them to keep us balanced. Depending on the task, the spine, shoulders, arms and head also influence how we perform the squat.
Weightlifters and CrossFit athletes train the squat directly, but the ability to squat also matters in sports where the connection is less obvious. In mountain biking, for example, coaches may assess squat mechanics and ankle dorsiflexion because riders use free weights to build strength. That training requires them to produce force while maintaining balance, which also matters when they have to control the bike during a demanding race and are already tired. In dance and acrobatics, squat depth matters for another reason. It contributes to the ability to jump high.
This does not mean that every athlete needs to squat in the same way. The deep squat is useful because it brings several movement capacities into one task. Research has associated squat depth with ankle dorsiflexion and hip mobility, so the depth someone reaches may tell us something about several capacities at once (Kim et al., 2015).
The squat also matters beyond sport. Research shows that older adults have more difficulty reaching deeper squat positions and use different movement strategies to get there (Honda et al., 2022). Maintaining the ability to move comfortably into and out of low positions is one part of preserving freedom of movement as we age. The deep squat gives us a practical way to observe several of the capacities involved.
Before we can assess the deep squat, we need to resolve an apparent contradiction. There is no single “perfect” squat.
There is no single perfect squat
How we squat changes with what we are trying to do. When we lift a heavy crate, we adapt to the load in front of us. When we squat down to spend time with a child, there is no external load to manage, so we have more freedom to organize the movement. A weightlifter may need to receive or stabilize a load overhead, which leaves less freedom in how the squat can be organized.
Different tasks invite different squat strategies.
This is more than a matter of style. Changing stance width or foot angle affects joint mechanics. Elevating the heels changes ankle and knee motion. Adding external load changes the movement again. A wider stance or a different foot position can also change the mobility demands of the task (Lorenzetti et al., 2018; Ghasemi et al., 2026; McKean et al., 2010; Demers et al., 2018).
For training and everyday movement, we adapt the way we squat to the task. In an assessment, we turn that around. We define the task and ask whether you can adapt your movement to it.
Rather than observing only how someone usually squats, we ask them to attempt a specific movement challenge. Can they maintain the defined stance, keep their heels on the floor, keep their arms alongside their body and still reach depth?
We are not trying to decide whether someone’s usual squat is right or wrong. The defined conditions give us a consistent task and make it easier to observe mobility, balance, coordination and movement strategy. If someone changes their foot position, reaches forward with the arms, lifts the heels or stops at a shallower depth, those adaptations are part of what we want to document.
The person in the recording above is not demonstrating a “perfect” squat that everyone should copy. The movement shows one person’s response to the task.
A defined movement challenge
For this assessment, the task is simple. Stand with your feet about hip width apart and pointing forward. Keep your heels on the floor and your arms alongside your body. Then squat as deeply as you comfortably can.
These instructions tell someone what to attempt. They do not tell us what the movement must look like to be worth assessing. If the person adapts the movement, those adaptations still give us useful information.
As we develop Glance, we are not trying to compare every squat with one ideal. We are exploring useful ranges for individual movement characteristics and documenting how each person adapts to the task. Once someone starts moving, depth is only one part of what we can observe.
One movement, many components
Squat depth tells us how low someone went, but not how they reached that position.
From the side, we can also observe how far the tibia inclines as the knees move forward, whether the heels stay on the ground, how much the trunk leans and whether spinal alignment changes during the movement.
These characteristics describe different parts of the squat, but they are related.
Trunk and tibia inclination show why these measurements need to be considered together. Their relationship affects the relative mechanical demands on the hip and knee extensors. For the reference movement, we look for enough forward movement of the tibia to reach depth without excessive forward trunk lean. When the tibia remains more vertical and the trunk leans further forward, the demand on the hip extensors increases relative to the demand on the knee extensors. Looking at either angle alone misses part of the movement (Barrack et al., 2021; Graber et al., 2023; Straub & Powers, 2024).
Spine flexion gives us different information. Forward trunk lean does not necessarily mean that the spine itself is flexing. A person can lean forward while maintaining a relatively neutral spine. For the reference movement, we are interested in whether someone can reach depth while maintaining that alignment. If spinal flexion increases, we want to document it as a separate characteristic (Straub & Powers, 2024).
Tibia inclination
Trunk inclination
For the initial assessment from the side, we are focusing on three measurements: squat depth, tibia inclination and trunk inclination. Together, they show how low someone went and how the tibia and trunk moved as they got there. Whether the heels remain on the floor is treated mainly as a control condition.
Spine flexion and arm strategy are planned for a later release. For the frontal view, we plan to consider head position and potential measures of asymmetry, including dynamic knee valgus, lateral pelvic shift, pelvic drop and frontal trunk asymmetry.
Measurement and interpretation are different steps. Someone may show limited tibia inclination together with greater forward trunk lean. That pattern is useful to record, but it does not tell us why they moved that way.
Measure the components first. Interpret the pattern second.
Once we preserve these observations, we still need to decide how to present them without losing useful detail.
Assessment has an information problem
A squat takes only a few seconds. A skilled coach can see a lot in that time. But seeing movement and preserving what happened are two different things.
A coach might notice limited tibial progression, greater forward trunk lean or a change in spinal position. Those observations can guide training immediately. Weeks later, however, it may be difficult to remember the movement in enough detail to make a useful comparison. If the assessment preserves only squat depth or a single overall score, the individual observations are lost.
This creates a practical trade-off. A single overall score is easy to record and compare, but it does not show which movement characteristics produced it.
The Functional Movement Screen illustrates this well. Its Deep Squat assessment reduces a complex movement to an ordinal score from 1 to 3. Research suggests that those scores are not arbitrary. Different score groups show measurable differences in squat kinematics (Butler et al., 2010; Macmillan et al., 2023). But the score cannot tell us everything about the individual components that produced it, and research has found that it does not map cleanly onto specific differences in joint mobility (Aleixo et al., 2024).
A score can be meaningful without preserving everything that went into it.
For a quick screen, that compression may be exactly what we want. But if the goal is to understand a movement in detail and revisit it later, preserving the individual measurements becomes valuable. Assessment is then not only about what we can see today. It is about creating a record detailed enough to ask a much more useful question next time:
What changed?
Measure the components, then track what changes
When we preserve the individual measurements, we can return to them later and ask a more specific question: what changed?
Imagine someone with limited tibial progression who spends six weeks doing exercises intended to improve ankle mobility. At the next assessment, the tibia moves farther forward, but overall squat depth looks almost the same. The overall result may look similar even though one measured component has changed. Other characteristics may still limit the depth reached in the reference squat.
A detailed profile makes that change visible while also showing which measurements stayed the same. It does not explain why the remaining limitations occurred, but it gives the coach a clearer basis for deciding what to examine next. That is why we measure the components separately.
We are building Glance to preserve and compare these individual measurements over time. A person records a short video while performing the squat. Glance uses computer vision to detect body landmarks in the recording and derives movement measurements from their relative positions.
For those comparisons to become useful, each measurement needs to respond to the movement characteristic it is intended to capture. At this stage, our checks are descriptive and exploratory rather than a validation of accuracy. We introduce controlled changes to the movement and inspect how the resulting measurements respond. In exploratory tests from the frontal view, for example, we simulated progressively greater leftward trunk lean, beginning with a neutral position and continuing through mild, medium and strong conditions. We then checked whether the signal for trunk position reflected those changes. This helps us examine whether the implementation behaves as expected before assessing its accuracy more formally.
Seeing the signal change is only part of the problem. We reviewed the controlled conditions with the coach on our team and asked when he would consider the trunk lean meaningful. His assessment gave us a practical reference for interpreting the measured change. It does not by itself establish a universal threshold or validate the accuracy of the measurement.
Neutral
Mild left
Medium left
Strong left
Put the movement back together
Breaking a squat into individual measurements makes it easier to document. But a squat is not simply the sum of those measurements. The measurements become more useful when we consider how they relate to one another.
Considering the relative orientation of the trunk and tibia helps us describe the distribution of demands between the hip and knee extensors more clearly than either angle alone (Barrack et al., 2021; Graber et al., 2023; Straub & Powers, 2024). A later measure of spine flexion could add another layer: two people can show similar forward trunk inclination while differing in how that inclination is distributed between the hip and spine (Straub & Powers, 2024).
Putting the movement back together does not mean collapsing the measurements into another overall score. It means considering related measurements together while keeping each result available. As we develop additional measurements from the side and frontal views, we can extend this combined description of the movement. We refer to that combined description as synthesis.
Break the movement apart to measure it. Put it back together to understand it.
Looking closer at movement
Squat depth alone does not tell us how a person reached that position. The individual measurements, the relationships between them and how they change between assessments can all provide useful information.
This is what we are exploring as we build Glance Movement Diagnostics. Glance uses computer vision to derive structured measurements from a short movement recording so they can be preserved and compared over time. Our aim is to help athletes, coaches and others understand what changed without reducing the movement to a single score.
Glance Movement Engine is in development
Release planned for late 2026.
A personal note from Karin
I would like to thank Witalij Kühne for taking the time to share his perspective on movement and coaching. Our conversations have led me to examine more critically what meaningful movement assessments should capture. They have also helped me ground my technical work in the practical realities of movement coaching. I am very grateful for his time, insight, encouragement, and continued support.
References
Aleixo, P., Atalaia, T., Bhudarally, M., Miranda, P., Castelinho, N., & Abrantes, J. (2024). Deep squat test: Functional Movement Screen: Convergent validity and ability to discriminate subjects with different levels of joint mobility. Journal of Bodywork and Movement Therapies, 38, 197–204. https://doi.org/10.1016/j.jbmt.2023.12.010
Barrack, A. J., Straub, R. K., Cannon, J., & Powers, C. M. (2021). The relative orientation of the trunk and tibia can be used to estimate the demands on the hip and knee extensors during the barbell back squat. International Journal of Sports Science & Coaching, 16(4), 1004–1010. https://doi.org/10.1177/1747954121997957
Butler, R. J., Plisky, P. J., Southers, C., Scoma, C., & Kiesel, K. B. (2010). Biomechanical analysis of the different classifications of the Functional Movement Screen deep squat test. Sports Biomechanics, 9(4), 270–279. https://doi.org/10.1080/14763141.2010.539623
Demers, E., Pendenza, J., Radevich, V., & Preuss, R. (2018). The effect of stance width and anthropometrics on joint range of motion in the lower extremities during a back squat. International Journal of Exercise Science, 11(1), 764–775. https://doi.org/10.70252/BWZE8275
Ghasemi, M., Emami, M., & Mohammadi Yaghoubi, U. (2026). Heel elevation increases ankle and knee range of motion during squatting in healthy adults: A systematic review with meta analysis. Sports Biomechanics, 1–17. https://doi.org/10.1080/14763141.2026.2619893
Graber, K. A., Halverstadt, A. L., Gill, S. V., Kulkarni, V. S., & Lewis, C. L. (2023). The effect of trunk and shank position on the hip to knee moment ratio in a bilateral squat. Physical Therapy in Sport, 61, 102–107. https://doi.org/10.1016/j.ptsp.2023.03.005
Honda, K., Sekiguchi, Y., Shimazaki, S., Suzuki, R., Suzuki, T., Kanetaka, H., & Izumi, S. I. (2022). Effects of aging on whole body center of mass movement and lower limb joint kinematics and kinetics during deep squat movement. Journal of Biomechanics, 134, 110996. https://doi.org/10.1016/j.jbiomech.2022.110996
Kim, S. H., Kwon, O. Y., Park, K. N., Jeon, I. C., & Weon, J. H. (2015). Lower extremity strength and the range of motion in relation to squat depth. Journal of Human Kinetics, 45, 59–69. https://doi.org/10.1515/hukin-2015-0007
Lorenzetti, S., Ostermann, M., Zeidler, F., Zimmer, P., Jentsch, L., List, R., Taylor, W. R., & Schellenberg, F. (2018). How to squat? Effects of various stance widths, foot placement angles and level of experience on knee, hip and trunk motion and loading. BMC Sports Science, Medicine and Rehabilitation, 10, 14. https://doi.org/10.1186/s13102-018-0103-7
Macmillan, C., Olivier, B., Benjamin-Damons, N., Wood, W. A., & Obiora, O. L. (2023). Altered sagittal plane mechanics is associated with Functional Movement Screen deep squat score. The South African Journal of Physiotherapy, 79(1), 1865. https://doi.org/10.4102/sajp.v79i1.1865
McKean, M. R., Dunn, P. K., & Burkett, B. J. (2010). Quantifying the movement and the influence of load in the back squat exercise. Journal of Strength and Conditioning Research, 24(6), 1671–1679. https://doi.org/10.1519/JSC.0b013e3181d8eb4e
Straub, R. K., & Powers, C. M. (2024). A biomechanical review of the squat exercise: Implications for clinical practice. International Journal of Sports Physical Therapy, 19(4), 490–501. https://doi.org/10.26603/001c.94600