Squat Technique: Optimizing Vertical Movement Part 2

Continued from part 1 of this series which can be viewed here

In the last section of our analysis, we looked at Pelvis GyroX data patterns to create time stamps for different phases of the squat, how to begin analyzing this movement in terms of depth and keeping the hips under the center of mass, and how to apply an external cue to create changes in the movement. In this next part of our analysis we will continue by looking at our 3rd measurement of the squat, Spinal Coupling.

Spinal Coupling:

To assess how well an athlete is able to maintain the position of their torso relative to their hips (I call this spinal coupling), especially at the bottom of this movement (< 90° knee angle), I like to use the H-Angle output from our sensors on the Pelvis (sacrum) and Torso (sternum). In optimal conditions, the offset of these two outputs should stay consistent throughout the full range of motion. 

As we see in James' squat above, at the 1st 0-cross of the Pelvis GyroX output (roughly 90° knee angle), his pelvis and torso angles match up at 41.5° and 41.8° respectively. This indicates good coupling of these two components through the first half of the eccentric phase of his squat.

As he proceeds to the bottom of the squat (2nd Pelvis GyroX 0-cross), we can see these two angle values split off in different directions as the torso becomes more aggressively tilted forward (from 41.8° to 35.5°), and the pelvis begins to posteriorly tilt (from 41.5° to 56.2°) resulting in an offset of these two angles of 20.7° and a visible rounding of the spine.

The addition of a heel lift alone reduces this offset from 20.7° to only 7.4°. To further correct this, we add an external cue in the form of a front or back loaded light weight. 

With the addition of a light (relative) cue as an unloaded bar on the back, James' pelvis and torso angles now line up symmetrically at 56.3° each, with 0° of offset. Adding this light weight cues the upper posterior chain to stay engaged through the hip flexion range of motion. This allows full squat depth, an optimal torso vs tibial angle, and a maintenance of torso and pelvic angle offset that reinforces good postural stability through the entire movement.

It is important to note that these values do not have to match perfectly as they do in this case, and the absolute angle values may vary depending on sensor attachment site. What is more important is that they maintain a constant relationship (or offset) throughout the range of motion. 

Let's now review these concepts with our runner subject, Wilson Kipsang.


Initial observations from Wilson's squat movement:

Depth- Due to his anatomy (lower body limb lengths) and also dorsiflexion range of motion, we see an altered Pelvis GyroX output. This one is lacking the same two distinguishing 0-crosses we saw in James' case. This means we are not achieving optimal depth of the squat. Based on the H-Angle output of the sensors on his thighs, his peak squat depth is about 23°. We will see how we can increase this shortly.

Torso vs Tibial Angle- In this assessment we opted to place sensors on his thighs as opposed to his tibias. What we can still observe though is that his torso is going through a total angular range of about 45.6°. In James' case, once external cues were applied, he was able to minimize his total torso angular range to just 33.7°. Visibly, we can also see that the torso is tilted forward at a more aggressive angle than the tibia.

Spinal Coupling- Wilson's spinal coupling presents as pretty good in our initial assessment, displaying an offset between his torso and pelvis of just 6° at the bottom of his squat. We will want to maintain this positive quality and monitor it for improvement as we work through some modifications.


With the addition of a 1.25" heel lift we see positive changes in all three of our movement criteria:

Depth- Improvement from 23° (measured on thigh) all the way down to 3°. This is a 20° increase in range of motion for the femur.

Torso Angular Range- The heel lift effectively decreased the total angular range of the torso from 45.6° to 34.9° (which is much closer to the optimal value James was able to achieve of 33.7°). Let's see if we can improve this further with the addition of a light weight to cue some additional postural stability.

Spinal Coupling- This heel lift alone improved Wilson's offset of torso and pelvis angles from an already favorable value of 6° to even better at around 0.5°-1°.

The final question will now be: Can we load this movement pattern and preserve these modifications and movement qualities?

Holding a kettlebell in a front-loaded position, we see that all movement qualities are preserved! The depth of the squat remains between 4-5° on the thigh, spinal coupling is held to within 2°, and the torso angular range actually improves to 33.9°.

With both of these cases, our focus will now be to preserve these movement qualities as we progress training load and simultaneously work to reduce reliance on these external cues. With these values now documented for us, we can easily track deviations from optimal movement and understand when our training load exceeds our functional capacity to perform these movements most effectively.


Joseph Cavarretta, MS, ACSM-EP

LEOMO Sport Scientist & Coach

Leave a comment

Please note, comments must be approved before they are published