Taekwondo Roundhouse Kick IMU Velocity Test: Quantifying Kick Power by Foot Angular Speed

Line up ten roundhouse kicks against a heavy bag with a force sensor bolted behind it, and watch the same athlete, throwing at what feels like identical effort, produce readings that swing 15 to 20 percent from one kick to the next. Coaches who have tried to build a power-testing routine around impact force run into this almost immediately, and the usual explanation — inconsistent effort — is often wrong. The bigger problem is where the foot lands on the bag, how the bag is still swinging from the last kick, and how far the athlete was standing when they chambered. Falco et al. (2009), studying black-belt taekwondo athletes kicking from three standardized distances, found that execution distance alone significantly changed both execution time and impact force — meaning two kicks that look identical on video can register very differently on a force sensor for reasons that have nothing to do with how hard the athlete actually kicked.

A shin-mounted IMU sidesteps that problem by measuring something the target never touches: the peak angular velocity of the shin segment as it whips toward contact. It works whether or not the kick makes contact at all, it isn't distorted by bag sway or pad compliance, and it gives a number that Gavagan and Sayers (2017) found reliably separated skill levels in taekwondo athletes when they tracked segment angular velocities with 3D motion capture. This guide walks through what that number actually represents, how to mount and calibrate a shin IMU for kick testing, a field protocol for repeated trials, skill-level reference bands, and how to read session-to-session change without a bag, a load cell, or a lab.

A force plate or a load-cell-instrumented bag measures the outcome of a kick — how hard the foot struck something — not the kick itself. Contact point shifts the reading independent of technique: a kick that lands flush transfers force differently than one that catches the target a few centimeters off-center, even at the same foot speed. Bag mass, hang angle, and residual swing left over from the previous rep all add noise a coach can't see from the sideline.

Distance compounds the problem. Falco et al. (2009) had experienced taekwondo competitors execute roundhouse kicks from near, mid, and far standardized distances and found that impact force and execution time both changed significantly with distance. In a gym setting where athletes rarely stand at the exact same spot twice, that alone can account for a good chunk of the rep-to-rep variability coaches blame on inconsistent effort. And a force-based setup has a more basic limitation: it can't measure a kick that never makes contact, which rules out shadow drills and any warm-up work where a coach still wants a power number without full-contact striking.

A roundhouse kick generates speed through a proximal-to-distal chain: the hip rotates first, the knee extends into the chamber, and the shin snaps through last, reaching its highest rotational speed in the final 30 to 40 milliseconds before the foot would strike a target. That final whip is where the kick's power actually lives, and it happens whether or not anything is standing in front of the foot. A gyroscope-equipped IMU strapped to the shin measures exactly that moment: the peak angular velocity of the tibia segment, in degrees per second, as it rotates through the kick's terminal phase.

This isn't a proxy pulled out of thin air. Gavagan and Sayers (2017) used 3D motion capture to compare roundhouse kick biomechanics across taekwondo athletes at different competitive levels and found that peak angular velocity of the kicking leg's distal segments separated skill groups more consistently than kick force measurements did — higher-level athletes reached higher peak rotational speeds in the shin and foot, a large, statistically significant effect between groups. The practical implication: angular velocity tracks the mechanical output of the kick itself, upstream of whatever happens at contact, which is exactly the variable a coach wants when comparing an athlete's kicks to their own history rather than grading a single strike.

One honest caveat: Gavagan and Sayers used lab-grade marker-based motion capture, not a wearable IMU, on a single cohort across a limited number of skill tiers. A shin-worn IMU is a validated, field-practical stand-in for that same segment-velocity measurement — highly sensitive to change within one athlete over time — but treat cross-study or cross-lab absolute comparisons cautiously, since marker placement and sensor location on the shin both shift the exact number reported.

Mount the IMU on the kicking leg's shin, anterior surface, roughly at the junction of the middle and lower third of the tibia — high enough to avoid ankle-guard interference, low enough to stay clear of knee-brace hardware if the athlete wears one. Strap orientation should run the sensor's long axis parallel to the tibia's shaft so rotation about the correct axis is what gets captured as the kick's angular velocity.

Calibration Sequence

1. Neutral standing hold (3 seconds): athlete stands square, kicking leg planted, establishing the sensor's zero-reference orientation before any kicking begins.
2. Chamber reference: the athlete lifts the kicking leg into a static chamber position and holds for 2 seconds, giving the app a reference range for where the kick's rotational phase begins so a walking step or stance adjustment doesn't get mistaken for the start of a kick.
3. Three calibration kicks: thrown at roughly 70% effort into open air or a light target, confirming the peak-detection algorithm correctly isolates the terminal-phase spike. A green check appears per kick when detection succeeds; re-strap after two consecutive misses.
4. Sampling check: confirm the sensor is logging at 800Hz or higher before a full testing set. A full roundhouse kick's terminal whip phase lasts well under 50 milliseconds, and a lower sampling rate risks missing the true peak between samples.

Standardize distance before anything else. Given Falco et al.'s (2009) finding that execution distance changes kick mechanics on its own, mark a fixed floor position for the lead foot and a fixed target height (roughly torso height, mid-line) for every session — even when testing without contact, keeping the athlete's setup distance consistent removes one of the largest sources of rep-to-rep noise.

Testing Steps

1. Warm up fully, including several sub-maximal roundhouse kicks, before any recorded testing set. Peak angular velocity on a cold kick underrepresents true capacity and will corrupt a baseline.
2. Perform sets of 5 kicks per leg at maximum intended effort, with 15 to 20 seconds of rest between reps within a set — enough to avoid pooling fatigue into a power test unless fatigue resistance is the specific thing being tested.
3. Discard any kick the app flags as a low-confidence detection — typically a chambering adjustment, a fake, or a kick where the athlete visibly pulled up short.
4. Record the peak angular velocity of each clean kick, then take the mean of the top 3 of 5 as that session's working number, discarding the two lowest as sub-maximal or technically flawed attempts.
5. Log this value against the athlete's personal baseline, built from at least 3 prior clean testing sessions performed under the same warm-up and rest conditions.

Test both legs. A meaningful lead-leg-to-rear-leg gap, or a gap that widens over a training block, is itself useful information for an athlete whose sparring or competition style leans heavily on one side.

These bands are a practical field translation of the skill-level separation Gavagan and Sayers (2017) reported, not a direct reproduction of their exact figures — their study used marker-based motion capture on a specific athlete cohort, and absolute numbers will vary with sensor placement, kicking leg definition, and measurement method. Use them to place an athlete's baseline, then track that individual's own trend from there.

These ranges describe typical output, not a pass-fail cutoff. An athlete sitting at the low end of their category with a stable, tightening trend over a training block is progressing normally; the number to worry about is a flat or declining trend against that same athlete's own prior sessions.

A single session's peak number matters less than how it moves against an athlete's own history. Track two things every session: the top-3-of-5 mean, and the coefficient of variation (CV%) across those three kicks, which tells you how repeatable the athlete's output is on a given day.

Two limitations are worth stating directly. Gavagan and Sayers (2017) worked with a single-cohort sample across a limited number of skill tiers, so the exact magnitude of the skill-level gap in peak angular velocity will not transfer identically to every population — treat the bands above as a starting orientation, not a fixed standard. And Falco et al.'s (2009) distance findings came from black-belt competitors kicking at standardized, coach-controlled distances in a lab setting; a live sparring or bag-work context introduces distance variability the protocol above is specifically designed to control for, but can't fully eliminate if an athlete's footwork drifts mid-set. Neither study used a wearable IMU directly, so the numeric bands here are an applied translation of their directional findings into a field-usable format, not a reported result from either paper.

Kick velocity testing earns its place in a program when it runs on a fixed schedule, not as an occasional curiosity.

• Weekly, post-warm-up: one full protocol per leg, logged against the rolling baseline — ideally the same day and time slot each week to control for diurnal and fatigue variation.
• Every 3–4 weeks: review the trend line rather than any single session, looking for a sustained shift rather than reacting to one noisy day.
• During a taper or competition week: a single light-effort check is enough to confirm the athlete feels normal; skip full maximal-effort testing sets so close to competition.
• After any lower-body injury: use the pre-injury baseline as a return-to-sport benchmark, since a kicking leg that has regained full strength and range of motion can still show a depressed peak angular velocity if the athlete is unconsciously guarding the movement.

Key References

• Gavagan, C. J., & Sayers, M. G. L. (2017). A biomechanical comparison of the roundhouse kick across the different levels in Taekwondo. PLOS ONE, 12(8), e0182645.
• Falco, C., Alvarez, O., Castillo, I., Estevan, I., Martos, J., Mugarra, F., & Iradi, A. (2009). Influence of the distance in a roundhouse kick's execution time and impact force in Taekwondo. Journal of Biomechanics, 42(3), 242–248.