How to Measure Seated Throw Power for Para-Athletics

Two seated throwers post almost the same distance at a meet, and one of them is already fielding questions. Her coach knows she has genuinely strong trunk rotation for her class; the classification panel sees a number that could just as easily belong to an athlete leaning hard on the backrest for a free push. Distance alone can't answer which one it is — a shot put or club throw is a single combined outcome of trunk drive, shoulder and elbow extension, release timing, and technique, folded into one final number. When a coach needs to show how much of that distance came from the trunk versus the arm, a field test that only records where the implement landed has nothing more to offer.

This guide walks through a field protocol that separates trunk-driven power from arm-driven power in a fixed throwing seat, using a trunk-mounted IMU, a wrist-mounted IMU, and a load cell on the trunk restraint strap. It covers what to measure, how to run a two-condition trial that isolates each contribution, what the published classification research says about trunk function and throw performance, and how to build the data into a training log and a reclassification file.

World Para Athletics seated throw classes group athletes by trunk and limb function precisely because trunk control changes how much force reaches the implement before the arm ever moves. An athlete with full trunk function can pre-rotate the torso, load the obliques, and whip the shoulder through a much longer power path than an athlete stabilized only by a chest strap and backrest. A tape measure at the landing point gives one number for a chain of contributions field observation alone struggles to pull apart — a classifier watching from the sideline can spot gross compensation, like an athlete rocking against the strap, but has no way to quantify how many degrees per second of trunk rotation, or how many newtons of strap load, actually produced the extra distance.

Chow and Mindock (1999), studying wheelchair discus throwers across three functional classes, found release velocity accounted for the large majority of variance in throw distance and that distance differed significantly between classes (p < 0.05) — but their own analysis noted that release velocity and trunk rotation range were both folded into a single measured outcome, distance, that couldn't on its own separate a technique effect from a true trunk-function effect. That gap between one outcome number and several underlying contributors is exactly what a distance-only field test still can't resolve today without instrumentation built to split trunk motion from arm motion at the sensor level.

Three numbers cover most of what a coach or classifier needs once trunk and arm can be measured separately: peak trunk angular velocity, peak wrist linear velocity at release, and peak restraint strap force during the acceleration phase. Strap force matters more than it might seem — an athlete braced against the backrest can still push against the restraint itself, converting what looks like isometric bracing into a real contribution to release velocity. A wrist IMU alone reads that as pure arm output; only the strap load cell catches it.

The seat itself has to match the athlete's actual competition configuration, since backrest height, strap placement, and footplate position set per class-specific seat regulations change how much trunk motion is mechanically available in the first place. Testing a class-restricted seat against a taller, more permissive backrest produces numbers that simply aren't comparable, and that mismatch is one of the most common ways a well-intentioned field test ends up misleading a coach about how much room an athlete has to improve.

Mount one IMU over the sternum or upper thoracic spine, secured with a low-stretch strap rather than tape so it doesn't shift during a maximal effort, and a second IMU on the throwing-side wrist. Both should sample at 200 Hz or higher to resolve the sharp velocity spike at release. Route a small strain-gauge load cell in line with the chest or hip restraint strap — most seating rigs can accept an inline load cell without changing the strap's actual restraint function, since the cell simply sits in the load path and logs force.

Calibration and Sync Sequence

1. Zero the load cell: athlete seated, strap fastened at rest, before any throwing motion, to set a true baseline restraint force.
2. Align both IMUs to a common axis convention: trunk sensor's primary axis to the sagittal rotation plane, wrist sensor's to the direction of release, confirmed with two practice rotations before recording.
3. Time-sync all three devices to a shared clock or trigger event, such as a hand-clap accelerometer spike, so trunk peak, wrist peak, and strap peak compare frame by frame rather than by approximated device clocks.
4. Run two submaximal practice throws confirming clean, un-clipped signal on all three channels before scoring a trial — a clipped wrist IMU on a fast release is a common early-session error.

Isolating trunk contribution means comparing the same athlete's release velocity under two conditions rather than reading one throw alone — a single trial, however well instrumented, can't separate what the trunk added because there's nothing to subtract it from.

In-Session Steps

1. Warm up with 4-5 submaximal throws in the competition seat configuration to confirm sensor sync.
2. Arm-only condition: trunk braced flat against the backrest, athlete cued to initiate from the shoulder only, 4 trials with full recovery (2 minutes minimum) between attempts.
3. Full-effort condition: athlete uses whatever trunk rotation and lean is available within their class's seat rules, 4 trials, same recovery interval.
4. For every trial, log peak trunk angular velocity, peak wrist linear velocity at release, peak strap force, and throw distance.
5. Discard any arm-only trial with a trunk angular velocity spike above roughly 15% of that athlete's typical full-effort peak — a sign the bracing cue wasn't followed — and any full-effort trial with an early release flagged by the timing gate.
6. Compute the trunk contribution ratio: (full-effort peak wrist velocity minus arm-only peak wrist velocity) divided by full-effort peak wrist velocity, averaged across retained trials.

The ranges below are a directional field reference built from published classification research and session-tracking patterns, not a certified pass/fail scale — classification decisions rest with accredited classifiers using their own assessment battery, and this protocol is a training and preparation tool, not a substitute for that process.

Individual variation inside each range is large — a well-trained F52 athlete with strong residual shoulder mechanics can outperform an untrained F55 athlete on raw distance despite a much lower trunk contribution ratio, so treat the ratio as a within-athlete tracking tool first. An elevated arm-only strap force reading also isn't automatically a compliance problem; it can reflect a genuinely difficult bracing instruction for that athlete's impairment, which is worth flagging to the athlete's own classifier rather than adjudicating alone.

The two studies behind this protocol looked at different pieces of the same problem, and neither alone gets a coach to a usable field ratio, which is why the isolation protocol above exists.

Chow and Mindock (1999) filmed wheelchair discus throwers across three functional classes and found release velocity was the strongest single predictor of throw distance, explaining the large majority of distance variance across throwers, with trunk rotation range adding measurable variance for the classes with greater trunk function. The study's limitation is a small, class-stratified sample filmed on a single throw type, with individual technique differences not fully controlled for, exactly the confound a within-athlete arm-only versus full-effort comparison sidesteps.

Bernardi et al. (2010) assessed trunk and arm function in Paralympic athletes across several seated sports, including seated throwing, and reported a meaningful positive correlation between a field-based trunk control score and throw performance, strong enough to support trunk assessment as a practical classification component. Their sample spanned athletes with different underlying impairments grouped by resulting trunk function rather than diagnosis, a limitation the authors flagged themselves, since two athletes with the same score but different impairments may respond differently to the same training.

Together, the studies support a practical reading rather than a strict cutoff: release velocity ultimately drives distance, trunk function is a major but not sole driver of that velocity, and any single metric alone misses part of what a paired trunk-and-arm measurement captures.

This protocol earns its place in a season only when it produces a trend a coach and, where relevant, a classifier can both read.

• Monthly: run the full two-condition protocol as a dedicated session, since the arm-only condition needs its own warm-up rather than folding into normal throw practice.
• Every throw practice: log full-effort trunk angular velocity and wrist release velocity from competition-style throws as a lighter trend check between full sessions.
• Before a seat or strap change: re-run the full protocol, since backrest height and strap placement change how much trunk motion is mechanically available and shift the ratio independent of any real change in ability.
• Ahead of a classification review: compile several sessions of ratio, strap force, and trunk velocity trend into a summary sheet, since a consistent trend helps a coach discuss an athlete's functional profile with more than a single meet's distance.

Key References

• Chow, J. W., & Mindock, L. A. (1999). Discus Throwing Performances and Medical Classification of Wheelchair Athletes. Medicine & Science in Sports & Exercise, 31(9), 1272-1279.
• Bernardi, M., Guerra, E., Di Giacinto, B., Di Cesare, A., Castellano, V., & Bhambhani, Y. (2010). Field Evaluation of Paralympic Athletes in Seated Sports: Implications for Classification. Medicine & Science in Sports & Exercise, 42(5), 1017-1025.
• Beckman, E. M., Newcombe, P., Vanlandewijck, Y., Connick, M. J., & Tweedy, S. M. (2014). Novel Strength Test Battery to Permit Evidence-Based Paralympic Classification. American Journal of Physical Medicine & Rehabilitation, 93(10), 830-848.