Measuring Arm Velocity in the Handball Wing Shot: Separating Arm Speed from Trunk Rotation

A wing player sets up tight to the post — eight degrees off the goal line, close enough that rotating the hips another fifteen would send the shot wide — and releases at 15 m/s where a backcourt teammate hit 22 m/s off a jump throw twenty minutes earlier. The coach checks the readout, assumes an arm-speed deficit, and pencils in more med-ball work for the week. That diagnosis is very often wrong. The wing shot and the open jump throw share a label in most training logs — jump throw — but they are not the same movement with a different backdrop. The angle to the goal limits how far a wing player can rotate the trunk before the ball sails past the far post, so the release speed that would normally come from pelvis and trunk rotation has nowhere to go. It has to come from somewhere else: shoulder internal rotation, elbow extension, and wrist flexion — arm velocity, in the narrow sense used throughout this guide.

That distinction matters for anyone comparing a wing throw's velocity flat against a backcourt jump shot's. Treating a lower wing-shot number as an arm-strength problem, when it is actually a consequence of a tighter angle, sends a player into the wrong corner of the gym. This guide covers a dual-sensor protocol for separating trunk rotation velocity from arm-segment velocity on the wing shot, comparing it against the same player's own open jump shot, and what that split should look like across shooting angles.

Overarm throwing in handball follows a proximal-to-distal sequence — pelvis, then trunk rotation, then shoulder internal rotation, then elbow extension, then wrist flexion, each segment peaking a beat after the one before it. Van den Tillaar and Ettema (2007), using 3D motion capture on experienced team-handball throwers performing standing overarm throws, documented exactly this ordering and found that forearm and wrist angular velocity correlated most strongly with final ball release velocity, while trunk rotation velocity's own correlation with ball speed was considerably weaker and more variable between throwers. Trunk rotation, in that data, looked more like it was setting up timing than directly producing the number a radar gun reads.

None of that changes for a wing player. What changes is how much trunk rotation is available to spend. An open jump throw from the backcourt or the 9m line lets a player rotate through roughly 60-90° relative to the goal line, because no sideline or goal frame constrains the ball's path. A wing shot from a tight angle — commonly 5-25° off the goal line, just outside the post — physically cannot use that much rotation without sending the ball wide. The player is not choosing to under-rotate; the angle is choosing it for them. Proximal segments contribute whatever the angle allows, and the arm makes up the rest.

Define the two variables cleanly enough to track them independently, throw to throw. Trunk Angular Velocity is the peak rotational speed (deg/s) of the thorax around its long axis in the second before release, captured by an IMU strapped across the sternum. Arm/Distal Velocity is the peak linear velocity (m/s) of the wrist in the direction of the throw, captured by a second IMU at the distal forearm — a proxy for the combined output of shoulder internal rotation, elbow extension, and wrist flexion, without isolating each joint separately.

From those two numbers, the app derives an Arm Contribution Index (ACI): the wrist segment's share of total resultant release speed, as a percentage. A higher ACI means more of the throw came from the arm; a lower ACI means trunk rotation carried a larger share. Because trunk contribution scales with how much rotation the shot geometry allows, ACI should shift predictably as shooting angle tightens — and that predictable shift, not a single universal number, is what this protocol tracks.

Mount one IMU across the sternum with a chest strap, oriented so its measurement axis aligns with the transverse plane — this is the trunk sensor. Mount a second IMU at the distal forearm on the throwing side, strap parallel to the forearm bones — the same wrist-mount position used for velocity and elbow-extension tracking elsewhere in this system, so a player already wearing one sensor only needs the trunk strap added.

Calibration Sequence

1. Neutral standing hold, both sensors (3 seconds): zero-reference orientation before any throwing begins.
2. Trunk rotation range check: the player rotates as far as comfortable toward the throwing side and back, giving the app their own rotational ceiling — rarely tested under game-realistic constrained angles otherwise.
3. Wing-angle-specific hold: standing at the actual wing position, rotate only as far as the shot angle allows without missing the frame — this becomes the constrained-angle reference live throws get compared against.
4. Three calibration throws at moderate effort from the wing position, confirming release-detection on both sensors is synchronized to a shared timestamp — a green check appears once both units agree on the release frame.

The number that matters is not the wing shot in isolation — it is the wing shot measured against the same player's open jump throw, because ACI is a relative, angle-dependent metric rather than an absolute one.

Session Steps

1. Mark two throwing positions: a wing position at the player's game-realistic angle (commonly 5-25° off the goal line, right up against the post) and an open position at 60-90°, typically the 9m backcourt line.
2. Have the player complete 8-10 full-effort throws from the open position first, establishing a baseline ACI and release velocity for an unconstrained shot.
3. Move to the wing position and complete another 8-10 full-effort throws, using the player's real in-game footwork and jump — a standing wing throw and a jumping wing throw are not interchangeable data points, so keep the technique consistent within each block.
4. For each throw, log Trunk Peak Angular Velocity, Arm/Wrist Peak Velocity, the timing gap between the two peaks in milliseconds, and the resulting ACI.
5. Average each block and compute the ACI delta between wing and open throws — this delta, not either raw number alone, is what identifies whether a player's wing mechanics are using the angle correctly.

Ten reps per position is enough to smooth out mis-timed release detections without asking for more throws than a normal session can absorb; drop any throw flagged with a release-detection mismatch between the two sensors rather than trying to hand-correct it.

These bands describe how ACI typically shifts as shooting angle tightens, applying the sequencing principles documented in the cited research across shot geometries rather than reproducing a single validated lookup table — treat them as a starting reference for where a player's numbers should roughly land, not a pass/fail line.

A wing player sitting well outside their expected band in either direction is worth a closer look. An ACI far lower than the band suggests they are over-rotating for the angle available, which risks either a wide shot or a foot violation on the sideline. An ACI far higher suggests almost no proximal contribution at all, which usually shows up as a shot that depends entirely on shoulder and elbow strength and fatigues faster across a match.

Neither of the two studies behind this protocol measured a wing shot from a constrained angle directly — both point at the underlying mechanism, which is why the bands above are an applied extension rather than a direct citation.

Van den Tillaar and Ettema (2007) analyzed standing overarm throws in experienced team-handball players using an optoelectronic 3D system and found forearm/wrist angular velocity correlated strongly with ball release velocity — the strongest relationship of any segment measured — while trunk rotation velocity showed a markedly weaker, less consistent correlation with the same outcome. The limitation: the throws studied were standing and unconstrained, run in a lab rather than from a game-realistic wing angle, on a small, specific sample. The finding establishes that arm segments dominate the velocity relationship in a general throw, not what happens once shot angle is restricted.

Wagner, Buchecker, von Duvillard, and Müller (2010) compared elite and club-level team-handball players on the jump throw and found a large gap in ball release velocity between groups, driven less by raw arm speed than by sequencing — the elite group reached peak pelvis and trunk rotation velocity earlier relative to release, giving the distal segments a faster-moving base to add onto. The limitation: the jump throw studied came from a relatively open angle rather than a wing-constrained one, the comparison was cross-sectional rather than longitudinal, and the elite-versus-club-level sample was small enough that the size of the sequencing advantage reads as directional, not as a fixed number to coach against.

Once a baseline ACI is established, training decisions get more specific than more shoulder strength for everyone. A player whose ACI is already high and appropriate for their angle but whose release velocity has plateaued needs arm-segment power work — banded internal-rotation throws, overload-ball work in wrist-dominant patterns — because the sequencing is already doing its job and the ceiling sits in the distal segments. A player whose ACI keeps drifting low relative to their own wing baseline has a different problem: they are trying to borrow trunk rotation the angle will not support, and the fix is technical — footwork and release-point cueing — before it becomes a strength conversation.

Run the comparison monthly, since a wing player's usable angle can shift as footwork or jump timing changes, and a baseline built in preseason may not describe how the player is throwing by midseason.

Key References

• Van den Tillaar, R., & Ettema, G. (2007). A three-dimensional analysis of overarm throwing in experienced team-handball players. Journal of Applied Biomechanics, 23(1), 12-19.
• Wagner, H., Buchecker, M., von Duvillard, S. P., & Müller, E. (2010). Kinematic description of elite vs. low-level players in team-handball jump throw. Journal of Sports Science and Medicine, 9(1), 15-23.