Arm forces in khatrah archery: Ariffin 2020 study

A single archer, world horseback archery champion in Korea in 2014, shot ten times with a 45 lb traditional composite bow using the Khatrah technique, the thumb release in which the wrist is adducted and the hand punches forward, rotating the upper end of the bow. During the shots ten muscles of both arms and the trunk were recorded with surface electromyography, and the five best trials were averaged and split into six phases: stance, nocking, pre-draw, full draw, release and follow-through.

On the bow arm the deltoid is the muscle that takes the most load, and it takes it all at once: it goes from 31 newtons in the stance to 132 at nocking, then 527 in pre-draw, 759 at full draw and 800 at release, falling back to 734 in the follow-through. The flexor carpi radialis of the same arm reaches 176 newtons at release, and this is the figure on which the authors rest their conclusion that with the Khatrah technique this is the most loaded muscle.

On the drawing arm the picture is different: the triceps is already at its highest in the stance, 387 newtons, and declines slowly to 357 in the follow-through, and the pectoralis major does the same, from 236 to 189. What rises instead is the latissimus dorsi, from 10 to 170 newtons between stance and release, and the middle trapezius, from 74 to 164. Averaged across phases, the bow arm starts less loaded than the string arm (102 newtons against 133 in the upper arm) and ends far more loaded (341 against 176), while the trunk stays busier on the string side.

The signals were collected with Delsys DE-2.1 differential sensors and the EMG Works 4.0.7 software, with electrode placement following the SENIAM recommendations. The ten muscles are medial deltoid, triceps brachii, biceps brachii and brachioradialis for the upper arm; flexor carpi radialis, extensor carpi ulnaris and extensor digitorum communis for the forearm; latissimus dorsi, pectoralis major and middle trapezius for the trunk. The signal was full-wave rectified and band-pass filtered between 20 and 450 hertz, sampled at 1,000 hertz and cleaned with a Butterworth filter, then normalised to each muscle's maximal voluntary contraction and smoothed by spline interpolation. The archer, 75 kg and 170 cm, warmed up with stretching and rested between trials.

The step that produces the newtons is the delicate one. The authors multiply the normalised value, that is the fraction of maximal contraction, by a maximum force worked out as the product of the muscle's physiological cross-sectional area and the specific tension of the tissue, 100 or 62 newtons per square centimetre depending on the muscle group. The areas and tensions are not measured on the archer: they come from three cadaver studies in the literature, chosen on the basis of weight and height. The trials were shot with one bow and one technique, with no target and no distance stated.

One archer, five trials kept out of ten, no standard deviations, no statistical test: the study describes a single case and says so. The forces in newtons add nothing to what the electromyography already says, because they are the same curve multiplied by a constant taken from anatomical tables; and that constant puts a ceiling on the values, which shows clearly in the deltoid of the bow arm, where the 800 newtons at release are 98% of the theoretical maximum of 820. Near-maximal deltoid activation with a 45 lb bow ought to raise a suspicion about the reference contraction, which the paper does not describe muscle by muscle.

The six phases are not bounded in time, which makes the figures for the first one hard to read, since the string arm comes out already at its peak there. On the forearm, surface electromyography also picks up the signal of neighbouring muscles, and here they are many and thin. There is no target, no distance and no score: where the arrows went is unknown, so nothing connects these values to the outcome of the shot. Fatigue and injury risk, which the paper discusses in several places, are not measured: there are no repeated trials to exhaustion, no reported pain and no follow-up. Finally, the reasoning the authors use to explain why the flexor carpi radialis comes first does not hold anatomically, since that muscle flexes and abducts the wrist and does not move the thumb and index finger. The journal is flagged among those with questionable editorial standards.

The study is about a traditional composite bow shot with the thumb, not the Olympic recurve, and it says nothing about where the arrows end up. What it shows is the shape of muscular effort through the shot in one very strong archer: on the bow arm the shoulder's load rises sharply between nocking and pre-draw and stays high until after the release, while on the string arm the latissimus dorsi and the trapezius peak towards the release and the triceps is working from the stance onwards.

For anyone shooting with a technique of this kind it is a pointer to where to look, not a measurement to set anything by: the bow arm's shoulder and the wrist that punches forward are the two places that take the most load, and they are also the ones traditional archers complain about most often. From here, though, nothing follows about how much load is too much, or which technique tires the archer least, because fatigue was not measured and there was one person. The newtons printed in the tables in particular should not be read as the force the muscle actually exerts: they are the percentage of its maximal contraction placed on a scale borrowed from cadaver studies.