Shooting parameters in young archers: Taha 2017

Thirty-two young Malaysian archers, 24 boys and 8 girls aged 13 to 20, shot one end of six arrows at 50 metres wearing four sensors. The measures were pelvic sway, movement of the bow hand, electrical activity of the finger extensor in the bow arm and the finger flexor in the drawing hand, and resting heart rate taken before the warm-up. The six-arrow score ranged from 17 to 57, with a mean of 44.125; resting heart rate ranged from 61 to 101 beats per minute, with a mean of 84.

A cluster analysis split the archers into 20 "high performance" and 12 "low performance". Between the two groups, pelvic sway was lower in the better archers (t = 4.19, p = 0.001) and the score was higher by 9.93 points on average (p = 0.010). For the finger extensor the authors report p = 0.050; flexor activity, bow movement and resting heart rate did not differ. A discriminant analysis placed 83.38% of the archers in the correct group with three variables and 90.63% with two.

The archers were on the Terengganu Sports Council development programme, with 3 to 6 years of shooting, and used their own bows and arrows; the bow type is not reported. After four practice arrows they shot the six scoring arrows at 50 metres on a range set up at Universiti Sultan Zainal Abidin. The Shimmer sensors were an accelerometer on the pelvis, a sensor held by a glove on the bow hand, EMG electrodes on the left finger extensor and the right finger flexor, and a heart rate sensor at the wrist. Data went over Bluetooth to an Android phone at 51.2 samples per second during stance, aiming and release.

The analysis first grouped the archers with a hierarchical cluster analysis on all the variables, score included; it then used discriminant analysis to check how well the same variables recovered the two groups, and finally compared the groups with t tests.

The weakest step is circular: the two groups are built from the same variables, score included, that are then used to recognise them, so the classification percentages say little. In the table, moreover, 18 plus 9 correctly classified archers out of 32 makes 84.4%, not 83.38%. The p values do not match the t values reported: with 30 degrees of freedom a t of 1.98 gives about 0.057 and a t of 2.02 about 0.052, so the extensor result, presented as significant, is above the threshold and appears to have been swapped with the flexor result. The test is described as paired, but it compares two separate groups.

Six arrows are few for classifying an archer. The units of the sensor measures are not given (extensor activity has negative values), and EMG at 51.2 samples per second only shows a rough trend. A mean age of 17.0 with a standard deviation of 0.4 is not compatible with a range of 13 to 20. The study is cross-sectional: it shows that those who swayed less scored more in that end, not that swaying less makes an archer score more. The same 32 archers appear in another paper by the group, on resting heart and breathing measures: in the registry they count once.

The finding that holds up best concerns body sway: in the archers who scored more the pelvis moved less during the shot, and the difference was clear even in a small sample. It is in line with other studies of postural stability in archery.

The idea that the better archers grip the bow less, which the authors read into the extensor activity, rests on a result above the threshold and on a roughly recorded signal. Resting heart rate did not separate the two groups, and nor did movement of the bow hand. For a young archer or their coach the study does not point to an exercise or a measure to chase: it describes differences seen in a single end.