Hold time and alignment in archers: Marsden 2015

Eight recurve archers from the Great Britain programme shot at 70 metres holding their aim for set times: first their usual duration, then double, then triple, 6 arrows per condition. A frontal camera followed six points on the body that describe alignment (both wrists, both elbows and both shoulders) and measured their velocity, that is how much they moved over time.

As the hold lengthened, five of the six points moved more. The wrist of the string hand went from 3.51 to 4.15 and then 4.60 centimetres per second; the bow wrist from 3.21 to 3.58 and 3.87; the bow shoulder from 3.20 to 3.62 and 3.80. Only the drawing shoulder showed no significant change (3.66, 4.07 and 3.84). The mean score per arrow fell from 8.60 to 8.13 and then 7.75, but the only statistically significant difference is between the usual hold and the triple hold.

Testing took place at the Lilleshall national centre, with the athletes' own bows and arrows. The archers (6 men and 2 women, aged 21 on average) did a 5-minute archery warm-up and then shot 6 free arrows, filmed from the side at 50 frames per second to work out each one's mean hold, timed from the string touching the face to release. They then shot 6 arrows at their mean hold, 6 at double and 6 at triple, always in that order, releasing on audible cues given at one-second intervals. Rest matched work time, with 3 minutes between ends, and at the end each archer did one maximal hold without shooting.

A Quintic camera at 150 frames per second, in the frontal plane and calibrated with a one-metre square frame, tracked reflective markers on the shoulders, elbows and wrists. For each arrow 200 frames were analysed, from one second before release to 0.3 seconds after, filtered and turned into mean velocities. Velocities were compared with repeated-measures analysis of variance and Bonferroni correction, scores with the non-parametric Friedman and Wilcoxon tests. The study is a master's dissertation in biomechanics at the University of Chester.

The order of conditions was not varied: triple always came last, so the effect of the long hold adds to the fatigue built up over the earlier ends and cannot be separated from it. The velocity statistics have 94 degrees of freedom, which shows that each arrow was counted as a separate case: the independent observations are eight archers, and significance is overstated as a result. The analysis window also includes 0.3 seconds after release, so the mean velocity mixes stability while aiming with the movement of the release and follow-through.

Hold times in seconds are not given, and the final maximal hold is not reported. There is only one camera, so movement in depth is missed. The link between instability and score is inferred from the two getting worse together, not checked arrow by arrow, and with 6 arrows per condition a difference of half a point per arrow is fragile. A hold timed by external cues is also not the same as one that drags on by itself in competition because of wind or hesitation. The text labels the elbow markers inconsistently (lateral or medial), and this is a dissertation that has not been peer reviewed.

For a high-level archer the study puts figures on a common impression: made to stay at full draw for two or three times their usual hold, these British archers lost points, from 8.60 to 7.75 per arrow at 70 metres in the most extreme case, and their wrists, elbows and shoulders moved a little more. The velocity differences are small, roughly 4 to 11 millimetres per second between the usual and the triple hold.

The study does not say what the best hold time is for any individual, nor whether letting down and starting again is better than staying on aim. The author suggests building holding endurance with isometric work or not going beyond one's usual time, but these are his own suggestions, which the test did not compare with anything else.