Draw before the clicker: Leroyer 1993 study

In eight archers, the final draw before the clicker was steadier in the elite archers and, for each archer, on the high-scoring shots. In the last seconds before the clicker the drawing hand first moves back quickly and then, over the final 5 seconds, slows to about a millimetre per second, covering about 5 mm over those 5 seconds. The authors measured how far this slow movement strayed from a steady progression: the deviation was smaller in the five elite archers than in the three university-level archers and, in all eight, smaller on high-scoring shots than on low-scoring ones.

In a second test, seven university archers held the shooting position until exhaustion, for 52 seconds on average. The mean frequency of the posterior deltoid EMG fell steadily, on average by 0.72% of its initial value per second, a classic sign of fatigue. Both the muscle signal and the bow signal showed a peak between 8 and 12 Hz, with a coherence between the two of 0.81 to 0.96: the muscle tremor was being passed on to the bow. The bow also showed slower oscillations, below 5 Hz, which did not change with fatigue.

Finally, six archers shot two series of 36 arrows, one normal and one with an audible signal that warned them whenever the drawing hand crept forward beyond a threshold. With the signal the draw became steadier in five archers out of six (in the sixth it got worse), but scores did not improve significantly.

A displacement sensor (a linear variable differential transformer, LVDT) was fixed to the riser of each archer's bow with a 150 g carbon bracket, and an iron core attached to the lower limb moved inside its coils. The archers adjusted their stabilisers to keep the usual feel. Close to the sight window the limb deflects only slightly, so the signal could be converted into millimetres of drawing-hand movement along the arrow's axis, with a scale factor for each archer. A switch triggered by the clicker marked the moment of the click.

In the first test, five elite archers aged 20 to 25 and three university-level archers aged 20 to 22, all men and using their own bows, shot 36 arrows at 90 metres in 12 ends of three arrows in 150 seconds, with 4-minute breaks. For each shot the signal over the last second and a half before the clicker, sampled at 70 Hz, was fitted with a straight line, and the mean distance of the points from that line became the measure of irregularity. High- and low-scoring shots were those more than one standard deviation above or below each archer's mean.

In the second test, seven university archers aged 19 to 33 held the shooting position until exhaustion, at about 40-50% of their maximal force measured on an ergometer. Surface EMG of the posterior deltoid and the bow signal were recorded together; the mean frequency of the EMG spectrum was followed over time, and the coherence between the two signals was calculated from 0 to 50 Hz. For the feedback test, with the signal taken from the bow and filtered below 5 Hz to exclude tremor, six archers shot the two series on different days.

The numbers are small: eight archers in the first test and seven in the second, all men, and six in the third, whose sex and level the paper does not give. The comparison between elite and university archers rests on five archers against three, and the link between steadiness and score is an association: it does not show that making the draw steadier raises scores. The only test in which steadiness was actually changed, the one with the audible signal, did not improve scores. The paper reports the score distribution of a single archer (a mean of 7.25 per arrow), so it is not clear how far apart the two levels were in points.

The sensor added 150 g to the bow; the archers said it did not bother them, but the effect was not measured. The fatigue test required holding the draw for almost a minute, far longer than a normal shot, so the tremor measured there need not be the same size in competition. The four ways the authors propose to reduce the effect of tremor (a less powerful bow, more maximal strength, a damper designed for tremor, and a single-motion shot that shortens the hold) are hypotheses that were not tested.

It is among the studies most often cited by the other papers in the collection. For archers, the point is that in the last seconds before the clicker the drawing hand is not still: it moves back slowly, by a few millimetres, and in the better archers, as on each archer's better shots, it does so at a more constant speed. The study does not say how slowly one should draw, nor that a steady draw is enough to score well.

The fatigue test involves holds far longer than a normal shot, but it points to a mechanism: as the deltoid tires, arm tremor reaches the bow at 8 to 12 oscillations per second. The audible signal made the draw steadier without adding points, and the authors put this down to the extra information the archer has to process while aiming. It is a useful warning for anyone using feedback devices on the draw in training: the measure can improve while the score does not.