Bow draw technique and balance: Kuch 2020 study

A single skilled recurve archer shot nine arrows at 12 metres using three different ways of bringing the bow to full draw, three arrows each: drawing from above, his usual technique, drawing from below, and a direct draw. Two force plates recorded the movement of the centre of pressure under the feet before the release and in the half second after it. Every arrow hit the centre.

After the release the three techniques did not differ on any measure. Before the release almost every measure was similar, and only two were lower with the draw from above than with the direct draw: the mean position of the centre of pressure in the body's front to back direction (61.7 mm against 72.1) and the sway along the shooting direction measured as root mean square distance (0.6 mm against 1.0), both at p = 0.09, below the 0.1 threshold the authors chose. The area of the ellipse containing 95% of the path was 10.5 mm² with the draw from above, 21.8 with the direct draw and 10.9 with the draw from below, with no significant difference.

Reaction time to the clicker gave a clearer picture: 161 ms on average with the draw from above and 163 with the draw from below, against 304 ms with the direct draw, which was also far more variable (standard deviation 149 ms against 34 and 22). Compared with values for quiet standing reported in the literature, the centre of pressure moved little but quickly before the release, which the authors read as the archer stiffening to make aiming easier.

The archer stood with one foot on each of two six-axis force plates (Sensix), and an ADXL354 accelerometer fixed to the riser marked the clicker fall and the release; all signals were recorded together at 1,000 Hz, with the force plate signals filtered at 6 Hz. From the centre of pressure the authors derived the standard balance measures: mean distance and root mean square distance around the mean position, mean and peak velocity, and the area of the 95% ellipse, each split by direction. For an archer standing side-on to the target, the body's side to side direction is the shooting direction.

The pre-release phase was the second before the clicker fell plus the reaction time, the post-release phase the 500 ms after the release; reaction time is the gap between the clicker falling and the release. The three techniques were compared with a Kruskal-Wallis test and Mann-Whitney pairwise comparisons, with a significance threshold of 0.1 rather than the usual 0.05, justified by the small number of arrows. The target was at 12 metres for technical reasons.

This is a three-page extended abstract on one person and nine arrows, three per condition: every comparison sets three values against three, with the significance threshold widened to 0.1. With fifteen measures compared in two phases, a difference at p = 0.09 can also arise by chance. At 12 metres every arrow hit the centre, so there is no way to link the balance measures to the score, as the authors acknowledge, and there are too few arrows to assess consistency.

The two new techniques were not the archer's usual ones, and a draw that has never been trained may be less stable and slower simply because it is new: the study does not separate the effect of technique from the effect of habit. The archer's exact level, beyond 'skilled', and his draw weight are not given.

The study does not say which way of drawing the bow is the right one. It says that, in a skilled archer, changing the draw did not upset balance before the release, and that the direct draw, which he did not normally use, went with a clicker reaction time almost twice as long and far less regular. The authors' closing recommendation, the draw from above for this archer, is the technique he had always used.

For anyone changing the way they raise and draw the bow, the most visible sign of the change in this case was the time between the clicker falling and the release, more than balance. Whether that time shortens after a few weeks of practice, and whether the same holds for other archers, the study does not measure.