Shot timing of world champion archers: Takai 2009 study

From video of the 2007 outdoor World Championships in Leipzig, the authors timed four phases of the shot for the men's and the women's recurve individual gold medallists and for all of their opponents, from the round of 64 to the final: seven matches each, 12 arrows per match. Of all the phases, the only one that tells the champions apart from their opponents is the aim, that is the time from anchoring to the release.

The men's champion aimed for between 2.65 and 4.37 seconds depending on the match, and in six matches out of seven the variability of his aiming time was lower than his opponent's. The exception is the round of 16, where not only the aim but every phase showed greater variability than his opponent's: he won that match by a single point, 109 to 108, and it is his lowest score of the tournament. In the other matches he scored 111, 112, 111, 112, 115 and 110 against opponents between 102 and 108.

The women's champion did the opposite in terms of duration: she aimed for between 0.61 and 1.24 seconds, while her opponents were between 2.03 and 4.60. For that very reason the relative variability of her aiming time came out higher than her opponents' in all seven matches: over durations that short, a tenth of a second counts for a lot. Her scores were 107, 106, 107, 110, 108, 109 and 108.

In the comparison of each athlete with themselves from round to round, aiming time is the one phase that does not change. In the men's champion the phase from the buzzer to raising the bow varies, and so does the one from raising the bow to anchoring; in the women's champion those two vary and so does the phase from the release to the end of the follow through. The aim, in both of them, keeps the same mean duration from the round of 64 to the final.

This is a case study of two athletes, with their fourteen opponents as the comparison. The event is the 44th Target World Championships, held in Leipzig from 7 to 15 July 2007, recurve division, individual events, all matches from the round of 64 to the final. In those years a match was 12 arrows, with 40 seconds per arrow and 30 when the athletes shot alternately.

The shot, usually described in eight steps, was grouped into four phases following the theory of movement phases: from the buzzer that opens the time to raising the bow, meaning the hands coming up before the draw; from raising the bow to anchoring, defined as the index finger of the string hand fixed under the chin or close to it; from anchoring to the release, which is the aiming time, described as the time during which the sight is used to align the aiming point with the centre of the target; from the release to the end of the follow through.

A video camera running at 30 frames per second, set in the middle tier of the stand, followed each athlete full length. The footage was played back frame by frame and the times were taken from it. For each phase the authors calculated the mean duration and the coefficient of variation, that is how much the time fluctuates from arrow to arrow relative to its mean. The comparisons between athlete and opponent, and those of each athlete from one match to another, were made with analysis of variance and Ryan follow-up comparisons, with a threshold of 5 per cent. The match scores come from the international federation's official results.

There are two athletes. The study is declared a case study and does not claim to generalise, but those who cite it often do: the aiming time of one men's and one women's gold medallist does not give a range that holds for elite archers, and the two cases point in opposite directions on duration.

The design confounds two things. The fourteen opponents are not a control group chosen to resemble the champions: they are the people the champions happened to meet, one per round, and in the round of 64 you meet whoever qualified worst. That the champions have a steadier aim may follow from their being champions or from their having been drawn against weaker opponents in the early rounds. The times of the men's champion's opponents, for that matter, follow no pattern: in the later rounds two of them aimed for less time than he did, 1.70 and 2.11 seconds against his 4.37 and 3.95.

The stopwatch is video at 30 frames per second, one image every 33 milliseconds, and the phases are identified by eye on that video; the paper does not say who did the reading, whether it was repeated, or how closely two independent readings agree. On a phase that lasts six tenths of a second for the women's champion, reading error is a far from negligible fraction of the measurement.

Reading the round of 16 as an effect of stress is an interpretation: the authors put forward processing efficiency theory and then immediately concede that assessing psychological stress in a real match is difficult and that more cases are needed. No measure of anxiety, heart rate or anything else was collected. In the same way, the idea that a short aim gives wind and rain less to work on is an argument, not a finding of this study: the weather in individual matches is not recorded.

Finally, there are no scores for individual arrows, only the match total: a long-aimed arrow cannot be linked to what it scored. And the aiming time that the coaching manual cited by the authors gives as ideal, one to three seconds, is contradicted by the men's case, who aims for more than three seconds in five matches out of seven and wins the tournament.

This is one of the few pieces of work that measures the shot in real competition, at a World Championships, on the people who won. What comes out of it is that there is no correct duration for the aim. Two world champions at the same tournament sit on different scales: he between 2.65 and 4.37 seconds, she between 0.61 and 1.24. Anyone looking up their own aiming time in the literature finds two incompatible answers here, and that is the information.

What the two have in common is something else: aiming time did not change from the round of 64 to the final, while the other phases did. The phases that varied from match to match in both champions are the ones before anchoring, that is the part of the shot where time is taken or lost; the part from anchoring to the release stayed the same at every round. This is an observation on two people rather than a rule, but it shows where the authors saw the steadiness.

The other useful episode is the men's champion's round of 16: the match won by a single point is also the one with the lowest score and the least regular shot across every phase. The authors cannot say whether the irregularity came from the pressure or the other way round, and they say so.

An archer takes from this neither a duration to adopt nor advice on what to correct. What they take is that aiming time is a measurement that can be lifted from a video, and that, in these two cases at least, the people who won repeat it identically round after round, whatever its length.