Fixed or changing archery distance drills: Saing 2022

In an Indonesian study of 20 archers, training at changing distances improved scores at 30 metres more than always shooting at the same distance. The 20 archers, drawn at random from a pool of 58, were split into two groups of 10 with similar starting scores.

On the final test of 36 arrows at 30 metres, the fixed-distance group went on average from 275.7 to 291.2 points (+5.6%) and the changing-distance group from 274.7 to 302.0 (+9.9%). In points, the mean gain was 15.5 in the first group and 27.3 in the second, and the authors report a statistically significant difference in favour of changing distance. All twenty archers improved: the smallest gain was 11 points, the largest 38.

The authors describe the design as a two-group comparison with a test before and after. All 20 archers took the initial test, were ranked by score and assigned to the groups in an A-B-B-A pattern so that both started level. One group trained by always shooting at the same distance; the other varied the target distance during training, in no fixed order.

The test is the Archery Accuracy Test: 36 arrows at 30 metres, scored as a total, with a test-retest reliability of 0.825 according to the authors. The groups were compared with t-tests, first within each group and then between the groups; the table for the latter shows the percentage improvements. The paper does not state how many weeks the training lasted, how many sessions there were, which distances were used or what kind of bow the archers shot.

The journal has questionable editorial standards and peer review is not stated. The text contains inconsistencies that affect the numbers themselves: for the changing-distance group the abstract and conclusions give t = 9.256 with a critical value of 2.447, while the table gives t = 15.051 with a critical value of 2.262, and only the second value matches the individual scores published in the paper. The between-group comparison is reported with 12 degrees of freedom, whereas two groups of 10 give 18 (and the critical value used, 2.101, is the one for 18); the table presents it in terms of percentage improvement, but the reported t of 5.723 is the one obtained by comparing the gains in points. One sentence of the results was left in Indonesian.

The information needed to repeat the study (duration, frequency, distances) is missing, and there is no group that did neither kind of training, so it is impossible to say how much of the improvement came from the training and how much from simply repeating the test. The test is shot at a single distance, 30 metres, and says nothing about competition distances. The archers' age, level and bow type are not reported. Crossref lists only the university department as author: the authors' names come from the PDF.

The idea being tested comes from motor learning research on varied practice: changing the conditions of an exercise rather than repeating the same task every time. Here the archers who changed distance gained almost twice as many points at 30 metres. With these inconsistencies in the numbers, and without knowing what the training actually involved, the result is weak evidence rather than a method to copy.

The cautious reading for a coach is that, in this study, alternating distances did not hold anyone back, and the group that did it improved more. Anyone who wants to try it can do so over a set period and measure the effect at the competition distance they care about, which is exactly the measurement missing here. In the registry, a study in a journal like this one never decides the status of a claim on its own.