Free weights or machines for archers: Subekti 2023

Twenty male archers aged 12 to 15 from the Bantul district of Java did six weeks of weight training in two ways: ten with free weights, dumbbells in hand, and ten on machines. The only measure is arm holding time in seconds on the Holding Bow Digitec Test, the Indonesian device that has the archer work against a resistance in a position close to the draw.

The free weight group went from a mean of 15.3 to 17.7 seconds, that is, 2.4 seconds more, an increase of 15.69%. The machine group went from 15.1 to 16.4 seconds, 1.3 seconds more, an increase of 8.61%. Both changes clear the statistical threshold, the first with a value of 7.856 and the second of 4.333. Comparing the two gains with one another, the difference of 1.1 seconds in favour of free weights clears the threshold in turn, with a value of 2.569.

The individual data are published, and they are all whole numbers: in the machine group three boys out of ten do not change by a single second, while in the free weight group the poorest result is a gain of one second. The authors conclude that both methods increase arm endurance and that free weights do better, adding that machines remain suited to a school setting with little supervision because they do not require a partner to act as a spotter.

No measure concerns shooting: no arrows, no score, no accuracy.

The design is a before-and-after on two groups, with no third group that trained nothing. The twenty boys make up the entire population considered, that is, every male archer of that age band in the district. After the pre-test they were ranked by score and paired in an A-B-B-A pattern, a way of distributing the best and the poorest equally between the two groups: it is not a draw, but it neutralises the starting level.

Training lasted six weeks over eighteen sessions, in the form of a circuit of eight stations. The free weight group did dumbbell presses, one-arm dumbbell presses, lateral raises, seated bent-over rows on an incline bench, rear delt work, front raises, side-lying lateral raises and single-dumbbell front raises. The machine group did low pulley lateral and front raises, rear delt reverse cable flyes, lateral raises on a Nautilus machine, pec deck rear delt laterals, shoulder presses, face pulls and standing lat pull-downs.

The outcome test is the Holding Bow Digitec Test, whose published validation the authors cite (0.895 against a critical value of 0.344 for holding time, reliability 0.944): it is the same device described in two other sources in this registry. The analysis uses paired t-tests within each group and an independent-samples t-test on the gains, after normality checks with Shapiro-Wilk and homogeneity checks with Levene's test, in SPSS 21.

Everything else about the training is missing: there is no record of the load, the number of sets and repetitions, the recovery between stations, the length of a session, or whether and how the load was increased over the six weeks. Nor is there any record of what the boys did with their own bows during those six weeks.

The measure is a timed hold, and the published data are all whole seconds, so the resolution of the instrument, as used here, is one second. On that scale the main result, a difference of 1.1 seconds between the two groups' gains, is worth little more than one step of the measurement: it takes only a few boys rounded one way to change it. And a mean gain of 2.4 seconds on 15.3 is a small number in absolute terms, whatever percentage is set beside it.

The content of the training is missing. Without loads, sets, repetitions and progressions there is no way to know what was compared: these could be two programmes of different demand, in which case the difference would concern the dose and not the equipment. A group that did not train is missing too, so the effect of the weights cannot be separated from that of six weeks of shooting and growing: the participants are 12 to 15 years old, an age at which strength changes on its own.

The sample is twenty boys from a single district, all male and all adolescent, so it says nothing about female archers or adults. It is not stated whether the assessors knew which group each boy was in, and there is no follow-up: whether the gain lasted is unknown.

The outcome is a hold, not a shot. No arrow was shot in the measurements, so the study does not allow anyone to say that free weights make you shoot better, or that a longer hold helps in competition: that step, present in the paper's introduction, remains a hypothesis borrowed from the literature. The discussion, for its part, leans almost entirely on studies of squats, leg presses and hormones in non-archery populations, that is, on lower-limb exercises that have little to do with holding at full draw.

One last note concerns the source record: in the registry the fields for design, number of participants, level and language are empty, while the paper states all of them (a before-and-after on two groups, twenty participants, youngsters aged 12 to 15, text in English). On Crossref the first author differs from the one on the publisher's page, as the record already notes.

The question the study asks is practical and sensible: if you are going to put young archers through weight training, dumbbells or machines? The answer it offers, on a timed hold and on twenty boys, is that the gain was slightly larger with dumbbells. It remains a gain of a few seconds on a test that is not shooting, over six weeks.

The observation that holds up best is the one the authors make at the end, and it does not depend on the numbers: machines require no spotter, and for that reason they fit better in a school gym where one coach supervises ten boys at a time. That counts as an organisational consideration, not a measured result.

What the study does not allow anyone to say is that an arm which holds longer scores more, because no points were counted. For the registry this source counts as evidence on arm endurance in young archers, measured with the Holding Bow Digitec Test, and not as evidence on shooting performance.