Fatigue and archery shooting: Squadrone 1994

Twelve archers from the Italian federation were split by FITA score into seven of intermediate level (1180 to 1300 points) and five of high level (above 1300). Each shot ten arrows at 12 metres, then completed a fatigue protocol, then ten more arrows. Throughout, the kinematics of 23 markers were recorded together with surface EMG from the finger flexors, the biceps of the drawing arm and the back muscles, ground forces on a platform, and the instants of the click, of the fingers leaving the string and of the arrow leaving the string.

The score does not change: the authors find no significant variation after fatigue, in any archer. What changes is how the bow is held. Lateral bow sway increases in all twelve, from a minimum of 6% to a maximum of 39%, 20% on average, and the increase is more marked in the less skilled: between FITA score and increase in sway the correlation coefficient is −0.65 (p < 0.05). Those who shot less well also lengthened the aiming phase after fatigue, with a correlation coefficient of 0.81 (p < 0.05).

The EMG shows that each archer repeats their own muscle pattern before and after fatigue, with a general rise in signal amplitude. In the group means the largest rise is in the biceps, more at intermediate level (16%) than at high level (12%); in the finger flexors it is 9% against 7%, and across the four trapezius channels it runs from 7% to 14%. The individual values, though, vary widely: two archers of the same high level show biceps rises of 21% and 5%, with some negative values on other muscles. Five archers out of twelve, three intermediate and two high level, activated the left-side back muscles more while holding at full draw, and in them the authors report a visible link between the alignment of the shoulder line with the bow arm and the rise in signal.

On the platform, some archers show significant changes in the length of the centre-of-pressure path in the second before release, a sign of faster corrections, but with no correlation to level; centre-of-pressure displacement along the two axes does not change, either individually or by group.

Data were collected at 12 metres from the target, with ten shots recorded before and ten immediately after the fatigue protocol. The protocol asked archers to come to full draw and remain in the aiming phase for 20 seconds, for six sets of ten repetitions with 20 seconds of rest between sets: sixty 20-second holds in all.

Kinematics, EMG and force platform were acquired together online and processed with the ELITE motion analysis system. The three-dimensional coordinates of 23 retroreflective markers 10 millimetres across were sampled at 100 Hz, with markers on the temporal bone and the mandibular joint for the head, the shoulders, elbows, wrists and hands for the arms, the iliac crests, knees, ankles and third metatarsal heads for the lower limbs, plus three markers on the spine and three on the bow.

Surface EMG was collected from the finger flexors and the biceps of the drawing arm and from the upper and lower back muscles, at 1000 Hz. The signal was full-wave rectified, integrated over 100-millisecond intervals and normalised to values obtained in standard maximum isometric actions. Ground reaction forces and centre-of-pressure displacement were measured with a Kistler platform at 1000 Hz and analysed with dedicated software that reconstructs the centre-of-pressure path and computes summary statistics.

To synchronise the variables with the phases of the shot, electrical switches were used to detect the closure of the clicker, the fingers leaving the string and the arrow losing contact with the string, sampled at 1000 Hz. The authors present their results as a brief synopsis: there are two correlation coefficients with their significance level, two tables of mean percentage EMG changes, and qualitative descriptions for the rest.

This is a four-page conference paper from 1994, and the results are declared by the authors themselves to be a brief synopsis. Almost all the numbers needed to assess them are missing: of bow sway only the percentage increases are known, not the values in millimetres; of the scores, which according to the authors do not change, no value is reported, before or after; of the EMG changes there are mean percentages but no standard deviations and no statistical tests. The two correlations are the only inferential results presented.

Those two correlations rest on twelve points. With twelve observations the 5% significance threshold sits around 0.58, so −0.65 and 0.81 clear it only narrowly and with a very wide margin of uncertainty: they are hints from a minimal sample, not stable estimates. And the second has a sign problem. The text says the less able archers lengthened the aiming phase, which would give a negative correlation between FITA score and the increase in duration; the coefficient reported, however, is a positive 0.81. One of the two is a misprint, and the paper does not make clear which.

There is no control condition. The ten arrows after always come after the ten arrows before, so the effect of fatigue is added to that of order and of arrows already shot, and no group repeated the sequence without the protocol in between. Nor is there any check that fatigue was actually produced: no measured loss of force, no scale of perceived effort.

The protocol itself is far from competition fatigue. Sixty 20-second holds at full draw, in six sets, are a prolonged isometric load; in competition archers perform many draws with release, over several hours, with breaks between ends. The two kinds of fatigue are not equivalent, so how much of this result transfers to a competition day remains open. The distance, 12 metres, also matches no competition format, and with the target that close a worsening of the hold shows up less in the score than it would at 70 metres: one more reason why the absence of change in the scores should not be read as absence of effect.

On the muscles, the channel labels are unclear. The methods say upper and lower back muscles, while the table reports four trapezius channels called «descendens» and «superior» on the right and left, which usually denote the same portion: which portions were recorded cannot be told. The group means then hide individual differences that the authors themselves show: between two archers of the same level the biceps rises 21% in one and 5% in the other, with negative values on other muscles. The link between shoulder alignment and the rise in signal, observed in five archers, is described as evident but not measured.

Finally, the recommendation in the conclusions, to train the more critical muscle groups for strength and endurance, does not come from these data: no training intervention was tested.

The result worth keeping is the decoupling between score and hold. After fatigue these archers' scores do not change, while lateral bow sway increases in all twelve, by 20% on average and up to 39%. That means fatigue shows up in the quality of the aim before it shows up in the result: an archer can close an end on the same points as before while already compensating for a less steady bow. It is why judging your own tiredness from the score is a crude yardstick.

The second point is that level protects. The increase in sway ran from 6% to 39%, with the low values among the stronger archers and the high ones among the intermediates, and the correlation with FITA score is −0.65. The same holds for the biceps of the drawing arm, which in the group means rises 16% at intermediate level against 12% at high level. Read together, the two findings point the same way: under fatigue the less experienced archer tends to add arm, and the bow steadies less.

Care is needed over how far this generalises, though. The protocol was sixty 20-second holds at full draw, a concentrated isometric load, not a competition day with many draws and releases spread over hours. And the shooting was at 12 metres, where errors in the hold weigh less than at 70. The mechanism the study illustrates is plausible; the precise numbers do not transfer.

There remains a pointer about how to watch yourself. The study shows that each archer's muscle patterns are highly repeatable, before and after, while the amplitude changes, and that two archers of the same level react to fatigue differently, one with the biceps rising 21% and the other 5%. Group means say little about this: what changes under fatigue is largely individual. On the platform, finally, fatigue did not move the centre of pressure, but in some archers it lengthened its path in the second before release: more corrections, not a different stance.