Field fitness tests

What gets measured

A field fitness test is one that can be run in a gym or beside the shooting line with very little equipment: a handgrip dynamometer, jumps, push-ups and sit-ups in a minute, the shuttle run from which VO2max is estimated, standing on one leg, a plank hold, pegboards for finger dexterity. It is the most widely used method in the physical training domain, mostly after 2015 and in archers from Indonesia, Malaysia and Turkey, many of them young. The two oldest studies were already looking at the strength the bow asks for: in 1967 49 beginner students drew a modified bow connected to a tensiometer, and in 1981, in six Japanese university archers, actual draw weight rose with shoulder and arm strength (r = 0.73 pushing, 0.50 pulling).

Next to the score

Almost every study sets the tests beside the score in a single session. Grip strength is the clearest example. When levels are compared, the better archers grip harder, but they are also bigger: in twelve young Malaysian archers the top six had 35.2 against 23.8 kg in the left hand and were twelve centimetres taller, and in a hundred Indian archers the experienced ones had three or four kilos more grip together with ten centimetres more height. Within a homogeneous group the link disappears: in twenty archers under thirteen grip did not predict the score (p = 0.925). The entry on grip and level stays at "reported".

The second limit is that the tests resemble one another. In 31 young Malaysian archers one-leg balance correlated with the score at 0.906, but strength, balance and endurance correlated with each other at between 0.83 and 0.91, and in the model with all three, balance added nothing (beta 0.129, p = 0.312). Of the nine samples that have set static balance beside the score, seven find nothing, and the entry on static balance is "contested". With many tests and few archers an isolated link can also arise by chance: the vertical jump rests on a single study of fifty young archers, with a beta of 0.39, while the standing long jump, a closely related quality, stayed below the threshold.

Then there is age, because in young archers strength, aerobic fitness, dexterity and score all grow together. Fine dexterity is linked to the score in 38 Turkish children and teenagers aged 9 to 15 (r = 0.39) and not in 45 adult traditional archers. Aerobic capacity splits the studies down the middle: two groups at around 46 ml/kg/min find it linked to the score, two groups of younger archers do not. In all four, VO2max is estimated from a field test rather than measured, and none accounts for age, height or experience.

Batteries for picking talent

Many studies use the tests to decide which young archers look promising. The Malaysian papers from a single research group split fifty young archers into high and low potential and recognise the groups with accuracies between 82.5 and 97.5%, but the groups are built from the same measures that later recognise them, and the only test on archers held out of training involves ten young archers, one of whom is placed in the other group. The Indonesian batteries for children assign bands that say little: in Manokwari the bands are drawn from the sample itself and 45% of the boys and of the girls come out as gifted, while in Sleman no child falls below the middle band. The more carefully built batteries, judged by seven experts or repeated a day apart (r = 0.893), are consistent or repeatable, but nobody has followed the tested children over time. The entry on talent selection therefore stays at "reported" and concerns current level only. The models are described on the page on machine learning, and the article on talent identification looks at the tests from the coach's side.

One trial or several

How many times to repeat a test depends on what is being measured, which is why the entry on repeated trials is "contested". For draw strength the 1967 data are clear: the best of three trials and the mean of the other two correlated at between 0.91 and 0.98, so a single trial ranked the archers almost as well as three. For the electrical activity of muscle the picture changes: in twenty experienced archers a single measurement reached 0.745 and the mean of five 0.936, but trial after trial the signal fell from 5.87 to 4.64 mV, and the hip angle weighed on the measurement almost four times as much as the differences between archers. For grip, the plank and one-leg balance, the studies on this page calculate neither an agreement index nor the typical error on their own sample, at most quoting them from other work. And a correlation says that two measures rise and fall together, not how far a single trial can stray.

Instruments that reproduce the draw

Generic tests measure strength some way from the shot itself. The simplest way to get closer is a bow scale: in the eighteen Indonesian recurve archers aged 13 and 14, strength measured by drawing the string averaged 15.47 kg, but the only tests linked to the score were the two full-draw holds with the fingers (rho 0.748 with the right hand and 0.632 with the left); another study points the other way, and the entry on holding time at full draw is "contested".

The next step is the instrumented riser. An Indonesian group started from a survey of twelve nationally certified coaches, who ran the holding test with bows of different draw weights from one centre to the next. The result was the Holding Bow Digitec Test, a 2.9 kg wooden riser on an 80 kg spring read by a load cell. In 33 adult archers it agrees with the manual test (r = 0.939 for strength and 0.974 for holding time), with almost identical means, 18.18 against 18.54 kg, and a repeat test gives 17.94 kg both times. Later users, such as the study on free weights, carried over the 2022 coefficients without measuring them again. The entry on the instrumented riser stays at "reported"; the one holding that specific strength is measured well only with an instrument that reproduces the draw is "not studied", because nobody has compared the riser with the dynamometer or with push-ups. Older still is the 1993 sound cue, which warned six archers when their draw crept forward: it is described with the other force sensors.

The trunk: a test and a direct measure

Only one study has set the trunk tests beside a direct measure of the muscle: in fifteen young elite archers assessed with ultrasound, the side plank hold on the non-dominant side went hand in hand with how much the back muscles at L5 thicken on contraction (r = 0.66 and 0.75), while the other two tests were linked to nothing. Twenty-four correlations were calculated without correction for multiple comparisons, and the entry on clinical trunk tests stays at "reported".

What to check

• whether the test was repeated, with how much rest, and whether there is an agreement index rather than just a correlation;
• whether VO2max was measured or estimated from a shuttle run;
• whether age, height and body mass enter the comparison, especially in young archers;
• whether the paper reports the correlations between tests: above 0.9 between different tests, they are measuring almost the same thing;
• where the high and low potential groups come from, and how many arrows make up the score, which in several studies is six.

How design, sample and journal are weighed is explained in How we read a study; the line on the physical profile tells the findings in chronological order, and the exercises are in the article on specific strength.

@misc{campagna2026_enresearchmethodsfitnesstestsarchers,
  author = {Campagna, Matteo},
  title = {Field fitness tests},
  year = {2026},
  howpublished = {Research, matteocampagna.com},
  url = {https://matteocampagna.com/en/research/methods/fitness-tests-archers},
  note = {accessed 2026-10-16}
}