Before buying a new bow, better limbs or a set of competition arrows, almost every archer asks the same thing: will more expensive equipment make me shoot better, or is the limit me? Two opposite answers go round the range. Some say that with the right kit the points more or less take care of themselves; others say a champion would score the same with a club bow. No study has tested the question in the form we ask it, by giving a group of archers higher-end equipment and looking at their scores a few months later, but research gets close to it from three directions.
The first is the shooting machine, which takes the archer out of the shot and shows how tightly the equipment groups on its own. The second is the experienced archer handed a bow that is not their own. The third is feel, meaning which equipment changes an archer notices and which they do not. Taken together, the results say something fairly precise about where the spread comes from, even though each of them, on its own, rests on few studies and few shots.
How tightly the equipment groups without the archer
Shooting machines were built to test bows and arrows, and that is exactly why they show what group is possible when nobody is holding the bow. The numbers are small. In 1990 a mechanical release machine, at 30 m, spread its arrows with a standard deviation of 0.9 centimetres, enough for 99.6% of them to be expected inside the ten. In 2020 a 114 kg machine with a 40 lb recurve put ten arrows out of ten into the 122 mm ten at 70 m on a windless day. In 2025, again at 70 m but indoors, the same arrow shot thirty times landed a median 1.3 centimetres from the centre of the group, with 95% of impacts within 3.5 centimetres: a seven-centimetre circle, only just wider than the X.
These results take a great deal of care. The first version of the 2020 machine weighed 24 kg, vibrated as it drew the string and lost precision, so it was rebuilt almost five times heavier; a prototype robot with a training bow, at 18 m, put the same arrow into a 16-centimetre circle. Even with a well-built machine the group is sensitive to the details of the arrow: at 25 m, with a compound on a machine bolted to the ground, changing nothing but the nock took the group of one arrow model from 44 to 27 mm. That is the evidence on which the registry reports that the type of nock changes the size of the group, and it is a reminder that, once the archer is removed, the group is decided by how the bow is built and how the arrows are prepared.
The comparison that matters to anyone who shoots, though, is between the machine and a person in the same setting, and there is only one. In 2022 a Japanese group filmed shots at 55 m with cameras running at 3,000 frames a second, on an indoor range with the air conditioning switched off to keep the air still. An archer shot two arrows 18 times each, with a vertical spread, measured as a standard deviation, of 107 and 108 mm; the robot, with different arrows, stopped at 34, and its improved version at 9, enough in the authors' view to stay inside the X at 70 m indoors. This is why the claim that a well-built shooting machine groups more tightly than an archer is reported: plenty of consistent evidence on machines and a single direct comparison, with an archer whose level the study does not give and who shot different arrows from the robot's. The comparison has not been made with the best archers on the same equipment.
The same difference in scale explains why a machine is useful for choosing arrows. In 2020 the 114 kg machine shot six arrows at 70 m, two of them defective: the four sound arrows stayed inside a circle with a radius of two standard deviations, the two defective ones landed outside it, and always on the same side, low for the one missing a vane and to the right for the other. The claim that a shooting machine at 70 m picks out defective arrows is reported, because the test is a single one and involves six arrows. The practical point lies in the numbers in the figure: a defect that moves the impact by a few centimetres, inside a group as wide as the archer's at 55 m, blends into the rest of the spread, and spotting it when shooting by hand means numbering the arrows and following them over a good many ends.
Where the expert archer is already close to the machine
If the machine groups better, the question is where the archer loses ground, and the speed at which the arrow leaves the bow does not seem to be it. In seven elite archers from the German squads, over thirty arrows at 18 m, the average arrow speed varied from shot to shot by between 0.23 and 0.58%. In a Japanese study of arrow aerodynamics an expert archer repeated her launch speed within 0.6% at 58 m/s, against 0.3% for the compressed-air launcher built for the study, and a compound shot from a machine and timed with the NIST reference chronograph varied by about half a percentage point. The claim that an expert archer repeats the arrow's speed from shot to shot almost as closely as a shooting machine is reported: the same order of magnitude, but in the only direct comparison, which involves a single archer, her variation is twice the machine's.
The German study adds the most useful piece. In the seven archers the time from the click of the clicker to the release varied by between 3.94 and 7.25%, on average more than ten times the variation in speed, and it was this variation, not the speed's, that went with the score: on its own it accounted for 66% of the differences in average score between the archers. With seven people a relationship that strong can arise by chance, and so the claim that among elite archers, those with a more consistent release time score higher is reported. The picture fits the machine studies all the same: even in a person's hands the bow returns its energy very regularly, and the gap with the machine has to be looked for in the rest of the shot.
An experienced archer with a bow that is not their own
The evidence closest to the question of anyone thinking of changing bow comes from a 1990 study of muscle activity at different shooting distances. Eight archers with FITA competition experience shot with their own bow and with a standard bow, the same for everyone, at 18, 25, 50 and 70 m. The imposed bow was lighter than theirs at 18, 25 and 50 m, from 133.4 to 174.6 newtons against personal bows of 178.5 to 214.8, and heavier at 70 m, at 235.4 newtons. Neither muscle activity nor group precision, measured as the contour of the best three arrows out of nine, changed significantly. Nine years later, in twenty young Swedish archers, the type of bow made no difference to score, perceived effort or average muscle tension. There, though, everyone shot their own bow, so the study says nothing about what happens when the bow changes in the same person's hands.
This is why the claim that giving an experienced archer a bow other than their own does not change their precision is reported, and it needs reading with three cautions. There are very few arrows, nine per distance. Precision is the contour of the best three, so it leaves out the six worst arrows of each series. And the average hides individual cases: the archer with the weakest pull had the poorest precision both with his own bow and with the heavier imposed bow, while the strongest did better with the heavier bows.
The same study holds a finding that concerns the choice of bow more than its price. The pulling strength of the eight archers ranged from 255 to 795 newtons, their bows from 178 to 215, and there was no relationship at all between the two, so much so that the authors describe the choice as arbitrary even among accomplished archers. Together with a study of 49 university beginners, in whom maximum strength was not related to draw weight, this is why the claim that stronger archers shoot heavier draw weights is contradicted. I cover how to pick a draw weight in the article on choosing your draw weight.
Which changes the archer feels
On the range most equipment choices are made by feel, and on how fine that feel is there are only a few observations, gathered in passing in studies designed for something else. In 1998, in a study of muscle activity in the shot, one arrow in five had been lengthened by 5 mm to shift the moment the clicker went off, and none of the nine archers from the German national squads noticed. In 2001 ten kyudo archers shot with four strings of different materials, commenting after every shot: with Kevlar, which eight of them normally used, "feels as usual" and "no particular impression" were the most common comments, while with hemp, the string most different mechanically, there were more comments on vibration and shock. The study does not say, however, that the archers did not know which string they were shooting. The claim that archers notice big changes in their equipment, not small differences is reported: for small differences there is the 1998 observation, for big changes only a hint, and neither study measures a threshold.
The most careful study of feel concerns barebow. In 2023 eleven weight set-ups on the same bow were measured with accelerometers and with an inertial sensor on the riser, and rated blind, from 1 to 5, by seven club archers. The rating dropped as the amplitude of vibration rose, above all on the riser next to the button, where the correlation reached minus 0.895, and as the bow moved further sideways after the shot: the bare bow moved 139 mm sideways and scored 2 out of 5, while the two favourite set-ups, two weights with the heavier one in the middle, held it to 17-19 mm and scored 4. The claim that in barebow, archers prefer the weight set-up that leaves the bow steadier is reported, on seven judges and four shots per set-up in the movement measurements.
The same study shows why feel is hard to turn into points. An advanced archer shot 30 arrows per set-up at 50 m, and the scores, between 242 and 262 out of 300, fell within the range expected by chance, except for the bare bow, which was below it at 228; by the authors' simulation, telling two set-ups apart by score would take 900 to 1,050 arrows each. It is the same reason why in the registry the claim that stabilisers improve shooting accuracy is not studied: the only comparison with and without stabilisers, on seven elite archers in 1990, finds no significant difference but is too small to rule out an effect, and nobody has linked a set-up to score with enough arrows. What stabilisers do to the bow I cover in the article on stabilisers.
A set-up built around the archer
If the bow alone does not move precision, one might expect a set-up built around the archer to do so. Here there are no measurements, only opinions, albeit from archers at the very top. A 2020 study interviewed eight archers from the South Korean Olympic squad, all with at least one Games behind them. One says that imposing the same style and the same equipment tuning on everyone does not help anyone improve, and that this is why they still have their old university coach look after their equipment; another, shorter than the rest of the squad and strong in the arms, describes how the wind was a real problem until their coach worked out the string tension that suited their strength and had them try a longer arrow. Nobody has compared archers with a set-up chosen on their measurements against archers moved to a standard one, and so the claim that bow set-up has to be built around the individual archer, not copied from a single model is not studied.
What moving up a level costs
The question also has a financial side, which research has looked at only through what young archers say. In 2025 ten archers at a secondary school in the Philippines, interviewed one by one, described their difficulties, and among them a theme of limited resources emerged: two named money as the main problem, and one said that the higher you go, the more advanced and expensive the equipment you need to buy. These are accounts without figures, so both the claim that equipment spending grows as the archer moves up in level and the claim that for young archers the biggest obstacle is the cost of equipment are reported: they describe how the matter is experienced, not how much is spent or how much cost weighs against other obstacles.
How far the studies go
Nobody has given a group of archers higher-end equipment and measured their scores before and after, and nobody has put the best archers next to a machine, with the same arrows, in the same setting. There is also a limit of method that affects every equipment comparison made on score: at 50 m, by the estimate in the barebow study, two set-ups can be told apart by score only with 900 to 1,050 arrows each, so a change worth a few points in competition can stay invisible both to a study and to an archer trying it out in one session.
What I take from it before buying
These studies do not tell you which bow to buy, and none of them promises points to anyone who changes equipment. They do show where the spread comes from clearly enough to guide a choice, and I take three things from them.
The first concerns comparing your own group with what the equipment can do. A well-built machine at 70 m keeps the same arrow inside a circle a few centimetres across, while in the only direct comparison the archer's vertical spread at 55 m was more than three times the robot's. If the group I shoot at my distance is much wider than what the equipment can deliver, the share a better bow can remove is small next to the share that depends on the shot, and the German study suggests where to look first: the regularity of the release. Measuring it takes your own numbers, collected over time, as I explain in the article on repeatability.
The second concerns preparation more than purchase. The same machines show that, with the archer removed, the group depends on details such as the nock or a damaged vane, which are things you check with numbered, matched arrows and a well-tuned bow rather than with a more expensive one. As for draw weight, it makes sense to choose it with your own strength and the distance in mind: in the eight archers of 1990 the choice looked arbitrary, and the archer with the weakest pull did worse with the heavier imposed bow, though that is a single case.
The third concerns feel. A big change, such as weights that leave the bow steadier after the shot, can be felt, and in barebow the archers' blind ratings followed the measured vibration; a difference of a few millimetres in the arrow, on the other hand, went unnoticed by nine national squad archers. So when I try a new component, feel tells me whether to keep it on trial, but I do not take it as proof that it scores more, and I treat a good impression over thirty arrows as still compatible with chance. As for a tailored set-up, it is reasonable to fit the bow to your own measurements and strength, knowing that what supports it today is the experience of coaches and the words of a few Olympic archers, not a measured comparison.