On the next target there is an archer putting six arrows into a fist-sized group, and anyone who has just started wonders what he does that they do not. From the outside the movement looks the same: raise the bow, draw, aim, and the arrow goes. Research has been asking this question for more than fifty years, and the first trace to set an expert beside a beginner dates from 1971: the force curve on the bow of a fourth-dan kyudo archer rose smoothly, while a beginner's rose in jerks.
Since then studies have compared experts and beginners on many things, from the repeatability of the shot to which muscles do the work, from balance to the length of the draw, and on to years of practice. The answers do not all carry the same weight, and almost all of them share one limitation that needs explaining before we get to the numbers.
How an expert is compared with a beginner
The design is nearly always the same: take two or three groups of different levels, measure them once and compare the averages. A comparison of this kind shows that the groups differ, but not why. If experts do something beginners do not, that something may be what made them experts or the result of years of arrows, and a measurement taken on a single day cannot tell the two apart. None of the studies in the registry has followed the same archers as they went from beginner to expert.
Then there is the question of what "expert" means: in one Japanese study three years of practice was enough, in a Belgian one the experts were Olympic archers, and in kyudo the level is the examination grade. Beginners range just as widely, from people who have shot for less than a year to people who had never picked up a sporting bow, so with every result it matters who was being compared with whom.
The same shot, one arrow after another
The difference measured most often is the repeatability of the shot, and it is also the one with the most material behind it. Even so, the claim that expert archers repeat the shot with less variability than beginners remains reported, not corroborated: every study is a comparison between different groups, with small samples and often no statistical test.
The clearest figure comes from a Belgian laboratory in 1990. Fifteen archers, three Olympians, six Belgian national-level archers and six beginners with a few months' experience, shot eighteen arrows at 18 m with electromyography on ten muscles. The authors classified each muscle's activation pattern from one shot to the next as identical, analogous, conforming (recognisable, but shifted in timing or amplitude) or different.
In the Olympians 92% of patterns were identical from one arrow to the next, and the rest were analogous, all in a single muscle. In the national-level archers identical patterns fell to 41%, and the variability lay mostly in the fingers of the string hand, before and during the release. In the beginners there were none: 79% of patterns were conforming and 21% different.
Other samples point the same way with different instruments. In 28 Chinese recurve archers the variability of muscle activation during the draw is around 10% in the experts and 21% in the beginners, on most of the muscles measured. Among high-school members of a kyudo club, the position of the drawing elbow changed between two shots by seven tenths of a millimetre in a third-year student and by 36 mm in a beginner. Finally, in 2026, a comparison of eight French elite archers and seven intermediate archers at 70 m measured the spread within each archer.
The spread of scores in the elite group was less than half that of the intermediate group, 0.91 points against 2.02, and the same held for the shift of weight over the feet, while draw-arm shoulder abduction varied about a third as much. On one measure, though, the variability went the other way: the peak speed at which weight shifts after the release varied more in the elite archers.
Not everything falls, and not for everyone
This is where the picture gets more complicated. In the same Chinese sample the consistency of activation while aiming does not distinguish the levels, and neither does the consistency of elbow angles; consistency during the draw separates experts from beginners clearly, but finds no difference between elite and intermediate archers on any variable. In fourteen archers from the Polish national team the draw was less dispersed than in fourteen people who had never shot a bow, above all at the centre of mass, yet on two measures of complexity the experts came out less regular, with more corrections.
Among archers who are already good, consistency separates them far less. As early as 1990, in eight elite and good county-standard archers, the consistency of posture at release did not distinguish the two groups, and in six elite archers from the Chinese national team the consistency of the movement did not separate arrows in the 10 ring from the rest. The more recent reading is subtler than "experts move less": in six archers from the Belgian national team, on arrows scoring 9 and 10, variability was not eliminated but channelled into directions that do not change the orientation of the arrow, a result that a 2025 review also picks up. The reading that follows is that what falls is not variability in general but the variability that decides where the arrow goes.
Each archer repeats their own shot
If experts repeat themselves, it is tempting to think they all repeat the same technique and that copying it would be enough. On this point the studies agree to an unusual degree, and the idea that elite archers share one correct technique is refuted in the registry. In the fifteen members of the US national team measured in 1990, the activation of the finger muscles varied on average by 16% within the same archer and by 52% between archers, and even the way the release was produced split into two distinct patterns. In video of the top fifteen men and top fifteen women in the world ranking, every phase of the shot shows at least two styles.
A common skeleton does exist: in 28 Chinese archers across three levels the number of muscle synergies is the same, around three. Within that skeleton, though, muscles, timing and geometry change from one archer to the next, so what the data link to skill level is an expert's consistency more than the shape of the shot. Consistency as a subject in its own right is covered in the article on repeatability and consistency.
Where the muscles work
The second difference concerns which muscles do the work. The advice on the shooting line is familiar, draw with your back and not your arm, and the evidence on archers that experts hold the draw with the shoulder and back more than the forearm points almost entirely that way; in the registry it is reported. In thirty Japanese archers split into three levels by their score at 70 m, in the second before the clicker dropped the lower trapezius of the draw arm worked at 54.0% of maximal contraction in the elite group, against 39.4% in the pre-elite and 34.0% in the beginners, while biceps and triceps worked less in the elite than in the beginners.
It is harder to argue that elite archers activate their muscles less than intermediate archers while aiming across the board, and the claim is contested. Between the eight French elite and seven intermediate archers, the elite activated 11 of the 12 muscles that differed between the groups less, from the fingers to the trunk, and only the middle trapezius of the draw arm more. In 27 Turkish archers, on the other hand, posterior deltoid and middle trapezius worked harder in the elite. The reading that holds up best is that as skill rises the work shifts from the fingers and arms to the muscles that hold the shoulder blade. No study yet says whether elite archers activate their muscles less because they are stronger, and so work at a lower share of their maximum, or because their technique loads those muscles less.
The same applies to the trunk. The idea that the transversus abdominis matters more for stabilising the trunk in more experienced archers is reported: two comparisons between groups point that way, and only one of them sets the transversus beside the superficial abdominal muscles. That comparison is in nine kyudo practitioners: the group of muscles active in the second half of the shot relied on the internal oblique and transversus abdominis in the four experts and on the external oblique in the five less experienced. In twenty-eight Indian archers the transversus abdominis worked at 22.9% of maximal contraction at anchor and 32.9% at release in the elite group, against 7.4% and 12.2% in the beginners, but the superficial muscles were not measured. Where the transversus was set against the score, in the eight French elite archers, it was more active after the release on the worse arrows, not less.
Balance before the release
When the string goes, the body has to absorb the push of the bow. Only one study has compared how experts and beginners do this, in 2024, and the claim that follows, that experienced archers prepare their balance before the release while beginners correct it afterwards, is reported. Nine Korean archers with an average of sixteen and a half years of practice and nine university students new to the sport simulated the shot on a force plate, without arrows, using a laser pointer on a target at 10 m. In the experts the load on the feet was already shifting before the release, in the opposite direction to the push of the bow; five of eight leg muscles peaked in the window just before the release, whereas in the novices the work came afterwards; and after the release the experts regained their balance sooner, with a similar maximum excursion.
The limitations weigh heavily: eighteen men, two trials each, no arrows and no scores. The other study on the subject, of nine elite Korean women archers, finds the anticipation in the last fifty milliseconds before the release, but does not find that a larger adjustment goes with better aim: sideways there is no link, and front to back, without a clicker, a larger adjustment went with less precise aim.
Elbow, string and face
Two differences in geometry rest on a single sample each. The first is the string elbow: in a Japanese letter from 1997, on thirty barebow archers filmed from the side, the angle between the two shoulders and the drawing elbow was about 155 degrees in the fifteen with at least three years of practice and 140 in the fifteen with less than a year. The authors read it as a higher elbow. The claim that expert archers keep the string elbow more open is reported, but the figures are averages of three shots read from a flat image, with no statistical test and no scores.
The second concerns the string against the face. In sixteen world-class recurve archers filmed at 4,000 frames per second, the string touched the face in four of them, three times on the chin and once on the chin and lips, while in the other twelve no contact was visible. The claim that in top-level recurve the string does not touch the face, or barely brushes it is reported, and it describes the top of the sport rather than comparing it with beginners. The same paper also calls that contact slight in its results and capable of altering the arrow's behaviour considerably in its conclusions, so whether it costs points is not known.
Timing and breathing
On the speed of the draw the answer depends on who is being compared, and the claim that higher-level archers complete the draw more quickly is contested. Between experts and beginners there is a difference: in 1997 the peak movement of the shoulders, going from set-up to draw, came at 0.38 seconds in the experts and 0.63 in the beginners, and in 2024, in 28 Chinese archers, the draw took 2.1 seconds in international masters and 2.5 in the second category. Among archers who are already competent the difference disappears: in sixteen young Malaysian recurve archers at 70 m, the best eight took 1.46 seconds against 1.36 for the other eight, a non-significant difference, and in the Chinese study the first category already drew as fast as the masters. The 1997 figure is read from a graph, with no test, and the 2024 one comes from a questionable journal.
On breathing there is even less. The idea that in expert kyudo archers breathing follows a fixed pattern even when they only imagine the shot is reported and rests on a single Japanese study from 1964, in kyudo. Fifteen students imagined their own shot with their eyes closed while a tube on the chest recorded their breathing: in the four graded practitioners the imagined shot ended with a slow exhalation lasting four or five seconds, while in the others breathing changed with no fixed pattern. Nobody actually shot, and for recurve and compound the question has never been measured.
Years of practice and skill level
That leaves the simplest explanation, that experts are experts because they have been shooting longer. The studies that set years against skill level point in opposite directions, though, and the claim that years of practice tell how good an archer is is contested. In the Chinese national team the official technical grades follow the years, ten to thirteen for internationals and six to eight for first class, but with no statistical test. Among sixty-two high-level archers, by contrast, years did not clearly separate European level from national level, 15.7 against 11.8 with p = 0.061, while arrows per week separated them well, 696.5 against 273.7. Finally, in thirty-six Japanese university kyudo practitioners, years of practice did not predict the percentage of arrows on target. None of these studies counted the arrows shot in those years, either, and only one counted the arrows shot per week.
The picture, dimension by dimension
| What was compared | What was found | Status in the registry |
|---|---|---|
| Repeatability of the shot | experts repeat themselves more, but not on everything and not among archers who are already good | reported |
| One elite technique | each archer repeats a pattern of their own | refuted |
| Drawing with the back | more work from the shoulder blade, less from the arm as skill rises | reported |
| Less muscle activity while aiming | less in fingers and arms, more in part of the shoulder blade | contested |
| Transversus abdominis | two comparisons between groups, only one with the superficial abdominals | reported |
| Balance before the release | a single comparison, without arrows | reported |
| More open string elbow | a single study from 1997, with no test | reported |
| String barely brushing the face | visible contact in 4 of 16 world-class archers | reported |
| Faster draw | yes against beginners, no among competent archers | contested |
| Fixed breathing pattern | a single kyudo study, with the shot imagined | reported |
| Years of practice | studies pointing in opposite directions | contested |
What I take to the shooting line
The picture does not give a list of things to copy, because it describes how experts differ and says little about how they got there. I take three points from it, knowing that none has been put to the test in an experiment on scores.
The first is to measure your own consistency rather than chase someone else's form. The simplest way is the spread of your scores: the standard deviation of arrow scores, over a few ends at the same distance, is exactly the measure on which the French elite and intermediate archers differed by more than double. A side-on video, always filmed from the same spot, shows whether elbow and shoulder return to the same place. The typical mistakes of people starting out are covered one by one in the article on common beginner mistakes.
The second is not to treat every feature of an expert as a cause. A more open elbow, a more active transversus or a shorter draw may be the result of years of arrows rather than the route to getting there, and the data cannot tell the two apart. There is more material on drawing with the back than on the rest, but even there nobody has compared archers taught to draw with the back against archers taught another way.
The third is to count arrows as well as years: the only study that counted arrows per week found that weekly volume separated national from European level better than years did. How many arrows to shoot, and on what evidence, is the subject of the article on arrows per week. How the movement settles with practice is the subject of the article on motor learning.
What is missing
Above all, what is missing is a study that follows the same archers from their first year onwards, with the same measures of variability, muscle activation and balance, and with enough arrows per person to link each measure to the score. Until there is one, we know how experts and beginners differ when measured on a single day, and much less about how one becomes the other.