"How do I improve my score?" is the question I hear most often, and the answer it usually gets is about a detail: change your anchor, raise your elbow, breathe better. Before the details, though, archery research says a few basic things that decide where your effort is best spent. An archer's level shows above all in how little the shot varies from one arrow to the next, not in how closely it resembles a model: elite archers do not share one correct technique, each of them repeats their own. A sizeable share of the score differences between archers goes, in a statistical sense, with mental factors such as confidence and not dwelling on mistakes, and archers who do better on attention tests tend to shoot higher scores, although the studies do not agree on this.
The starting point is a measure, the score, and it is the yardstick I use to judge every lever that follows: body stability, the release, the gaze, the mind, the heart and training load. For each one I say what the studies have measured, how well it holds up and what I do with it when I coach; when something has only been seen in other sports, or has not been studied in archers, I say so.
Where to start: measuring your level
The most useful number for knowing where you stand is the average score per arrow, because it lets you compare rounds of different lengths. Archers who measure themselves, though, often take the nine points per arrow of the very best as their benchmark, and the comparison does not hold. Across the six samples of competing archers reported in the studies, from five countries, the average sits between 5.9 and 7.65 points per arrow, mostly around seven: sixty Indian archers with national competition experience start from about 7.4 over seventy-two arrows under World Archery rules. Nine is the level of national-team archers, and even there it is an average rather than a floor: eight French recurve archers in their country's top twenty send one arrow in three to the eight ring or below at seventy metres.
These numbers need reading alongside the format, because distance, target face and number of arrows shift the average before level does. Groups of known level, from adult beginners to competitive archers in Hong Kong, measured one at a time line up in the expected order, but these are one-off comparisons between different archers, and nobody has followed the same archers as they moved up a level. The score per arrow tells you where an archer stands under those conditions, not their place in some absolute ranking.
There is also a reason I use the score as the lens for judging everything else: it moves little from one round to the next. In the only study that has set the stability of the score beside that of laboratory measures in the same archers, in the seven university recurve archers of the training group a simulated 72-arrow round at 70 metres varied by 3%, with a typical error of 4 points, while the electrical activity of the muscles before the release varied by 8 to 55% depending on the muscle and the group. Whether it is the most repeatable measure of all cannot be said: in the same study strength shows higher consistency, two shoulder angles show the same 3% variation, and the two sessions had twelve weeks of training between them. What holds is more modest, and it is enough to work with: for the same archer, the score of a full round fluctuates much less than an electromyographic trace or a strength test. That is why, when I assess a lever, the first thing I look at is whether the score has moved, over enough arrows and under the same conditions.
That leaves the question of how far a result can be predicted. The common view is that data from the moment, heart rate or how you feel that morning, say a lot about how the competition will go, and the studies give a contested picture. In an indoor competition between adolescents, shooting time and heart rate variability before each arrow together explained 11.8% of the score of a single arrow, while the difference between beginners and experienced archers alone weighed about twice as much. In thirty-two Indian archers, a model trained on one session and tested on the next was out by about six points in three hundred, and among the predictors years of practice came first and the score in previous competitions third; in second place, though, came resting systolic blood pressure, measured before the warm-up, which is exactly a figure from the day itself. Nobody has yet compared these models with the simplest prediction of all, the average of the last few competitions. It is still the benchmark I use with the archers I coach, because they already have it and it does not depend on how they feel that morning.
What sets better archers apart
The first sign of level is repeatability. Experienced archers repeat the shot with less variability than beginners, and the comparisons always point the same way: in fifteen archers measured with electromyography, the Olympians repeated the same muscle pattern on 92% of shots, the national-level archers on 41% and the beginners on none. These are comparisons between different groups measured once, though, and among archers who are already experienced the picture changes, because consistency of movement does not separate the arrows in the ten from the rest. What drops as the level rises, then, is not variability in everything but variability in the things that decide where the arrow goes, and those are what is worth working on; I cover this in repeatability and consistency.
The second is that there is no single model to copy. The idea that elite archers share one correct shooting technique has been refuted by independent samples measured with different instruments since 1990: each archer repeats their own way of shooting with great consistency, and that way differs between archers. In the fifteen archers of the United States national team, the muscle pattern of the release varied by 16% within the same archer and by 52% between archers, and a model trained on the trajectories of shoulders, elbows and wrists identified which of four national-level athletes was shooting in 98.7% of cases. There is a common skeleton, since the number of muscle synergies is the same at every level, but within that skeleton timing, angles and muscles are a personal signature. Copying a champion's elbow angle, then, does less for you than making your own repeatable.
The third concerns the mind. In a statistical sense a sizeable share of the score differences between archers goes with mental factors, but the share is much lower than the percentages that do the rounds at the range. In the largest study, of American recurve archers in 1986, seven measures explained 58% of the differences in score, and the mental block (dwelling on mistakes, confidence, visual imagery) accounted for 26%, almost twice as much as strength and build combined; widening the sample to archers with some missing data, the share from the same analysis fell. In sixty-one athletes from the Chinese national team, psychological factors explained about a third of the variation in results. These are associations collected at a single point in time: they say how far mental measures go together with the score, not how many points you would gain by changing them.
The fourth is the least certain. Whether archers who hold concentration better shoot higher scores is contested, and the disagreement follows the instrument used to measure it. In the largest piece of work, over a hundred archers up to world level, self-reported concentration did not go with the score (r = 0.16). Attention tests, which are tasks done against the clock, almost always find the same sign: in sixty-one archers from the Chinese national team, a pen-and-paper test was linked to performance in the following week's competition, with a correlation of 0.43. Here too there are exceptions, and none of these studies says whether training attention raises the score.
1 · Body stability while aiming
How still the body stays around the release is among the signals the research most often links to the score. It is measured through the centre of pressure: the archer shoots standing on a force plate and you watch how much their weight shifts over their feet. The studies point almost entirely the same way, namely that the less the centre of pressure moves, the better the arrow tends to be, and within the same archer the picture is consistent: comparing the best arrow with the worst, on the worst the centre of pressure travels more than twice as far. But how and when also matter. While aiming, elite archers show smaller but faster sway than intermediate archers, and in 39 elite archers measured at the World Cup it was the peak speed of sway after the release that predicted the score, while the sway in the second before did not enter the model. So I read stability together with the follow-through, rather than as a number to drive down at any cost.
An expert's body, moreover, does not wait for the release to organise itself. In nine experienced Korean archers compared with nine novices in a simulated shot without arrows, the load under the feet was already shifting before the release, against the push of the bow, and five of eight leg muscles reached their peak before the loose, whereas in the novices the work came afterwards. It is a single comparison, on eighteen people and with no score, and in nine elite women archers a larger anticipatory adjustment did not go with better aiming: it is a clue to how the body works, not a yardstick for judging an athlete.
The usual explanations for the gap are strength, proprioception and breathing, and for none of the three do the archery studies show a solid link. That strength work makes the aim steadier has been measured in students who had never shot, with four weeks of side planks and bow tremor going down. Shoulder position sense, set against the score, gives results that do not agree, with a clear link at only one angle out of six. As for breathing, there is no single rhythm to teach: in the only data on archers, before the shot one breathed at 20 to 26 breaths a minute and the other at 7 to 15, and the rhythm varies from one archer to another.
Stability can be trained, but whether training it raises the score is a different matter. Core exercises have given mixed results even on measured balance, and in before-and-after studies of club youngsters without a control group the score always goes up. The most robust result points the other way: in thirty national-level archers assigned at random, six weeks of balance work alone did not separate their score from that of archers who added nothing. With the archers I coach I use these exercises as part of their preparation and check their effect on the score, not on the wobble board. The programmes are set out in posture and stability and in balance, core and proprioception.
2 · The release
The release is the shortest moment of the shot: in high-speed footage of world-class archers the string leaves the fingers in about two milliseconds, 1.9 on average. What prepares it, though, shows up in the muscles. Compared with weaker archers, elite archers almost always activate the muscles of the fingers, arms and trunk less while aiming, while part of the trapezius that holds the shoulder blade works harder: in thirty Japanese archers the lower trapezius reached 54% of maximal contraction in the elite group against 39% and 34% in the other two groups, and the biceps and triceps worked less. The picture recurs across three laboratories but not for every muscle, and the reading that holds best is that as the level rises the work moves from the fingers and arms to the shoulder blades. I go into this further in muscle synergies.
There is a precise rule going round about the fingers: the release should be passive, the fingers relax rather than open. As an instruction, electromyography refutes it. In the fifteen archers of the United States national team the release split into two stable patterns, eight who relaxed the flexors and seven with an active burst of the extensors, and their average score for the year was almost identical, 1,209 against 1,213 out of 1,440. Fingers that stay curled, the kind you see in footage of the best, are the result of a good release, not the command to give your hand.
What seems to matter more is consistency in time. In seven archers from the German national squads, those who kept the time between clicker and release most consistent had the highest average per arrow, and the relationship held even when one archer at a time was taken out of the calculation. With seven people it is a clue rather than a measurement, but it fits well with what I said about repeatability and with what happens around the clicker.
A consistent release is not built with words, and telling someone to "release more softly" does little. External feedback helps more, such as slow-motion video reviewed with your coach (I explain how to use it in video analysis and self-coaching). As for attention during the shot, in motor learning studies in other sports, focusing on something external usually leads to better learning than focusing on the body; in archery the only trial that compared the two, watching the sight or feeling your own body, found no difference in score, with six arrows per condition, too few to detect a small effect. The full picture is in attentional focus.
3 · The gaze
The best-studied phenomenon is the quiet eye, the steady fixation on the target in the seconds before the release. In ten Olympic-level archers from the British squad, compared with ten beginners at thirty metres, the quiet eye lasted 1.96 seconds against 0.76, and with the two groups taken together longer durations went with smaller errors. The difference lay in the onset: in the experts the fixation began earlier, it did not end later. It is a comparison between levels, though, and read that way the length of the quiet eye may simply be another way of saying that someone has been shooting for years; in archery nobody has yet checked that training it raises the score.
The weight of eyesight itself also needs scaling back. In a hundred and two recurve archers, from beginner to Olympic champion, eleven visual screening tests, from near and distance acuity to colour vision, were not related to the score, and only depth perception stood out. The way I work, then, gaze training makes sense within the real shot, on when and how the target is fixated, more than as an eye exercise done away from the bow.
4 · The mind
If a share of the score differences goes with mental factors, the next question is whether training them raises the score, and here the archery studies with a comparison group split two and two. In favour, twelve archers from the Hong Kong squad gained 11.3 points over seventy-two arrows after twelve weeks of mindfulness while their matched controls dropped three, and thirty-six Turkish beginners gained more with eight weeks of imagery or slow breathing than a group that only trained. Against, eighteen Greek archers, with two official competitions as the measure, showed no difference after four weeks of imagery and relaxation, and sixteen Ukrainian archers could not be told apart from controls after a week of autogenic training. The two null results come from the shortest programmes, which suggests that duration matters, but nobody has put that to the test. Tools and programmes are covered in the article on mental skills training.
There is also the question of how time at the range is used, and the only direct measurement concerns a single group. Ten national-level archers at a Malaysian sports school, filmed in three sessions in the run-up to competitions, spent 83.3% of their time in closed practice, almost all of it technical, and not a minute on competition-like practice. Their coach explained that on other days the squad held scoring sessions, and the study does not say whether adding competition-like practice would change the result. I mention it because it describes a habit I see often, plenty of technique arrows and few scored ones; whether it pays to change the balance has not yet been measured in archers.
5 · The heart and arousal
Heart rate variability (HRV), how much the time between one heartbeat and the next varies, is a widely used indicator in sport, and in archery its results have to be read one at a time. In seventeen adolescent archers followed through an indoor competition, in the twenty seconds before each arrow the experienced archers had a higher RMSSD than the beginners, which is read as more parasympathetic activity; in the same study, though, the score of a single arrow rose with the component usually read as sympathetic, and the idea that parasympathetic activity is higher on the best arrows remains contested.
The same goes for heart rate. In 1990, in four elite archers, the cardiac cycle lengthened over the last beats before the good arrows, but in the same year, in twenty-eight elite archers, the heart sped up rather than slowed in the seconds before the release, and the most recent measurement puts the lowest point two or three seconds after the release, once the arrow has gone: whether the heart slows before the shot is contested. The link between a lower heart rate and a higher score also shows up in observational studies, but when an intervention has shifted the heart rate the score has not followed. The most cautious reading is that the slowing accompanies a good shot without causing it.
On arousal, the data point to a personal zone rather than one window for everyone. Three top Brazilian archers followed for a season had optimal zones that differed from one another, and the zone changed with distance; even inside their own zone, each of them still shot some mediocre ends. It is a sample of three, and nobody has checked whether deliberately bringing yourself back into your zone raises the score.
HRV has another use too, as a gauge of load. In endurance sports, read as an average over several days and against the same athlete's own profile, it has tracked form and freshness in some top athletes, while a single morning's value remains ambiguous; it has not been studied in archers. With the archers I coach I use it that way, as a trend alongside what the archer feels and never in place of it.
6 · Training load
How many arrows to shoot is one of the questions I am asked most often, and the first thing the data show is how differently competitive archers train: volume varies by an order of magnitude from one setting to another, from the 4.3 hours a week reported by seventy-five German competitive archers to the 36 of fifteen national-level university compound archers in Taiwan. These are figures reported by the archers themselves, in studies set up for other purposes, and nobody has set them against the score with the same yardstick: the right volume for an archer cannot be read off a table. I discuss it in how many arrows per week.
In my experience the signs that the load has gone too far are fairly easy to recognise: the bow arm shakes more in the last ends, aiming times get longer, tension in the shoulder or forearm does not clear from one day to the next, sleep gets worse. When I see them together I cut the volume rather than add to it.
Periodisation, organising the season in cycles that alternate volume, intensity, technique and recovery, comes from endurance and strength sports. In archery no study has compared archers trained with and without periodisation, and nobody has measured the effect of tapering before a competition, neither how far to cut the arrow count nor for how many days. The guidance that circulates on recovery weeks and tapers comes from general training theory. I use it, because the reasoning holds up, but I treat it for what it is: a hypothesis to test on the archer in front of me, with the score as the measure.
From data to intervention
All these levers can be measured, some with tools within reach of a club (a camera, a heart rate strap, a well-kept training diary), others only in a laboratory. The tool, though, matters less than the method, which is a simple cycle: you observe where the archer stands, turn the observation into a number, choose a single intervention and, after the right amount of time, measure again, with the score as the final test.
The difference from working by feel lies entirely in that "single". An archer who goes by feel tries ten things at once and does not know which one worked, whereas the method changes one variable at a time and allows for some ideas not holding up. You also have to accept that improvement does not show session by session: one arrow says little, a week of arrows starts to say something, and to say that an archer has changed you need enough arrows to get past the normal variation between one round and the next. How many you need I explain in how many points it takes to say one archer is better, and the Did I really improve? tool does the sum for two rounds.
For the same reason the starting point changes from one archer to another. The levers apply to everyone, but the one limiting your score today may be stability for one archer, release timing for another, handling the competition for a third. That is the point of an initial assessment: working out where to step in first, instead of spreading your effort across everything.
Where to go next
Every claim cited in this article has its own entry in the evidence registry, with its status, its year-by-year history and the studies behind it. The entries are grouped into six domains, and each domain has a page listing them with their status: technique and biomechanics, physical training, mind and attention, equipment and tuning, competition and scoring, health and recovery. If you are after the answer to a specific point, start from the right domain.