Sooner or later every archer does the sum: "I finished my seventy-two arrows at 70 metres on 580. Is that a lot or a little, and what level does it match?" On the range the usual answer is a single number, the nine points per arrow of the best in the world, and anyone who uses it as a yardstick nearly always comes away feeling a very long way off. The studies that have measured archers of different levels make it possible to build a fuller ladder, from the first course to the national team, and to see how little that nine is a floor even for the archers at the top.
One limit comes first, because it runs through the whole article: none of these numbers comes from a collection designed to build a ladder. The score is almost always the description of the sample in a study that was measuring something else, such as strength, anxiety or the muscles of the release, and each group shot at its own distance, on its own target face and with its own number of arrows. The ladder is therefore pieced together one study at a time, and the pieces never fit perfectly.
Why think in points per arrow
The average score per arrow is the total divided by the number of arrows: 580 over 72 comes to about 8.06. Its advantage is that it puts rounds of different length on the same scale, an end of six, a 36-arrow test, a 72-arrow qualification round, the old 144-arrow FITA round over four distances. What it does not do is make different distances and faces comparable, because eight points per arrow at 18 metres on the 40-centimetre face and eight at 70 metres on the 122 do not describe the same archer.
With that caveat, the registry lists as reported the claim that the average score per arrow tells you an archer's level. There are plenty of measurements and they fall in the expected order: from secondary school pupils who score under three points per arrow at 10 metres after a basic course, and university students who sit between 3.7 and 4.6 at the same distance after sixteen hours of teaching, up to competitive archers in Hong Kong who finish a 72-arrow outdoor qualification round on about 625 out of 720, roughly 8.7 per arrow. Direct tests, on the other hand, are few. The two studies that compare levels within one sample define those levels by the score itself, so the separation between groups is a foregone conclusion, and nobody has followed the same archers as they moved up. The status therefore stays at reported, and the number should be read for what it is: it says where a group stands under those conditions, not a ranking that holds for everyone.
The top level at 70 metres
On the nine the registry is more confident. That at 70 metres a top archer averages about nine points per arrow is corroborated, because samples with nothing in common land in the same narrow band, between 8.65 and 9.14. Eight Korean university archers with eight to fourteen years of practice average 9.12 over twelve arrows each. Three Malaysian national-team archers, measured arrow by arrow in training, come in at 8.93, 8.86 and 9.14. The reference archer in a study of a score-recording method finishes seventy-two arrows on 649, which is 9.01, and thirty-one elite Korean university and school archers, over fifteen arrows, sit between 8.65 and 8.86.
The same band comes down from the top of the results. At the 2007 outdoor World Championships the men's recurve champion won his twelve-arrow matches with scores between 109 and 115 out of 120, above nine per arrow, and across the head-to-head matches of 120 events between 2013 and 2019, read from the official results, the mean in the sets is 9.15. That last figure mixes recurve and compound and only takes matches that went to a shoot-off, so in the registry it counts as context.
Two caveats apply before using nine as a yardstick. The first is size: the samples run from one to thirty-one archers, with a few dozen arrows each and often in training rather than competition. The second matters more, because nine points per arrow is the level of national-team archers and not what you see at a club. The figure below lines up eleven groups measured at 70 metres, the top level included: they are not every group in the registry, but they cover the ladder from the Cypriot first division to the national team.
Competitive archers in the studies sit around seven
When studies measure archers who compete without being at the top, the number drops considerably. The registry lists as corroborated the claim that in studies of competitive archers the mean score per arrow sits around seven points, not nine: six independent samples, in five countries, fall between 5.9 and 7.65. The most complete is a group of sixty Indian archers from four academies, with at least three years of practice and national competitions behind them, who start from about 7.4 per arrow over seventy-two arrows under World Archery rules. Ten Brazilian archers with competition experience score about 7.2 in a simulated sixty-arrow indoor round at 18 metres, thirty-eight Indonesian club competitors sit between 7.0 and 7.1, twenty Indonesian university archers have a median of 7.65 over seventy-two arrows at 30 metres, and twelve Turkish university recurve archers, with six and a half years of practice, score 6.6 over thirty arrows indoors.
The lowest value in the series, 5.9 per arrow in national-level compound archers at 18 metres, is also the one to trust least, because for that division at that distance it is hard to explain. The highest among the samples gathered on this question, 8.75 in Indonesian juniors at 30 metres, comes from one of the easiest formats and still falls short of nine. The two groups in the figure that compete without being at the top, the Taiwanese university archers and the Indonesian sub-elite juniors, also stay below nine, although above 8.5. These numbers cannot be compared to the decimal point; what holds is the second half of the claim, namely that archers who compete without being in a national team do not, in the studies, average nine.
One of these samples adds something that anyone who trains alone ought to know. In the thirty-eight Indonesian competitors the standard deviation between arrows shot by the same archer is 2.4 and 2.5 points in the two groups: at that level a single arrow swings by more than two points around the mean, which is why a single end says little about the level of the archer who shot it.
The step between club and national team
Between the band around seven and the band around nine there is a step that some older studies measured on the FITA round. The registry lists as corroborated the claim that about 150 points on that round separate a club archer from a national-team archer. The widest comparison comes from amateur American recurve archers measured between 1982 and 1984 and split into four groups: annual averages of 1272 points at world level, 1221 at upper collegiate and national level, 1157 at collegiate level and 1054 in the weakest group, with spreads that widen further down. Two independent Turkish samples find the same step. In the first, national-team archers score 1296 and 1322, men and women taken separately, against 1157 and 1147 for club beginners, with about twice as many years of training; in the second the comparison is 1303 against 1152, and the same work also gives the bottom of the ladder, students on 288 points after an eight-week course.
| Group | Points out of 1440 | Per arrow |
|---|---|---|
| United States, world level (1982-1984) | 1272 | 8.83 |
| United States, upper collegiate and national | 1221 | 8.48 |
| United States, collegiate | 1157 | 8.03 |
| United States, weakest group | 1054 | 7.32 |
| Turkey, national team | 1303 | 9.05 |
| Turkey, club beginners | 1152 | 8.00 |
| Turkey, students after an eight-week course | 288 | 2.00 |
The per-arrow score in the table is the total divided by 144 and blends four distances, from the longest to the shortest, so it sits higher than what the same archer would score at 70 metres. A step of 150 points is therefore worth about one point per arrow, the move from an archer who averages around eight on the round to one who averages around nine. It should be read as an order of magnitude rather than a conversion, because it comes from different eras and formats and from groups described in passing in studies designed to measure something else.
The step shows further down as well. Among teenagers measured at a real indoor competition, those with about four years of practice scored more points per arrow than those with less than one, by a clear margin; which group an archer belonged to, however, accounted for about a quarter of the variation in scores, which says that level is not the whole story. Within a single sample, too, the gaps stay wide. Thirty Japanese school and university archers, split into three groups by their ranking over seventy-two arrows at 70 metres, score 633, 593 and 557.5, the three Japanese points in the figure, and the groups also differ in age and years of training. Even within one selection the spread is wide: at a Korean national selection over 50 metres, the archers placed first to tenth finish thirty-six arrows on 339.5 out of 360 and those placed twentieth to thirtieth on 323.3: sixteen points apart, among archers who all have at least ten years of practice.
At the top, most arrows are not tens
An average of nine does not mean nine with every arrow, let alone a ten nearly every time. The registry lists as corroborated the claim that even in competition at the highest level most arrows are not tens, and the counts behind it are among the largest in the literature. In the outdoor qualification rounds of more than three thousand Chinese professional archers between 2010 and 2016, the mean score per arrow is 7.91 and one arrow in five, 20.7%, lands in the ten; these figures come from the working version of the study, the only one that is open access. In the recurve head-to-head matches at the Olympic Games from 2012 to 2020, the 2021 World Championships and the 2021 European Grand Prix, across 6,374 three-arrow sets, the first two arrows are both tens in only 13.0% of cases, while in 48.9% neither is, and the last arrow of the set is worth 8.25 on average. In eight high-level French archers, 549 arrows at 70 metres, tens make up 30.2%, nines 40.1% and arrows of eight or less 29.7%.
The French group also shows how much the share of tens varies between archers at the same level: from seven to thirty-two out of about seventy-two arrows. There is a physical reason, and a single study measures it. A competitive archer filmed at high speed in a sports hall with no wind, at 55 metres, had a vertical standard deviation on the target of 107 and 108 millimetres with two arrows of the same model, while the ten at 70 metres is 122 millimetres across. The group is already as wide as the ring the archer is aiming at, and the ten cannot be repeated at will, at any level. Why groups open up with distance has an article of its own.
The arrow away from the centre
The opposite tail, the arrow that lands well out, is a more delicate matter. Here the registry is divided: the claim that even top archers put the occasional arrow away from the centre is contested. Almost every study that reports the individual arrows of high-level archers finds it. In four elite Australian archers, at 30 metres on the 60-centimetre face, one arrow in eight scored between five and seven and was judged poor by both archer and coach. In six archers from a Belgian national squad, ninety arrows each at 18 metres, the best shot no sixes or sevens and the weakest one six and fourteen sevens. And in the head-to-head matches recorded by World Archery from 2013 to 2019, the worst shoot-off arrow scored one point, against a mean of 9.09.
One study, however, points the other way. Seven elite Indian archers, over 210 arrows indoors at 18 metres, never dropped below eight: they are the third group in the figure. The study is small, counts thirty arrows each at a short distance and appeared in a regional journal, but it is a contrary result in archers, and under the registry's cautious reading that is enough to keep the status at contested. How often a stray arrow happens to an archer of a given level, over full rounds and in competition, nobody has yet counted.
Where the research stops
The picture has three gaps, and they are worth keeping in mind before comparing your own score with these numbers. The first is format: each group shot at its own distance, on its own face and with its own number of arrows, and there is no collection that gives, division by division, the score per arrow over seventy-two arrows at 70 metres and sixty at 18. The second is time, because every comparison is cross-sectional, different groups photographed once, and nobody has followed the same archers as they moved up. The third is the tail of poor arrows: nobody has counted how many arrows away from the centre are normal for an archer who shoots 1250 on a FITA round and how many for one who shoots 1350.
On top of these comes the quality of the measurements. Many samples are small and were measured in training, the top level rests on groups of one to thirty-one archers, and some of the lower groups on the ladder come from journals with questionable editorial standards, which under the registry's rules never decide a status.
How I use these numbers on the range
The first use is the simplest: divide the total by the number of arrows and compare it only with groups that shot at the same distance on the same face. At 70 metres, among the groups in the figure, the Cypriot first-division archers, the Malaysian state-level group and the lowest Japanese group sit between 7.6 and 8, the middle Japanese group, the Taiwanese university competitors and the sub-elite juniors between 8.2 and just over 8.6, and the elite groups and national teams from 8.65 upwards. This is a reading of the pieced-together ladder, not an official table: it helps you see which zone you are in, not mark yourself out of ten.
The second concerns expectations. An archer who shoots around seven and holds every session up against nine is measuring themselves against a level that, in the studies, belongs to national-team archers. The step between club and national team on the old FITA round is worth about one point per arrow, and in the first Turkish sample the national-team archers had about twice as many years of training as the club beginners. In the samples measured, then, that distance goes with extra years of training, but nobody has followed the same archers as they covered it, and no study says how many years a particular archer will need.
The third concerns individual arrows. If even at the top, at 70 metres, about seven arrows in ten are not tens and three in ten score eight or less, judging a session by its number of tens, or an end by one seven, will lead you astray. Working out how much of a difference in score is noise takes a lot of arrows, and I have done the sum in an article on how many points it takes to say one archer is better than another.