Sooner or later every archer gets filmed and discovers they have a rhythm of their own. Some come up with the bow and the arrow is gone almost at once; others hold aim for so long that whoever is watching holds their breath too. That is when the question tends to come up: how long should a shot last, from the moment the bow goes up to the moment the arrow leaves? On the shooting line there is a fairly precise answer, about one second to draw and one or two to aim, and with it the idea that every archer has a tempo of their own to keep, arrow after arrow.
The phases of the shot have been timed in competition and in training, with video, force plates and sensors on the bow and on the clicker. The figures move away from the common answer in two places. The first is that aiming lasts longer than people say and takes up the largest share of the shot. The second is the one that surprised me most: in the largest sample, the length of the phases says almost nothing about the score of the individual arrow, while the few indications there are, still small, point to how consistent the timing is rather than to how long it lasts.
The whole cycle, phase by phase
The most complete timing comes from sixteen young recurve archers from a Malaysian state squad, sixteen years old on average, who each shot 36 arrows at 70 m in front of a video camera set up to one side: 576 arrows in all. The shot was split into four phases: from raising the bow to the pre-draw, from the pre-draw to the anchor, from the anchor to the release, which is the aim, and from the release to the end of the follow-through. On average the raise took 2.85 seconds, the draw 1.48, the aim 3.32 and the phase after the release 2.69, for a whole cycle of 10.33 seconds, against the roughly 40 the rules allow for each arrow.
Those averages cover a wide range. The shortest cycle lasted 6.24 seconds and the longest 16.68, and the phase that stretched most was the aim itself, from 0.56 to 10.04 seconds across all the arrows. Other studies give values that are close but not identical: in an experienced coach on a force plate the draw took about 1.2 seconds and the aim about 1.6, while in six teenage archers followed in competition with a wrist sensor the great majority of actions, from draw to release, fitted within four seconds.
Before comparing these figures with your own, bear in mind that each study sets the boundaries of the phases in its own way. In the same paper on the experienced coach, the electromyography trace placed full draw and aiming between 0.7 and 1.4 seconds, at a different point from the one given by the force plate. In 1984, with ten university archers, fixing only the hold at three seconds still left individual differences so wide that the authors had to set a rhythm for the whole movement.
The wait before raising the bow
Timing in competition turned up a phase that laboratory studies barely show. A Japanese study went back over the footage of the 2007 World Outdoor Championships, following the two recurve world champions and all their opponents, from the round of 64 to the final, over seven matches each of 12 arrows. Between the signal and raising the bow, 6.7 to 21.8 seconds passed in the men's matches and 4.0 to 25.5 in the women's, and this wait was the longest phase of the shot in 27 series out of 28; the only exception was the women's champion in the final, with 4.01 seconds of waiting against 4.07 of raise and draw.
It was also the phase that changed most from one round to the next, since averaged over each match it went from 6.7 to 17.9 seconds for the men's champion and from 4.0 to 15.3 for the women's. That is why the registry lists the claim that in competition the longest phase of the shot is the wait before the bow goes up as reported: it comes from a single tournament and from head-to-head matches, with the timing of that format, and the study does not say whether a longer wait helps or costs points.
The draw and the aim
On the draw the figures are fairly close to each other: in the sixteen Malaysian archers it went from the pre-draw to the anchor in 1.48 seconds on average, and in the coach on the force plate in about 1.2. Comparisons between levels, on the other hand, depend on who is set beside whom. In a study from the 1990s of fifteen experienced archers and fifteen beginners, the movement of the shoulders between set-up and draw peaked at 0.38 seconds against 0.63, a value read off a graph and without any statistical test. Among the Malaysian archers, all of them already competent, the eight best took a median 1.46 seconds over the draw and the eight weaker ones 1.36, a difference that was not significant and, if anything, ran the other way. The claim that higher-level archers complete the draw more quickly is therefore contested.
The aim takes longer. In the sixteen Malaysian archers it was the longest phase, 3.32 seconds on average, and took up about 69% of the time spent on draw and aim together. In four elite archers measured in 1990, the aim on good arrows lasted between 2.12 and 5.82 seconds with recurve and barebow and reached 8.40 with the compound. Every measurement agrees that the phases of the shot last only a few seconds, but the split in the common answer, one second to draw and one or two to aim, is confirmed by a single archer, the coach on the force plate, while the small elite group from 1990 goes against it, so the status is contested.
The idea that the draw and the aim share the time almost evenly has even less support. The only study that separates the two phases with boundaries close to those of the claim, in a journal the registry has no reservations about, is the Malaysian one, and it finds an aim more than twice as long as the draw; with a single study, and one that goes against it, the status is contradicted.
At the 2007 World Championships the men's champion aimed for between 2.65 and 4.37 seconds on average per match, and the women's champion for between 0.61 and 1.24, after a raise and draw of about four seconds; his opponents sat between 1.70 and 6.86 seconds, hers between 2.03 and 4.60. The two gold medallists of the tournament were therefore at opposite ends of the range, and in those data there is no aiming time of the winners.
The last seconds before the click
Within the aim, with the recurve, comes the final pull, the slow movement that brings the point past the clicker, and two independent groups have measured it with figures that are close. In eight archers at 90 m the string hand went from about 500 mm per second at the end of the draw to about 1 mm per second while aiming, which is about 5 mm over the last 5 seconds before the clicker. A strain gauge on the clicker of seven international junior archers showed, in one archer, a final pull starting 3 to 2 seconds before the click with about 5 mm of draw, and in another about 2.5 mm in the last second. The claim that the final pull lasts two or three seconds and adds only a few millimetres of draw is reported, because the samples are small and the phase has no fixed boundaries: within the same archer its start shifted by almost a second from one arrow to the next.
Then the clicker goes, and from there to the arrow leaving runs the most regular interval of the whole shot. Since 1987 six research groups, with different instruments and at distances from 8 to 90 m, have measured it at between 120 and 240 milliseconds, and in the largest sample, thirty-nine elite archers measured before a World Cup stage, the average was 163 milliseconds, with a spread of 10.4% across the whole group. It is one of the few points in this article on which the registry holds a corroborated claim: between one and a half and two tenths of a second pass between the clicker sounding and the arrow leaving. I look at that interval, and at what fills it, in the article on the clicker.
After the release
The last phase, from the release to the end of the follow-through, lasted 2.69 seconds on average in the sixteen Malaysian archers. It is the only one whose length was linked to the score of the individual arrow, with a weak negative correlation, r = -0.111 over 576 arrows, and the eight best archers finished it more quickly than the others, a median of 2.40 seconds against 2.72. Another study, of eight Korean university archers, had found on the better shots a draw-side shoulder that rotated more slowly during the follow-through, so the claim that a quicker release and follow-through go with higher scores remains contested. The two studies measure different things, after all, and a shorter final phase can mean a cleaner follow-through just as easily as a bow arm that drops sooner.
Timing and the arrow's score
This is the part I did not expect when I started reading these studies. In the sixteen Malaysian archers the correlation between the total time of the shot and the arrow's score was -0.006, no relation at all, and the same held for the raise (-0.004), the draw (0.031) and the aim (0.016). Back in 1990, an experiment on seven elite archers at 70 and 90 m had found no relation between aiming time, drawing time and precision, although it said so in a single line and without reporting the figures.
The picture is not clean, though, and the claim that the time spent on the phases of the shot does not predict an arrow's score is contested. Besides the phase after the release, there is a study of four elite Turkish women archers, 108 arrows at 18 m, in which the full draw, from the first activation of the trapezius to the clicker going off, took 5.6 seconds on the 8s, 5.0 on the 9s and 4.8 on the 10s, with a weak correlation (r = -0.280), while the time spent at full draw did not change. These are small effects, one of them from a tiny sample, and they sit in phases other than the one people argue about most.
That phase is the aim, and the question of whether holding aim for too long makes the shot worse is the most contested of the group. The only experiment with figures, a master's thesis on eight elite archers from the British squad, found that when they held aim for three times their usual time the average score at 70 m fell from 8.60 to 7.75 points per arrow; the conditions always came in the same order, however, and fatigue could not be separated from duration. On the other side, the eight best Malaysian archers aimed for longer than the weaker ones, a median of 3.38 seconds against 2.72, and took 10.46 seconds per arrow against 9.74. Finally, in eight elite French archers the length of the aim was linked to the arrow's score in only two cases out of eight, and in opposite directions, which helps to explain why group averages cancel out: they put together archers who behave in contrary ways.
Consistency of timing
If the duration says little, consistency is what is left to look at, meaning how much the time of a phase changes from one arrow to the next for the same archer. The indications are few and small, but nearly all of them point the same way. Two university archers, with 18 arrows each at 70 m and accelerometers on their arms, aimed for 4.2 seconds on low-scoring arrows, 3.0 on middling ones and 2.9 on high ones, and between the middling and the high arrows only the spread changed, with a standard deviation of 1.4 seconds against 0.4; the study runs no statistical test. Seven archers from the German senior and junior squads, with a microphone recording the clicker and the release, had an average interval of between 144 and 176 milliseconds, and none of them repeated it exactly: even in elite archers the time between the clicker and the release changes from arrow to arrow. In that group, those who varied less had the higher average, which is why the claim that among elite archers, those with a more consistent release time score higher is also reported, with the caveat that with seven people even a strong relation can arise by chance.
The 2007 World Championships study, the one most often quoted on this point, also complicates it. The men's champion held a steadier aiming time than his opponent in six matches out of seven; the women's champion, who aimed for a second or a little more, was less steady than her opponents in all seven. What the two gold medallists shared was something else: the aim was the only phase that the round of the tournament did not move, while the wait before raising the bow and the draw stretched and shrank from one match to the next. The claim that champions hold more consistent aiming times than their opponents is reported, and goes no further because the result in favour and the one against come from the same study.
What I take from it on the shooting line
These studies do not hand over a shot time to copy. They say that a cycle lasts around ten seconds, that the aim takes up the largest share of it, and that the length of an individual arrow predicts little about its score. No study with checkable figures has measured what happens to the score when a group of archers changes its rhythm for a few weeks, so the three points I take from this should be read for what they are.
The first is to measure your own shot before judging it. A phone on a tripod, to one side of the archer, and a few ends filmed in training are enough to time the four phases frame by frame: the Malaysian study did just that, with a camera at 30 frames per second, 1.5 m from the archer. More than the average, what matters is how far it spreads from arrow to arrow and in which phase: if the aim moves from two seconds to six while the raise stays the same, that is where the shot is changing. There is a practical guide to filming in the article on video analysis.
The second is not to chase another archer's timing. The two world champions of 2007 aimed in opposite ways, and the better Malaysian archers aimed for longer than the weaker ones. A long or short aim is not, on its own, a fault to correct, and shortening or lengthening it on command means changing the shot without knowing whether it will pay off. I go into aiming time, and the window in which each archer releases, in more depth in the article on aiming time.
The third concerns consistency, and it is the one I am most careful with. The indications suggest that repeating your own timing matters more than having any particular timing, but they come from two archers, from seven archers and from the two champions of a single tournament, in which the women's champion was the least consistent of all. This is how I use it: I know my own aiming interval, and when an arrow falls well outside it I treat it as information to look at on video rather than as a mistake to punish. The same goes for the wait before the raise, the longest phase in competition, which is usually given a shape through a pre-shot routine: the footage also shows how closely it repeats.