Joint angles at recurve release: Santosh Kumar 2024

Five Indian recurve archers, men of international level from the SAI training centre in Sonipat, were filmed at the moment of release at 70 metres, and 18 angles were measured on each recording: shoulder, elbow and wrist on both arms, head and trunk inclination, bow verticality and several angles between the bow arm, the drawing forearm and upper arm, the shoulder line, the arrow and the line of force. Some mean values: 96.12 degrees at the bow-arm shoulder, 118.99 at the drawing shoulder, 185.07 at the bow-arm elbow, 3.45 degrees between arrow and line of force, and 179.96 for bow verticality.

Performance is the arrow score, averaging 9.97 points (range 9.9 to 10). With only five archers a correlation is significant only above 0.878, and fifteen of the eighteen angles exceed it, with values from 0.881 to 0.998; head inclination (0.540), bow verticality (0.280) and the angle between bow arm and line of force (0.821) stay below. Almost all the angles also correlate with one another, with coefficients up to 0.999. The authors infer that shoulders, elbows and wrists play a decisive part in performance.

Footage was shot with a Casio EX-F1, a camera recording at 60 to 300 frames per second, from four viewpoints (above, side, front and the inner side); the archer stood in a 1 by 1 metre box marked on the ground at the shooting line, with the centre of the target 5 feet above the floor. Each archer had three trials and all were counted, but the paper does not say how many arrows per archer were scored or how the score was reduced to a mean. Angles were measured at release with Siliconcoach Pro 8 and Tracker, then correlated with score using Pearson's correlation in SPSS 20, with significance set at 0.05. The archers were 24 years old (plus or minus 5), about 6 feet tall, with FITA scores between 650 and 720.

The main problem is statistical. The five archers' scores run from 9.9 to 10, with a standard deviation of 0.044: in practice they all shot tens, and the difference between them comes down to a few hundredths of a point. With five people, if one has a slightly lower score and also an angle unlike the others', that angle will correlate almost perfectly with performance; and if more than a hundred correlations are tested at the 0.05 threshold, many will pass it by chance. Coefficients nearly all above 0.88 and up to 0.999, all positive, across very different variables point to this rather than to a genuine link between posture and score.

The text is also inconsistent in places. Two pairs of angles appear twice with different values: the drawing forearm and the line of force (7.56 and 7.05 degrees), and the bow arm and the line of force (2.97 and 7.40). In the discussion the strongest link with performance is credited to the angle between arrow and line of force, whereas in the list of coefficients the 0.998 falls on the angle between the drawing forearm and the line of force, and the arrow angle has 0.901. A bow-arm elbow at 185 degrees and a drawing elbow at 17.81 degrees imply a measuring convention the paper does not explain, just as it does not explain how the views were combined or whether the angles were measured in the image plane; there is no repeatability check of any kind. The same sample, with the same description and filming set-up, appears in another paper by the same authors on linear variables and in a 2025 anthropometric study; the methods section of this paper still refers to 'linear variables' and to a regression that never appears in the results. The registry flags the journal as having questionable editorial standards.

For anyone who shoots, this study does not say which shoulder, elbow or wrist angle earns more points. It describes five international-level archers at one precise moment, the release, and their mean values can serve as a descriptive reference: a bow-arm shoulder at about 96 degrees, for instance, and a line of force 3 to 4 degrees off the arrow. The correlations with score do not hold up, though: with five archers almost all on 10, an angle that matters cannot be told apart from one that varies by chance. A coach who used them to correct posture would be starting from information the study cannot provide.