Matteo Campagna
ITEN
Menu
Menu

Body movement alone tells you which archer took the shot

1 study · 4 archers measured

ForContextAgainst202420262025, for, n = 42025, context, n = 4
Each circle is a study: its position gives the year, its row whether it is for, against or context, its size how many participants it measured. Light circles are studies on other populations or without archers.

A single 2025 study says so, on four Korean national-level archers filmed in training between February and May 2024. From 5,984 shots, from the starting position to the release, a pose estimation program extracted only the trajectories of the shoulders, elbows and wrists, and the best neural network identified who had taken the shot in 98.7 per cent of cases on the test data; in the performance table the other networks tried run from 98.5 down to 89.7 per cent, and one stops at 55.9, so the result also depends on the tool. It is a sign that each archer's movement carries a recognisable signature, in line with the studies that find different traces from one elite archer to the next. But the choice is between only four people, who may also differ in build, position relative to the camera and day of filming, and the paper does not say how the shots were split between training and testing: if near-identical shots from the same session ended up on both sides, the accuracy is inflated. No measurement is linked to the score.

Also phrased as: "every archer has their own shot"; "movement signature"; "personal shooting style"; "recognising an archer by how they shoot"; "movement fingerprint"

In practice

Archers say that everyone has their own shot, and that a coach can recognise an athlete from a distance, before even seeing their face, from the way they raise and draw the bow.

Evidence

  • 2025
    For Archers Published later

    4 Korean national-level archers, all men, 5,984 shots filmed in Full HD at 60 frames per second: using only the coordinates of the shoulders, elbows and wrists, the best network, a bidirectional GRU, attributes the shot to the right archer in 98.7 per cent of cases on the test data, never used in training. In the performance table the plain GRU reaches 98.5, the bidirectional LSTM 96.6 and the simple recurrent network 89.7. The paper does not say how the shots were split between training and testing.

    Deep Learning Sequence Network for Identifying and Analyzing Archery Shooting Patterns · Jihoon Park, Hyongjun Choi

  • 2025
    Context Archers Published later

    In the same work the plain LSTM stops at 55.9 per cent, with an F1 of 49.3, a result the authors put down to overfitting or insufficient tuning: how well the archer is recognised depends on the tool as well as on the movement. The task is choosing between four people, and no measurement is linked to the score or to where the arrow lands.

    Deep Learning Sequence Network for Identifying and Analyzing Archery Shooting Patterns · Jihoon Park, Hyongjun Choi

Status history

  1. 2025 Reported The first data come from four Korean national-level archers filmed in training over four months. From almost six thousand shots a program extracts the trajectories of the shoulders, elbows and wrists, and the best network attributes the shot to the right archer in 98.7 per cent of cases on the test data. The movement seems to carry a personal signature, but the task is choosing between four people, the network may also have learned build and framing, and it is not known how the shots were split between training and testing. A single study, on four archers: reported.

Related

Related articles

Research lines

Studies describe groups of archers, and your own numbers show how far the finding applies to your shooting.

See the full assessment
Cite
APA

Campagna, M. (2026). Body movement alone tells you which archer took the shot (Version 1). In Evidence Registry. Retrieved 16 October 2026, from https://matteocampagna.com/en/registry/technique/identifying-the-archer-from-movement

BibTeX
@misc{campagna2026_riconoscerearcieredalmovimento,
  author = {Campagna, Matteo},
  title = {Body movement alone tells you which archer took the shot},
  year = {2026},
  howpublished = {Evidence Registry, matteocampagna.com},
  url = {https://matteocampagna.com/en/registry/technique/identifying-the-archer-from-movement},
  note = {Version 1; accessed 2026-10-16}
}

Other claims in this domain

See all claims on technique and biomechanics