An archer's score is predicted better by experience and past results than by moment-to-moment data

The summary today

The two studies that set the predictors against each other do not give a clear answer, and they are few and indirect. In an indoor competition among adolescents, shooting time and heart rate variability before each arrow together explain 11.8 per cent of the score of the single arrow, while the difference between novices and experienced archers on its own weighs about twice as much. In thirty-two Indian archers a machine learning model, trained on one session and tested on the next, is off by about six points out of three hundred, and at the top of the predictors are years of practice, first, and scores in previous competitions, third; in second place, though, is resting systolic blood pressure, taken seated before the warm-up, a measure from the day itself that ranks above past results, while heart rate around the release comes sixth and tenth. Two small Indonesian groups then show that the internal ranking changes little over time: a correlation of 0.898 between two very different training periods and of 0.985 between the start and the end of six weeks. These are associations, the comparisons between predictors come from different analyses, and the best model explains little more than half of the variability: nobody has yet compared a model with the simplest forecast, the average of the last few competitions. With a measure from the day in second place, the status is contested.

Also phrased as: "predicting a competition score"; "past scores predict future scores"; "heart rate and score"; "prediction models in archery"

Evidence

• 2011
  For Archers Published later

  7 novices (0.4 ± 0.3 years of practice) and 10 experienced archers (4.1 ± 0.9 years), adolescents in an indoor competition, 1,000 arrows: in the regression, time per shot (beta 0.311), low frequency power (0.070) and the LF/HF ratio (0.064) together explain 11.8 per cent of the score of the single arrow (R squared 0.118). The effect of experience group on the average score per arrow has an eta squared of 0.24. The experienced archers also spend longer in the final aim (28.1 against 20.2 s), so the two kinds of data are not independent.

  Autonomic nervous system modulation during an archery competition in novice and experienced adolescent archers · Andres E. Carrillo, Vasilios X. Christodoulou, Yiannis Koutedakis, Andreas D. Flouris

• 2018
  Context Archers Published later

  116 archers: the score estimated before the competition is associated with the result with an R squared of 0.305 in the group with a strong goal orientation (82 archers) and of 0.160, not significant, in the group with a weak orientation (34). Journal with questionable editorial standards.

  The Impact of Pre-Competition Estimated Results for Elite Archers on Performance According to Achievement Goal Theory Questionable venue · Utku Isik

• 2021
  Context Archers Published later

  9 young sub-elite archers from a centre in East Java, 36 arrows before and during the pandemic restrictions: an average drop of 21.7 points, but a correlation between the scores of the two periods of r = 0.898 (p = 0.001); for arm endurance r = 0.700 and for pulse r = 0.651, not significant. Conference proceedings.

  The Training Performance Of Sub Elit Adolescent Archery Athletes Before And During Covid-19 Pandemic · Fani Ravena, Ajeng Tirtasari, Irmantara Subagio, Azizati Rochmania, Kunjung Ashadi

• 2024
  Context Archers Published later

  10 middle-school pupils on an after-school archery course in Sleman (Indonesia), six weeks of meditation with no control group: correlation between initial and final score r = 0.985 (p < 0.001). Regional journal.

  Pengaruh latihan meditasi terhadap akurasi memanah siswa ekstrakurikuler panahan SMP di Kabupaten Sleman · Muh Batistuta Fitoni Nuradila, Yudik Prasetyo, Heru Prasetyo

• 2025
  For Archers Published later

  32 Indian male archers (15 elite, 17 not) at 18 m indoors, a model trained on 960 arrows from one session and tested on the 960 of the next: the best one (linear CatBoost, 35 variables) is off by 6.26 points out of 300 with an R squared of 0.575, on an average score of 279.5; the other eleven models range from 6.73 to 9.61, linear regression 7.49 with R squared 0.393. By importance, years of practice come first and previous competition scores third, while heart rate 2 s before and 5 s after the release comes sixth and tenth. Higher values for years of practice, age and past scores push the predicted score up; with quadratic heart rate terms the model gets worse (6.26 becomes 7.33, R squared from 0.575 to 0.418).

  Can we predict archery performance using shooting heart rate and other biophysical parameters? A machine learning approach · Chandra Sekara Guru, Uma Mahajan, Anup Krishnan, Karuna Datta, Deep Sharma

• 2025
  Against Archers Published later

  In the same model the second most important predictor is resting systolic blood pressure, taken seated on the day of testing before the warm-up, and it ranks above previous competition scores: a measure from the moment that weighs more than past results. The authors link it to other studies in which lower resting blood pressure goes with stronger archers.

  Can we predict archery performance using shooting heart rate and other biophysical parameters? A machine learning approach · Chandra Sekara Guru, Uma Mahajan, Anup Krishnan, Karuna Datta, Deep Sharma

Status history

1. 2011 Reported The first study to set the two kinds of data side by side follows seventeen adolescents in an indoor competition, a thousand arrows: shooting time and heart rate variability before the arrow together explain 11.8 per cent of the score of the single arrow, while experience, just under half a year against four, explains about a quarter on its own. The two numbers come from different analyses and the experienced archers also take longer per shot, so the comparison is indicative: a single study, by association, reported.
2. 2018 Reported A study of 116 archers looks at a different kind of data from the moment, the score each archer expects before the competition: it explains 30.5 per cent of the result in those with a strong goal orientation and a non-significant share in the others. The journal has questionable editorial standards and the figure does not set expectations against experience, so it stays as context and the status does not move.
3. 2021 Reported Nine young sub-elite archers followed through a normal period and a period of reduced training lose almost twenty-two points on average, but the internal ranking stays almost intact, with a correlation of 0.898 between the scores of the two periods. It shows that past results anticipate present ones even when conditions change a great deal, but it does not set them against data from the moment: it is context, and the status remains reported.
4. 2024 Reported Ten pupils on an after-school archery course, followed for six weeks, confirm how stable the ranking is: the initial and final scores correlate at 0.985. A tiny sample, with no comparison against other predictors, and the status does not change.
5. 2025 Contested Thirty-two Indian archers, a model trained on the arrows of one session and tested on the next: the error is about six points out of three hundred. At the top of the predictors are years of practice, first, and scores in previous competitions, third, but in second place is resting systolic blood pressure, taken before the warm-up: a measure from the day that ranks above past results. Heart rate two seconds before and five seconds after the release comes sixth and tenth. The study therefore points two ways, and counts once in favour and once against: the status becomes contested.

Related

Related articles

• How to improve your archery score: where to start and what to work onHow to measure your level, what sets better archers apart and the levers on your score, from release to mind and load, and how well the evidence holds.
• Talent identification: tests and assessment batteriesWhat talent tests in archery actually measure: coaches' ratings, physical tests, attention, years of practice and predictive models, and where their limits lie.
• The future of archery: technology, research, evolutionWhat has already been measured about technology in archery: virtual reality, sensors, video, the quiet eye, camera scoring and shooting machines.

Research lines

• 102 studiesFrom the total on the scorecard to the arrow's coordinates
• 48 studiesWhat predicts the score, and how talent selection tried to use it

Other claims in this domain

• CorroboratedAbout 150 points on a FITA round separate a club archer from a national-team archer13 studies · 492 archers measuredStatus since 2003
• CorroboratedAn archer's arousal rises with what is at stake, and keeps rising as the competition goes on3 studies · 23 archers measuredStatus since 2011
• CorroboratedArchers are more aroused in competition than in training5 studies · 50 archers measuredStatus since 2025
• CorroboratedAt 70 metres a top archer averages about nine points per arrow7 studies · 72 archers measuredStatus since 2012
• CorroboratedEven in competition at the highest level, most arrows are not tens4 studies · 3,212 archers measuredStatus since 2024

See all claims on competition and scoring