Physical testing characteristics better explain draft outcome than in-game movement profile in junior elite Australian rules football players

Jacob Jennings; Daniel W Wundersitz; Courtney J Sullivan; Stephen D Cousins; Gerry Tehan; Michael I Kingsley

doi:10.1016/j.jsams.2021.07.005

Physical testing characteristics better explain draft outcome than in-game movement profile in junior elite Australian rules football players

J Sci Med Sport. 2021 Dec;24(12):1284-1289. doi: 10.1016/j.jsams.2021.07.005. Epub 2021 Jul 16.

Authors

Jacob Jennings¹, Daniel W Wundersitz², Courtney J Sullivan², Stephen D Cousins², Gerry Tehan³, Michael I Kingsley⁴

Affiliations

¹ Holsworth Research Initiative, La Trobe Rural Health School, La Trobe University, Australia; La Trobe University Bendigo Pioneers, Australia.
² Holsworth Research Initiative, La Trobe Rural Health School, La Trobe University, Australia.
³ School of Psychology and Counselling, University of Southern Queensland, Australia.
⁴ Holsworth Research Initiative, La Trobe Rural Health School, La Trobe University, Australia; Department of Exercise Sciences, Faculty of Science, University of Auckland, New Zealand. Electronic address: Michael.Kingsley@Auckland.ac.nz.

PMID: 34364809
DOI: 10.1016/j.jsams.2021.07.005

Abstract

Objectives: The aim of this study was to investigate the extent to which Australian Football League (AFL) draft outcome is associated with physical performance and/or in-game movement profile.

Design: Observational cohort design.

Methods: Physical testing results and in-game global positioning system (GPS) data were collated from Victorian-based, draft-eligible participants in the under 18 boys NAB League competition (n = 450; age = 17.1 ± 0.3 y). Players were grouped by position as nomadic, fixed-position or fixed&ruck.

Results: Individually, variables that best distinguish drafted and non-drafted players were: estimated V̇O₂ max (all-position, nomadic, fixed&ruck: d = 0.60, 0.64, 0.53), standing vertical jump (d = 0.57, 0.58, 0.53), running vertical jump (d = 0.52, 0.51, 0.56), AFL agility (d = 0.49, 0.44, 0.67) and 20-m speed (all-position, nomadic: d = 0.50, 0.61). Factor analysis prior to binary logistic regression assessed the probability of factors influencing position-specific draft outcome. AFL agility (all-position, fixed&ruck: OR = 4.58, 15.86), anthropometry (all-position, nomadic, fixed, fixed&ruck: OR = 2.55, 2.06, 11.41, 7.99), and jumping (all-position, nomadic, fixed&ruck: OR = 1.75, 1.69, 2.68) were the factors most associated with positive draft outcome. More game involvement (fixed&ruck: OR = 2.22), sprinting (all-position, fixed&ruck: OR = 1.45, 2.06) and less non-sprinting activities (all-position, nomadic: OR = 0.64, 0.61) were associated with positive draft outcome. The fixed&ruck model was the best performing (χ²(115) = 30.59, p < 0.001, AUC = 84.7%).

Conclusions: Physical testing-related factors were most likely to influence draft outcome, where larger and more agile players were desirable draft picks. In-game movement profile had some bearing on draft outcome in all positional groups with the exception of fixed.

Keywords: Draft; Performance; Selection; Talent identification; Team sports.

Publication types

Observational Study

MeSH terms

Adolescent
Anthropometry
Athletes* / statistics & numerical data
Athletic Performance* / statistics & numerical data
Australia
Cohort Studies
Geographic Information Systems
Humans
Male
Personnel Selection* / statistics & numerical data
Predictive Value of Tests
Retrospective Studies
Team Sports*