Repository logo
 

A Technological Approach to Performance Analysis in Competitive Street Skateboarding

Files

Thesis(18.35 MB)

Date

2025

Authors

Supervisor

Cross, Matt
Neville, Jonathon
Cronin, John

Item type

Thesis

Degree name

Doctor of Philosophy

Journal Title

Journal ISSN

Volume Title

Publisher

Auckland University of Technology

Abstract

Street skateboarding made its Olympic debut at the Tokyo 2020 Games and is now part of the core Olympic program. The sport’s rapid evolution, combined with its inherently subjective and freestyle nature has resulted in limited clarity around what constitutes performance and resulting outcomes. The overarching aim of this thesis was to use technological approaches to identify what constitutes successful performance in competitive skateboarding. The thesis comprised four sections: 1) understanding current knowledge and perspectives – ‘drop in’; 2) developing methods to objectively describe performance – ‘get set’; 3) applying methods to characterise and distinguish success – ‘send it’; and 4) translating findings into practice – ‘land bolts’. A scoping review of the literature (n=19 studies) identified physical and technical demands but revealed lacking understanding of what drives success in judged competition formats. This ambiguity was confirmed from the findings of a survey of sponsored Olympic (n=4, 25.5±2.89 years old) and amateur competitive skaters (n=10, 25.8±12.21 years old) which highlighted the need for clearer performance evaluation criteria. Style was rated as the top contributor to success, however the importance of its underlying objective aspects (e.g., speed, power) varied by skill-level and discipline. To obtain an objective understanding of such criteria, a notational analysis framework was developed leveraging publicly available broadcast footage, and used to capture trick, obstacle, and execution elements of trick attempts (TA) during best-trick (BT) and runs (RUN) from the Tokyo 2020 Games. Frames involving obstacle interactions were less reliable (mean absolute difference (MAD): intra=2.04-2.26, inter=3.62-4.35 frames). Take-off (MAD: intra=1.43, inter=3.32) and landing (MAD: intra=1.33, inter=1.55) frames could be reliably identified, particularly by an experienced rater. This objective method was applied to characterise and compare actions during the men’s and women’s competition. Men’s scores were greater than the women’s (β=1.71-1.85, p<0.001) and descriptively more variable; men demonstrated greater trick selection whilst women displayed mostly regular stance grind tricks, notably in RUNs. How these differences were reflected in judge preferences, overall competition outcome, and where training should be focused, was explored using several linear regression models to determine which factors contributed to score differences. RUN models (RM2: M=0.750, W=0.829) were better fit than BT for both divisions (RM2: M=0.302, W=0.520). Using larger, feature obstacles was related to scoring in BTs (βW-BT=0.86, pW-BT<0.001); all BTs in the men’s division used feature obstacles. Successful women distinguished themselves by flipping their board (βW-BT=1.10, pW-BT<0.001) and varying take-off stances (βW-BT=1.10, pW-BT=0.002); whereas successful men introduced uniqueness by combining tricks (βM-BT=1.38, pM-BT<0.001). In RUNs, there appeared a sweet spot number of TAs (βRUN=3.67 to 4.58, pRUN<0.001; 8 to 9) for a high score, albeit dictated by a quality-quantity trade-off possibly leading to the negative, non-linear relationship (βRUN=-2.00 to 1.65, pRUN<0.001). Bails were also critical; RUN scores decreased with each bail (βRUN=-0.39 to 0.40, pRUN<0.001). The findings from this thesis encompass the first objective understanding of skateboarding performance. Although more insights as to how these findings transfer across competitions is warranted, coaches can leverage the foundational performance model presented to focus training efforts and competition planning within the evolving landscape of Olympic street skateboarding.

Description

Keywords

Source

DOI

Publisher's version

Rights statement

Permanent link

http://hdl.handle.net/10292/19958

Collections

Doctoral Theses