Zhang, G (2023) The Analysis of Team Tactical Behaviour in Football Using GNSS Positional Data. Other thesis, Liverpool John Moores University.

Preview

Text 2022guangzezhangmphil.pdf - Published Version Available under License Creative Commons Attribution Non-commercial. Download (7MB) | Preview

Abstract

Tactical analysis in football is an emerging field focused on assessing the collective movement of teams. Advanced player tracking technology systems facilitate the data collection for tactical analysis. GNSS tracking systems is currently the most popular player tracking technology in football application and is mainly used in physical monitoring. It also captures players positional information as geographic coordinates (i.e., latitude and longitude coordinates) which requires extra data pre-processing for tactical analysis as opposed to Cartesian coordinates (i.e., X, Y coordinates). Given the lack of a comprehensive workflow on pre-processing raw GNSS positional data for calculating tactical measures in previous publications, this thesis aimed to present a workflow that provides exemplar data, processing steps, potential issues, and corresponding solutions. With the presented workflow, not only sport scientists but also practitioners are able to engage in tactical analysis using GNSS tracking systems and bring in their own understanding and perspective. In other words, GNSS tracking systems could play an important role in both physical and tactical analysis in real-world application.
Collective movements and actions vary as the match progresses along. The second objective was to use GNSS positional data to compare team tactical behaviour in different phases of a competitive match. The presented workflow was applied in data pre-processing of this analysis as a proof of concept. Although team tactical behaviour in football has been widely studied in recent years, there is no previous study that analyses team tactical behaviour in phase of attack, defence, and transition, based on tactical measures measured by positional data. In this thesis, effective playing time of a professional football match was divided into phase of in possession (IP), attack-to- defence transition (ADT), out of possession (OOP), and defence-to-attack transition (DAT). Team length, width, length per width ratio (LpW ratio), surface area, stretch indices, and interpersonal distance were calculated and compared to explore the difference of team tactical behaviour between phases. The findings showed that the team tactical behaviour during each phase was in line with the offensive and defensive tactical principles. The team presented a more dispersed and wider formation while in possession than other phases. The difference of all team tactical behaviour between IP and DAT indicated the potentiality of distinguishing defence-to-attack transition from in possession when analysing offensive tactical behaviour. Moreover, there was no significant difference across all tactical measures between defence-to-attack transition and defence, which implied that a short period of time was required for the team to switch to attack mode. In the future, the difference between transitions, attack, and defence should be valued in tactical analysis. Combining multi-type data with multi-disciplinary knowledge could inform stakeholders of dynamic team moving pattern and benefit decision making process. However, data quality (e.g., positional data and synchronisation of positional data and event data) should be prioritised in this type of study.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Football; Tactical analysis; Player tracking technology; Positional data; Match phases
Subjects:	T Technology > T Technology (General) G Geography. Anthropology. Recreation > GV Recreation Leisure > GV561 Sports > GV711 Coaching G Geography. Anthropology. Recreation > GV Recreation Leisure > GV561 Sports
Divisions:	Sport & Exercise Sciences
SWORD Depositor:	A Symplectic
Date Deposited:	20 Mar 2023 09:54
Last Modified:	20 Mar 2023 09:54
DOI or ID number:	10.24377/LJMU.t.00019091
Supervisors:	Olthof, S, McRobert, A and Kempe, M
URI:	https://researchonline.ljmu.ac.uk/id/eprint/19091

View Item