Insights from the Application of an Agent-Based Computer Simulation as a Coaching Tool for Top-Level Rugby Union

Authored by Michael K. Lauren, Kenneth L. Quarrie, David P. Galligan

Date Published: 2013

Sponsors: No sponsors listed

Platforms: No platforms listed

Model Documentation: Other Narrative Mathematical description

Model Code URLs: Model code not found

Abstract

We discuss an attempt to use a computer simulation as a method to allow coaches to develop and test tactics in the sport of rugby union. Such approaches are common in military science, and appear promising as a method in sport. An agent-based modelling methodology is used, where agents are software entities which represent players and are capable of making decisions for themselves. We discuss how such an approach allows a coach to estimate the potential success of a planned move, by creating a simulation in which the defence responds in a realistic and unscripted manner, which ultimately may help to identify ideal attacking or defensive patterns for a given circumstance. This in turn reduces the need for extended experimentation with new moves during training sessions. We give a description of how the model was employed by a top international rugby team, the All Blacks, and what lessons were learned.

Tags

Agent-based modelling Decision Making Non-Linear Dynamical Systems Rugby Football Strategic Planning