Meeting Abstract
The outcome of any interaction between competing agents is determined by the physical capabilities of each combatant and the strategy they employ – whether this is for animals in natural settings or humans in sport. By measuring physical capabilities and defining strategies for both parties in a dyadic duel, we can see how these interact to determine the probability of success for each individual. Applying optimisation theory within a game-theoretic context allows us to identify the strategy for each individual that maximises their probability of success. We have used soccer penalties as a model system. For shooters, the relevant physical capabilities are their kicking accuracy across a range of speeds and their strategy can be defined by where they aim, how fast they kick and how much deception they use. For goal-keepers, their physical capabilities are their speed of movement and ability to predict shot direction, and their strategy can be defined by when they choose to dive relative to the ball being kicked and their choice of direction. We have experimentally measured these variables and developed a model that identifies the optimal strategy for both shooters and goal-keepers. Throughout this project a number of interesting performance trade-offs have been investigated – How does speed affect accuracy? Does kicking the ball harder make the shooter more predictable? Does using deception really work – does it decrease accuracy? We discuss the implications of our results for both soccer penalties and other games between competing agents when there are trade-offs between performance traits.
