Models for Performance in Sport

Examples relating to sport are used as illustrations in this e-book. Most of these examples are based on simulations that were developed by Larry Weldon from Simon Fraser University in Vancouver, Canada in order to show students that even simply-stated probability models can give valuable (and sometimes surprising) insight into sporting performance.

When individuals or teams compete in sporting events, we usually concentrate on the skill levels of the competitors. However there is also a considerable random component in sport results — when matches are replayed, the results are often markedly different. So how much do the reported results reflect skill levels and how much are they simply random artifacts?

We will use simulations to investigate this question in a few sporting contexts.

Tennis

In the first section, simulation methods are introduced with a simple model for a tennis match.