Team sport is a dynamic, intermittent field sport that involves high intensity bouts of running, frequent accelerations and decelerations, and complex body movements. Moreover, it is played over a prolonged time course of 60 min and includes multiple stoppages (commercial breaks and play-offs), which can extend game times to upwards of three hours. Furthermore, the physical demands of different positions within a team may differ, necessitating nuanced considerations when utilising tracking systems and associated metrics in these contexts.
The availability and affordability of tracking systems has led to an increased interest in their use for quantifying training and competition characteristics within team sports. This can aid objective decision-making for the prescription of external load, with the goal of promoting a healthy training environment and the prevention of injury [1, 2].
Practitioners often use descriptive data to inform their planning process, thereby influencing the selection of appropriate internal loads and ensuring that training is targeted towards specific competitive characteristics (e.g., game speed and distance). However, the utilisation of tracking data requires a critical process to ensure that this information is meaningful and useful in the sport’s unique context.
There are a multitude of metrics available from various tracking systems, which can be manipulated and analysed using numerous analytical techniques. As a result, there is a risk that the vast array of available metrics can cause confusion in the applied setting. Decision makers require simple, accurate and coherent feedback, allowing them to make informed decisions on the application of tracking data to the context of their particular sport and playing position.
Ongoing monitoring of the physical characteristics of a team sport also offers the opportunity to explore the evolution in these characteristics. For example, a change in the way that players restart the game after a kickoff in men’s professional Australian football has been linked to an increase in sprint distances and match flow speed.
The selection of tracking technologies and related metrics is heavily influenced by the sport’s ecological characteristics, such as playing dimensions, player density, positional characteristics and game rules. For example, GPS is impractical at elite level basketball due to the large and varying court size, whilst optical tracking and IMU devices are more suitable.
The relationship between a team sport athlete’s physical and skilled outputs can also be explored by overlaying tracking data with rich qualitative work domain analysis methods. For example, aggregated high-intensity running and handballs measured by optical tracking in elite netball have been shown to exhibit trivial relationships when analysed using linear mixed models and conditional inference trees. However, when the same data is overlaid with qualitative work domain analysis, strong associations between skilled actions and tracking data have been observed.