The Right Way to Synchronize Tracking and Event Data:Using Domain Knowledge to Optimize Algorithms

aluable new insights can be obtained by combining tracking and event data in soccer anal-<br/>ysis. However, how to synchronize the two data streams, is rarely discussed. Non systematic<br/>errors in the timestamps, and synchronizing with cost functions result in suboptimal synchro-<br/>nization, which hinders further analysis. Within this proceedings we will introduce a com-<br/>putationally optimized implementation of the Needleman-Wunch algorithm, by using domain<br/>knowledge about the game. The optimized version is over 70 times more efficient in terms of<br/>time constraints and memory usage. On top of that, we show that the properly synchronized<br/>approach translates back to practice with better performing xG models. Taken together, this im-<br/>plementation is a training-free, high-quality synchronization algorithm, with low computational<br/>cost that solves existing issues. On top of that, all data and code used for this proceedings is<br/>fully open-sourced and available in the DataBallPy package.