Supervised sequential pattern mining of event sequences in sport to identify important patterns of play: an application to rugby union

TL;DR

This paper applies a supervised pattern mining algorithm to rugby event sequences to identify key play patterns that differentiate scoring outcomes, providing more insightful results than traditional unsupervised methods.

Contribution

It introduces the use of safe pattern pruning (SPP) for supervised sequential pattern mining in sports, revealing more relevant and sophisticated play patterns for coaching analysis.

Findings

Identified key patterns discriminating scoring and non-scoring plays.

Supervised SPP outperformed unsupervised algorithms in pattern diversity.

Patterns included linebreaks, successful lineouts, and errors.

Abstract

Given a set of sequences comprised of time-ordered events, sequential pattern mining is useful to identify frequent subsequences from different sequences or within the same sequence. However, in sport, these techniques cannot determine the importance of particular patterns of play to good or bad outcomes, which is often of greater interest to coaches and performance analysts. In this study, we apply a recently proposed supervised sequential pattern mining algorithm called safe pattern pruning (SPP) to 490 labelled event sequences representing passages of play from one rugby team's matches from the 2018 Japan Top League. We compare the SPP-obtained patterns that are the most discriminative between scoring and non-scoring outcomes from both the team's and opposition teams' perspectives, with the most frequent patterns obtained with well-known unsupervised sequential pattern mining algorithms when applied to subsets of the original dataset, split on the label. Our obtained results found that linebreaks, successful lineouts, regained kicks in play, repeated phase-breakdown play, and failed exit plays by the opposition team were identified as as the patterns that discriminated most between the team scoring and not scoring. Opposition team linebreaks, errors made by the team, opposition team lineouts, and repeated phase-breakdown play by the opposition team were identified as the patterns that discriminated most between the opposition team scoring and not scoring. It was also found that, by virtue of its supervised nature as well as its pruning and safe-screening properties, SPP obtained a greater variety of generally more sophisticated patterns than the unsupervised models, which are likely to be of more utility to coaches and performance analysts.