T-pattern analysis in soccer games: Relationship between time and attack actions

Sports performance consists of a multiple series of strategies that tend to follow one another. Performance analysis in team sports is usually focused on primary (fundamental skill execution), secondary (scoring) and tertiary (match result) outcomes. While there is general agreement over measuring secondary and tertiary outcomes, literature does not show a unanimous agreement over a unique measure of the primary level of performance. The aim of this study was to investigate primary performance outcomes through an analysis of temporal patterns. In particular, we were interested in verifying if changes in tertiary performance outcomes may be related to changes in primary ones. We selected three soccer matches played by a top club during the Serie A league over the 2012-2013 season in which there was a change in match result between first and second half (tertiary level of performance). The methodological approach was based on observational design, supported by digital recordings and computer analysis. Data were analyzed with theme 6 beta software, which detects the temporal and sequential structure of datasets, revealing repeated patterns that may regularly or irregularly occur within a period of observation (Tpatterns). Striking differences were found comparing first and second half temporal patterns, especially when the final match outcome showed an improvement of the first half's one. Our results suggest that theme software and T-pattern enhance research opportunities by identifying a useful tool to study the link between primary and tertiary level of performance, making this an elective research and support instrument for sports analysis.