Abstract
The aim of this study was to evaluate the effect of tactical and contextual dimensions on the offensive performance, in terms of degree of penetration or creation of goal scoring opportunities (GSO), of direct attacks in the English Premier League (EPL) and Spanish La Liga. In total, 10,078 team possessions from 40 random La Liga and EPL matches were initially selected. From that, 2,384 possessions categorized as open play direct attacks were selected for analysis, including a first univariate binary regression analysis and a further construction of adjusted binary multivariate logistic regression models. Four independent tactical dimensions ("Initial zone," "Initial opponent position," "Initial opponent pressure," and "Width of the possession") and four independent contextual dimensions ("Match location," "Match status," "Quality of opponent," and "Quality of the observed team") were initially analyzed to predict penetration performance on one hand and scoring opportunity performance on the other hand. The results showed that the distribution of the type of attacks was different in La Liga and EPL (χ (2) = 11.015, p = 0.001), but direct attacks were not different in La Liga and EPL in terms of performance. Three of the four tactical dimensions ("Initial zone," "Initial opponent position," and "Width of the possession") showed association with "Penetration performance" (p < 0.01), while only "Width of the possession" showed association with "Scoring opportunity performance" (p < 0.01). Two of the four contextual dimensions ("Match status" and "Quality of observed team") showed association with "Penetration performance" (p < 0.05), while none of them showed association with "Scoring opportunity performance." In conclusion, direct attacks are frequent in EPL and La Liga but very ineffective offensively in terms of creation of goal scoring opportunities. The tactical dimensions that explain a higher effectiveness in terms of getting penetrative performance are vertical progression through the field, attacking against a defensive low-block, as well as starting the team possessions from the defensive zones instead of middle zones, while the only tactical dimension that explains higher performance in creating GSOs is vertical progression. Contextual dimensions, such as "Match status" and "Team level," also influence the offensive penetration performance during direct attacks but not GSO performance.