Passes allowed per defensive action: a measure of pressing or defensive activity?

The high press is a collective out of possession strategy designed to reduce the time and space available to opponents in possession, in the opponent's half of the pitch. To measure the high press, player and ball movements need to be tracked, and a complex algorithm developed. Consequently, a proxy measure for the high press was developed, passes allowed per defensive action (PPDA) which could be calculated without recourse to x y coordinate data, a resource only available for the highest-level football competitions. PPDA values are under reported in scientific papers and their utility unknown. The objective was to consider PPDA in relation to defensive activity encapsulated within a playing style framework. All matches in the 2017/18 English Premier League (EPL) were accessed through Opta Sportsdata. Variables analysed using Spearman's Rho, Wilcoxon signed-ranks and Factor Analysis were PPDA, ball recoveries in the middle and attacking thirds, progressive passes and possession percentage for home and away teams classified as highest, middle and lowest ranked teams. Teams with the lowest median PPDA tended to finish higher in the league (PPDA related to points gained at home (Spearman's Rho = −0.52) and away (−0.57), with colloquially recognised high press teams demonstrating low average PPDA values. PPDA was associated with home advantage, but relative team strength had a greater effect with stronger teams having lower PPDA values than weaker teams, irrespective of venue (medium and large effect sizes for Wilcoxon signed ranks tests except when the lowest ranked teams played at home against the middle-ranked teams ( p = 0.8). High possession teams tended to play a low percentage of progressive passes and a low PPDA, indicative of an indirect style of play, against an opponent with a high PPDA, suggesting a low block defence. Individual teams displayed large variations in PPDA values suggesting PPDA, in conjunction with other metrics, could be informative for describing defensive playing style, which is dependent on match context, evident here in matches where possession was unequally shared. Future research should consider the utility of PPDA in closely contested matches but in conjunction with spatial-temporal tracking metrics.