A Retrospective Evaluation Of Team Performance Indicators In Rugby Union

AbstractThe purpose of this investigation was to demonstrate a method to validate a set of Performance Indicators (PIs). Data from one team (65 matches) was analyzed using 19 PIs categorized into six groups. Frequencies of each PI were recorded at 10- minute time intervals along with match outcomes (win/loss). Odds ratios (OR) derived from logistic regression (>2.0 or <0.5) were used to identify those PIs that contributed to match outcome. Eight PIs were subsequently identified: Ruck Quick had the largest positive effect on match outcome, within the first 20 minutes of the game and also within the 60-70 minute time interval; Territory (Intro 22) had a positive effect on match outcome, its impact was greatest in the first 20 minutes of the match. Two PIs from the Offence group (Gainline + and Gainline +P) had smaller positive effects on match outcome. Ruck Slow had the largest negative effect on winning a match for all time intervals. Other PIs that had negative effects on match outcome included turnover handling, ruck lost and turnover contact. Retrospective match data was used to investigate the efficacy of PIs. This post-hoc confirmation provides a framework for the assessment of PIs in performance analysis.logistic regressionmeasurementvalidity