Past research has tried to predict future performances in soccer (Harrop and Nevill, 2014) and identified team performance indicators (PIs) in rugby union (Robertson, Back & Bartlett, 2016). However, there are no similar approaches published for professional rugby league. Therefore this study will identify performance variables that best predict point`s difference, which is suggested to give more meaningful information on match performance than match outcome alone, in professional rugby league matches and can therefore be deemed as PIs. Opta rugby league data collected from all 27 rounds of the 2012, 2013 and 2014 Super League season amounting to 567 matches were extracted from Opta data sheets using visual basic programming and processed in Microsoft Excel (v2013, Microsoft Inc., Redmond, USA). A comprehensive list of action variables were used, these variables were made relative to each other i.e. home teams` carries subtracted by the away teams carries, this approach reduces the amount of variables included in the analysis. Action variables that related directly to points scored such as tries and conversions were removed where appropriate. Linear regression with Backwards stepwise method was used to analyse the data in IBM SPSS Statistics package (v21, IBM Corp., New York, USA). Backwards-stepwise linear regression was used to assess the ability of action variables to predict the point`s difference in professional rugby league matches. The final model had 15 action variables included with an R2 value of .877 (F (15, 551) = 261.93, p < .001). This study demonstrates a methodology that enables relevant and meaningful information to be produced that can be used by coaches and analysts. For example how performances on PIs can affect the final point`s difference e.g. a one unit increase in relative total sets leads to 1.1 points added to the team`s final score. Future research could build upon this research by analysing more matches, comparing PIs across different competitions and finally, develop positional PIs.
© Copyright 2017 Journal of Human Sport & Exercise. University of Alicante. Alle Rechte vorbehalten.