Use this URL to cite or link to this record in EThOS:
Title: Methodological advances for assessing individual and team performance in elite rugby union
Author: Jones, Nicholas Michael Pringle
Awarding Body: University of Wales, Swansea
Current Institution: Swansea University
Date of Award: 2006
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Appropriate methodologies were devised for constructing and assessing individual and team performance in rugby union with an emphasis on providing practical solutions for the coach or performance analyst. The Noldus 'Observer Video-Pro5 analysis system (Noldus Information Technology, 1995) was used for all data collection with appropriate tests for reliability and validity conducted. Parsons and Hughes (2001) indicated varying skill demands of different rugby playing positions but within position differences were not investigated. The first study consequently created performance profiles for all playing positions, via the use of individual performance indicators (Pis), using all players used in 22 matches of the domestic season of a professional male rugby union team. A novel transformation to account for the time a player spent on the field was devised. Furthermore it was suggested that the appropriate descriptive statistics for presenting non-parametric summary information was via the median and confidence limits for a population median. Significant differences were observed between individuals of the same position for the most frequently performed Pis (passing, carrying and tackling for forward positions and passing, carrying, tackling and kicking for the backs) of all the playing positions tested. The findings suggested that while general positional performance profiles appear to exist, intra-positional differences may occur due to variations in an individual's style of play and physical attributes. Hunter and O'Donoghue (2001) suggested specific indicators that differentiated successful and unsuccessful rugby union teams, although between team differences may have contributed to this. Twenty further matches of the same team used for study 1 were analysed using additional Pis designed to analyse team performance. Some of the existing individual Pis were also modified, with off the ball behaviours added to enhance individual profiles. Only two of the team Pis (lineout success on the opposition throw and tries scored) revealed statistical differences between winning and losing performances although a general trend of superior performance was found when the team was winning. The individuality within positional roles that was found in study 1 was further tested and revealed that only one of 13 players' ball-in-hand behaviours differed significantly between two different seasons despite a considerable change in the remainder of the playing personnel. Bracewell (2003a) used control charts to create individual performance scores although no attempt was made to encapsulate team performance. Thus, objective methods of scoring team performance were presented using a single score measure of performance through the use of PI weightings (study 3a), and secondly via the combination of comparative scores for a match (study 3b) using the same 20 matches as study 2. Study 3a calculated a single score using PI weightings based on correlation coefficients between 31 Pis and two elite coaches' assessment of overall match performance. These coefficients squared were multiplied with the performance value of each PI in a given match and combined to form the single score. Of the models tested, the one containing all Pis was found to have the smallest mean bias for scores out of 100 for both wet (4.18) and dry (1.14) conditions, a high correlation (r= 0.77 wet, 0.85 dry) and no significant difference (p= 0.35 wet, 0.88 dry) with the coach scores. This suggested that the model predicted coach scores and thus match performance well, although some variance remained. Further work is needed to assess the applicability of this approach, preferably using coach evaluations for validation purposes only. Study 3b used 18 Pis from the 20th match of the same sample, standardised relative to the previous 5 and 19 matches producing distributions of median 50 and interquartile range 15. The standardised values were plotted on a 'form chart' to provide a visual assessment of each PI on one scale. This, coupled with non-standardised descriptive statistics, provided comprehensive and simple to understand feedback on performance relative to previously accomplished standards that can easily be used within a practical setting for any multi-faceted sport. This thesis has investigated individual and team Pis and found that rugby union performance is best characterised by a number of comparative Pis. Future research needs to utilise this methodology to assess comparative strengths and weaknesses between different teams.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available