The physical demands of Super 14 rugby union

Available online at www.sciencedirect.com Journal of Science and Medicine in Sport 14 (2011) 259 263 Original research The physical demands of Super 14 rugby union Damien Austin a,b,, Tim Gabbett a,c, David Jenkins a a The University of Queensland, School of Human Movement Studies, Australia b Sydney Roosters Rugby League Football Club, Australia c Brisbane Broncos Rugby League Football Club, Australia Received 14 February 2010; received in revised form 3 December 2010; accepted 20 January 2011 Abstract Objectives The purpose of the present study was to describe the match-play demands of professional rugby union players competing in Super 14 matches during the 2008 and 2009 seasons. Design The movements of 20 players from Super 14 rugby union team during the 2008 and 2009 seasons were video recorded. Methods Using time-motion analysis (TMA), five players from four positional groups (front-row forwards, back-row forwards, inside backs and outside backs) were assessed. Results Players covered between 4218 m and 6389 m during the games. The maximum distances covered in a game by the four groups were: front row forwards (5139 m), back row forwards, (5422 m), inside backs (6389 m) and outside backs (5489 m). The back row forwards spent the greatest amount of time in high-intensity exercise (1190 s), followed by the front row forwards (1015 s), the inside backs (876 s) and the outside backs (570 s). Average distances covered in individual sprint efforts were: front row forwards (16 m), back row forwards (14 m), inside backs (17 m) and outside backs (18 m). Work to rest ratios of 1:4, 1:4, 1:5, and 1:6 were found for the front row and back row forwards, and inside and outside backs respectively. Conclusion The Super 14 competition during 2008 and 2009, have resulted in an increase in total high-intensity activities, sprint frequency, and work to rest ratios across all playing positions. For players and teams to remain competitive in Super 14 rugby, training (including recovery practices) should reflect these current demands. 2011 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved. Keywords: Time motion analysis; Training; High-intensity exercise 1. Introduction Since turning professional in 1995, rugby union has experienced a number of law changes in order for the game to remain attractive to spectators and be competitive with other football codes. 1 Indeed, evidence 2,3 suggests that the game has become faster and more physically demanding as a result. While studying under-19 rugby union players (who played 70 min games), Deutsch et al. 4 found that props and locks covered 4400 m during a match, while the average distances for the back row forwards, inside backs and outside backs were 4080 m, 5530 m, and 5750 m respectively. The authors found that players spent 85% of their time in low-intensity activities and 15% in high-intensity activities. In a subsequent investigation of the match demands on 16 forwards and 12 Corresponding author. E-mail address: Damiena@sydneyroosters.com.au (D. Austin). backs in ten Super 12 rugby games during the 2001 season, Duthie et al. 5 found forwards to have higher work to rest ratios than backs (1:6 vs. 1:20) and while both front-row and back-row forwards had the same work to rest ratio, the inside backs averaged 1:15 while the outside backs averaged 1:24. Duthie et al. 6 also found that Super 12 rugby players performed most of their sprints over distances between 10 m and 20 m; forwards sprinted on average 15 m on 13 occasions, and backs 20 m on 24 occasions. In a more recent study with players in the English Premiership competition, Roberts and colleagues 6 found average distances covered by the tight forwards to be 5408 m, the loose forwards 5812 m, the inside backs 6055 m and the outside backs 6190 m. These distances are markedly higher than those reported by Deutsch et al. 4 in the under-19 competition, and this is likely to be due, at least in part, to differences in the duration of the games (i.e., 70 vs. 80 min). Eaton and George 8 also studied the demands of players in the English 1440-2440/$ see front matter 2011 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jsams.2011.01.003

260 D. Austin et al. / Journal of Science and Medicine in Sport 14 (2011) 259 263 Premiership competition and found work to rest ratios of 1:8 for the forwards and 1:13 for backs. They also found the average sprinting distances for forwards ranged from 6 m to 9 m, while for backs the range was between 13 m and 15 m. Roberts et al. 7 found that backs sprinted an average of 23 times during a match, while the forwards sprinted on 16 occasions. Differences in the findings reported by Duthie et al., 5 Eaton and George 8 and Roberts et al. 7 may be due to differences in the styles of play between the competitions studied and also differences in the interpretation of the laws by the Northern and Southern Hemisphere referees. There were also differences in methodology with regard to analysing player movements; Eaton and George 8 and Roberts et al. 7 used computer analysis to match locomotor activities with speed settings based on m s 1, while Duthie et al. 5 used video analysis and measures of locomotion based on assessor interpretation. Nonetheless an increase in the demands on players would be consistent with how, at least anecdotally, observers believe the game has evolved over the past decade. 8 While research has established that rugby union is a game characterized by periods of high and low-intensity intermittent activities, research is yet to examine the physiological demands on players in the Super 14 competition. The purpose of the present study was therefore to examine the matchplay demands (e.g. the frequency, duration, and distance of locomotor and discrete activities) in Super 14 rugby union. 2. Methods The movements of 20 players from Queensland s rugby union team during the 2008 and 2009 seasons were video recorded; seven Super 14 games played at Suncorp Stadium, Brisbane, Australia were assessed. Players were clustered into each of the four positional groups used by Duthie et al. 5 ; these were front row forwards (age 23 ± 2 years, height 183 ± 2 cm and weight 114 ± 4 kg), back row forwards (age 26 ± 3 years, height 183 ± 4 cm and weight 103 ± 9 kg), inside backs (age 22 ± 1 years, height 179 ± 6 cm and weight 87 ± 3 kg) and outside backs (age 24 ± 3 years, height 182 ± 4 cm and weight 100 ± 12 kg). Five players in each group were analysed and filmed randomly over seven games. Players were not filmed more than once during the study. Only one player from each of the positional groups was filmed per match. Seven games account for a 50% of the Super 14 season and could be considered a reasonable sample size. The study was approved by an Ethics Committee of The University of Queensland and all subjects involved in the study provided informed consent prior data collection. Video recordings were made using three video cameras (HITACHI DZ-GX5060SW, Hitachi LTD., Japan) positioned on stationary tripods. Cameras were positioned on the halfway line approximately 30 m above the playing field. Each camera operator followed one player for the entire duration of each match investigated. Each player was filmed with at least a 10-m radius around him so as to allow a view of his surroundings and field position. Each was filmed for the entirety of the match, including breaks in play and time on the bench. The filming of seven matches allowed for the averaging of data for each positional group, differences in environment, oppositional standard and referee interpretation. Each movement was coded as one of eight categories of locomotion (standing, forward walking, backward walking, forward jogging, backward jogging, forward striding, forward sprinting, and lateral movement), non-running intense activities (e.g. tackling, static holds or lifts, and scrimmage) or one discrete activity (e.g. kicking). These extended previously used definitions by Duthie et al. 5 and the codes used are provided in Table 1. Video footage was analysed using a simple hand-notation game analysis system; for each movement activity, frequency, distance covered and duration were recorded. The information was transferred to a specifically designed computer program for analysis. Initiation and completion of each movement were established and the duration and distance of each activity calculated subjectively, based on the movement definitions and standard line markings of a rugby union field. Reliability for the analysis was considered moderate to high. High-intensity work was considered to include forward striding, forward sprinting, tackling, static holds, and scrummaging (scrums, rucks and mauls). Low-intensity work included standing, forward and backward walking, lateral movement, and forward and backward jogging. Kicking was not included in the intensity analysis. Low-intensity activities were considered rest periods as opposed to high-intensity activities that were classified as work. The duration of each interval of high-intensity work was divided by the duration of the following rest interval. This gave a work to rest ratio for that passage of play or game. All match information relates to the full duration of a game. Reliability of the time-motion analysis was primarily assessed by re-analysing approximately 40 min of footage from one player four weeks apart. In addition, a second researcher assessed the same 40 min; this was used to determine the accuracy of the primary investigator in assessing the video footage. Each experimenter was denied knowledge of results by the others and approximately four weeks separated the first and second analyses. 9 The reliability of the coding technique was assessed using the typical error of measurement (TE) and also expressed as a coefficient of variation (CV). 10 Previous studies have applied this method of reliability to TMA in rugby union 6 and field hockey. 11 TE and CV calculations were based on four measurements (frequency, total time, relative time (%), and relative distance (%)) of the 12 key movements (standing, forward walking, backward walking, forward jogging, backward jogging, striding, sprinting, lateral, tackling, static holds, scrummaging and kicking). Absolute TE values for frequency, total time, relative time (%), and relative distance (%) were 0.04, 0.39, 0.27, 0.62 respectively for inter-coder reliability and 0.2, 1.17, 0.1,

D. Austin et al. / Journal of Science and Medicine in Sport 14 (2011) 259 263 261 0.21 respectively for intra-coder reliability. CV (%) values for frequency, total time, relative time (%), and relative distance (%) were 0.4%, 4.5%, 5.6%, 4.1% respectively for intercoder reliability and 1.2%, 2.9%, 4.3%, 3.7% respectively for intra-coder reliability. Data for duration, percentage time, total time, number of activities, and mean duration and frequencies for activities and high-intensity efforts, are presented as the mean and standard deviation (SD). Analysis of variance (ANOVA) was used to establish differences in the variables between each playing position. Residuals from the ANOVA were checked for normality. A difference between front row forwards, back row forwards, inside backs, and outside backs, resulted in a posthoc test (Tukey) being used to identify where the differences were.. All statistics were run on SPSS (v17.0 for Windows) with an alpha of 0.05 set a priori. 3. Results The mean (±SD) total distances covered each game by the front row forwards, the back row forwards, the inside backs and the outside backs were 4662 ± 659 m, 5262 ± 131 m, 6095 ± 213 m, and 4774 ± 1017 m respectively (Table 2). The front row forwards and outside backs covered significantly less distance than the inside backs (P < 0.05 and P < 0.05 respectively). The maximum distances covered in a game by the four groups were: front row forwards (5139 m), back row forwards, (5422 m), inside backs (6389 m) and outside backs (5489 m). Forward walking and forward jogging comprised 65% of the total distance covered by the front row forwards; these same two activities comprised 63% of the total distance for the back row forwards, 56% for the total distance inside backs and 58% of the total distance for the outside backs. Striding and sprinting contributed to 31%, 32%, 38% and 33% respectively of the total distances covered by the front row and back row forwards and the inside and outside backs. There was a significant difference in total striding distance covered by the inside backs (1399 ± 91 m) compared to both the front row forwards (940 ± 257 m) (P < 0.05) and outside backs (864 ± 356 m) (P < 0.05) (Table 3). Total average sprinting distances covered by the four groups were: front row forwards (501 ± 163 m), back row forwards (547 ± 55 m), inside backs (918 ± 253 m) and outside backs (558 ± 282 m). The only significant difference in sprinting distance was between the front row forwards and inside backs (P < 0.05). The average match time for front row and back row forwards, and inside and outside backs was 84 ± 9 min, 92 ± 9 min, 89 ± 1 min and 72 ± 20 min respectively. There was a significant difference in total average striding time for the inside backs (426 ± 17 s) compared to both the front row forwards (285 ± 72 s) (P < 0.05) and outside backs (249 ± 101 s) (P < 0.05). Average total sprinting times for the four groups were: front row forwards (110 ± 36 s), back row forwards (141 ± 8 s), inside backs (203 ± 47 s) and outside backs (131 ± 74 s). The only significant difference (P < 0.05) in total time sprinting was between the inside backs and front row forwards. There was no significant difference between the four positional groups in time spent tackling. As would be expected given the specific positional demands, the front row and back row forwards had significantly greater (P < 0.05) average durations for static holds (11 ± 8 s and 15 ± 9 s respectively) compared to the inside and outside backs (0.0 ± 0.0 s each). The front row and back row forwards also had significantly greater (P < 0.05) average total duration involved in rucks and mauls (554 ± 153 s and 606 ± 194 s respectively) compared to the inside and outside backs (165 ± 10 s and 120 ± 52 s respectively). The back row forwards spent the greatest amount of time in high-intensity exercise (1190 ± 241 s), followed by the front row forwards (1015 ± 222 s), the inside backs (876 ± 161 s) and the outside backs (570 ± 91 s). Work to rest ratios of 1:4, 1:4, 1:5, and 1:6 were found for the front row and back row forwards, and inside and outside backs respectively. There was a significant difference (P < 0.05) in the number of times players strode 40 m between inside backs and outside backs compared to front and back row forwards (11 ± 1 and 15 ± 1, and 6 ± 5 and 8 ± 3 respectively). Outside backs were found to stride 5 m significantly less (P < 0.05) than the front row and back row forwards and inside backs (5 ± 1, 13 ± 2, 8 ± 2 and 13 ± 1 respectively) (Table 4). The outside backs sprinted 40 m significantly more times than the front row and back row forwards (4 ± 2, 1 ± 1 and 1 ± 1 respectively; P < 0.05). The outside backs sprinted 5 m, significantly fewer times (P < 0.05) than the front row and back row forwards and inside backs (3 ± 1, 6 ± 1, 6 ± 1 and 5 ± 1 respectively) (Table 4). Average distances covered in individual sprint efforts were: front row forwards (16 m), back row forwards (14 m), inside backs (17 m) and outside backs (18 m). Mean frequency of high-intensity running during matchplay for the front row forwards, back row forwards, inside backs and inside backs was 43 ± 12, 52 ± 14, 56 ± 28 and 41 ± 15 respectively for striding and 29 ± 10, 31 ± 6, 44 ± 22 and 27 ± 12 respectively for sprinting. Mean frequency of non-running high-intensity activities during match-play for front row forwards, back row forwards, inside backs and outside backs was 20 ± 4, 19 ± 4, 25 ± 13 and 20 ± 7 respectively for tackling and 62 ± 13, 68 ± 15, 17 ± 7 and 14 ± 5 respectively for scrummaging. 4. Discussion The aim of the present study was to examine the matchplay demands of Super 14 rugby union players competing in matches played in 2008 and 2009. The present data suggest that the current match-play demands have resulted in play-

262 D. Austin et al. / Journal of Science and Medicine in Sport 14 (2011) 259 263 ers spending more time in high-intensity running activities; sprint frequency, and work to rest ratios increasing across all playing positions to previous studies. 6 8 The maximum distances covered in the present study ranged from 5139 m to 6389 m. These are less than the distances (6680 7227 m) found recently by Cunniffe et al. 12 who used GPS to assess distances and game activities for two elite players from the UK in a trial game but are similar to the distances reported by Roberts et al. 7 who found that tight forwards covered 5408 m, loose forwards 5812 m, inside backs 6055 m and outside backs 6190 m. Roberts et al. 7 assessed 40 min of match-play and doubled the values to gain estimates for an 80 min game. Although matches in the English Premiership are played under slightly different laws and the methods of analysis used by other research groups differ, the available data collectively show an increase in total distances covered by modern professional rugby union players. Players studied in the present investigation were found to spend significantly less time standing and jogging, and greater relative time in high-intensity activities than the Super 12 players studied in the 2000 and 2001 season 5 with similar methodologies used. In all four playing groups, the present study has found that players during the 2008 and 2009 seasons spent, on average, 7% less time standing, 4% less time jogging, but 4% more time striding, 2% more time sprinting and 2.5% more time in non-running intense activities compared to the Super 12 players studied by Duthie et al. 5 Front row forwards in the present study were found to spend 21% of their time striding, sprinting, tackling or scrummaging. This compares to 19% for the back row forwards, 17% for the inside backs and 14% for the outside backs. These estimates are significantly higher than those reported by Eaton and George, 8 and Roberts et al., 7 where forwards on average spent a greater amount of time in high-intensity activities compared to the backs (11% and 7%, and 11.5% and 3.8%; Eaton and George, 8 and Roberts et al. 7 respectively). Differences in the percentages between studies could be due to the difference in law interpretation and/or methodological differences between the investigations (including differences in how various movements were assessed). The average number of sprints for all positional groups in the present study was 40, which is significantly greater than the average of 18 reported by Duthie et al. 5 Given that the present study used a similar methodology to that used by Duthie et al., 5 the increase in the number of sprints suggests that Super rugby is faster than it was in 2000/2001. The average numbers of sprints for the present positional groups were: front row forwards (33), back row forwards (39), inside backs (53) and outside backs (36). These compare to 8, 14, 29 and 24 sprints per game found by Duthie et al. 6 for the front row forwards, the back row forwards, the inside backs and the outside backs respectively. Eaton and George 8 reported that forwards sprinted on average four times, while backs averaged 13 sprints in a match. This is substantially less than the values reported by Roberts et al., 7 who also studied the English Premier competition and who found that forwards and backs averaged 16 and 23 sprints respectively each game. This suggests that changes in styles of play, law changes and the intensity of the game has increased since Duthie et al. 5 studied the Super 12 competition in 2000 and 2001. While the present data impress the need for players to concentrate their sprint training over distances less than 20 m, some longer sprints (30 60 m) must also be accommodated. In addition, training for high-intensity exercise needs to mimic as closely as possible the highly variant and position-specific activities such as tackling, rucking, mauling and scrummaging. Players assessed in the present study completed most sprints over distances between 11 m and 20 m, which are similar to the distances reported by Duthie et al. 6 but longer than those reported by Roberts et al. 7 who found that forwards sprinted on average 10 m and backs 9 m. Inside backs in the present study were found to cover significantly more of their sprints over distance between 11 m and 20 m compared to the other positional groups. For the outside backs, 20% of their sprints were over 20 m and 8% were over 30 m. Both inside and outside backs completed more sprints over 40 m than front row and the back row forwards. The present work to rest ratios were different to those reported by Duthie et al. 5 ; present work to rest ratios of 1:4, 1:4, 1:5 and 1:6 were found for the front row and back row forwards, and inside and outside backs respectively. In comparison, Duthie et al. 5 reported the same positional groups to have work to rest ratios of 1:7, 1:6, 1:15 and 1:21. Deutsch et al. 13 results are comparable to Duthie et al., 5 who quantified movement patterns of 29 players in eight Super 12 competition games in a similar methodology to the current study, finding work to rest ratio s of 1:7, 1:8, 1:21 and 1:23 for front row forwards, back row forwards, inside backs and outside backs. A recent study by Eaton and George 8 in the English Premier competition reported similar data to those of Duthie et al. 5 with work to rest ratios for forwards and for backs of 1:8 and 1:15 respectively. In the present study, minimum works to rest ratios were: front row forwards (1:4), back row forwards (1:3), inside backs (1:4) and outside backs (1:5). Differences in the work to rest ratios between the present Super 14 data and those from the English Premiership suggest that the law changes may be at least partly responsible for different demands on players competing in the Super 14 competition, however differences in the methodology used by Eaton and George 8 to assess movement activities relative to speed (m s 1 ), could account for some of the differences between the studies. The significant decrease in rest periods for both the inside and outside backs identified in the present study and in the study by Eaton and George 8 tends to suggest a quicker distribution of ball to the backs, due to shorter breakdowns, and/or changes to the strategies teams use in match-play. In summary, since the work undertaken by Duthie et al. 5 on Super 12 players in 2000 and 2001, the physiological demands on players in match-play have changed. Differences

D. Austin et al. / Journal of Science and Medicine in Sport 14 (2011) 259 263 263 in sprint data between the present investigation and recent work with players from the English Premiership suggest that the modern game of rugby union, and in particular the Super 14 competition during 2008 and 2009, have resulted in an increase in total high-intensity activities, sprint frequency, and work to rest ratios across all playing positions. Practical implications The training regimes of strength and conditioning coaches need to meet the adjusted work to rest ratios of specific positional play. Need to reflect the increased speed at which the game is now played. The increased demands on players at the elite level mean greater physiological fatigue; recovery practices are therefore particularly important. Acknowledgements There has been no financial assistance with the project. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jsams. 2011.01.003. References 1. Duthie G, Pyne D, Hooper S. The applied physiology and game analysis of rugby union. Sports Med 2003;33:973 91. 2. Brooks J, Kemp S. Recent trends in rugby union injuries. Clin Sports Med 2008;27:51 73. 3. Quarrie K, Hopkins W. Changes in player characteristics and match activities in Bledisloe Cup rugby union from 1972 to 2004. J Sports Sci 2007;25:895 903. 4. Deutsch M, Maw G, Jenkins D, et al. Heart rate, blood lactate and kinematic data of elite colts (under-19) rugby union players during competition. J Sports Sci 1998;16:561 70. 5. Duthie G, Pyne D, Hooper S. Time motion analysis of 2001 and 2002 Super 12 rugby. J Sports Sci 2005;23:523 30. 6. Duthie G, Pyne D, Marsh D, et al. Sprint patterns in rugby union players during competition. J Strength Cond Res 2006;20:208 14. 7. Roberts S, Trewartha G, Higgitt R, et al. The physical demands of elite English rugby union. J Sports Sci 2008;26:825 33. 8. Eaton C, George K. Position specific rehabilitation for rugby union players. Part I. Empirical movement analysis data. J Phys Ther Sport 2006;7:22 9. 9. Duthie G, Pyne D, Hooper S. The reliability of video based time motion analysis. J Hum Mov Stud 2003;44:259 72. 10. Mueller WH, Martorell R. Reliability and accuracy of measurement. In: Lohman TG, Roche AF, Martorell R, editors. Anthropometric Standardisation Reference Manual. Champaign, IL: Human Kinetics; 1988. p. 83 6. 11. Spencer M, Lawerence S, Rechichi C, et al. Time-motion analysis of elite field hockey during several games in succession: a tournament scenario. J Sports Sci 2005;4:382 91. 12. Cunniffe B, Proctor W, Baker J, et al. An evaluation of the physiological demands of elite rugby union using global positioning system tracking software. J Strength Cond Res 2009;23(4):1195 203. 13. Deutsch M, Kearney G, Rehrer NJ. Time-motion analysis of professional rugby union players during match-play. J Sports Sci 2007;225:461 72.