International Journal of Sports Physiology and Performance, 2015, 10, 732 -739 http://dx.doi.org/10.1123/ijspp.2014-0541 © 2015 Human Kinetics, Inc.
Original Investigation
Analysis of Physical Collisions in Elite National Rugby League Match Play Cloe Cummins and Rhonda Orr Objective: To investigate the impact forces of collision events during both attack and defense in elite rugby league match play and to compare the collision profiles between playing positions. Participants: 26 elite rugby league players. Methods: Player collisions were recorded using an integrated accelerometer in global positioning system units (SPI-Pro X, GPSports). Impact forces of collisions in attack (hit-ups) and defense (tackles) were analyzed from 359 files from outside backs (n = 78), adjustables (n = 97), wide-running forwards (n = 136), and hit-up forwards (n = 48) over 1 National Rugby League season. Results: Hit-up forwards were involved in 0.8 collisions/min, significantly more than all other positional groups (wide-running forwards P = .050, adjustables P = .042, and outside backs P = .000). Outside backs experienced 25% fewer collisions per minute than hit-up forwards. Hit-up forwards experienced a collision within the 2 highest classifications of force (≥10 g) every 2.5 min of match play compared with 1 every 5 and 9 min for adjustables and outside backs, respectively. Hit-up forwards performed 0.5 tackles per minute of match play, 5 times that of outside backs (ES = 1.90; 95% CI [0.26,3.16]), and 0.2 hit-ups per minute of match play, twice as many as adjustables. Conclusions: During a rugby league match, players are exposed to a significant number of collision events. Positional differences exist, with hit-up and wide-running forwards experiencing greater collision events than adjustables and outside backs. Although these results may be unique to the individual team’s defensive- and attacking-play strategies, they are indicative of the significant collision profiles in professional rugby league. Keywords: global positioning system, GPS, impact force, hit-up, tackle Portable global positioning system (GPS) devices coupled with integrated microsensor technology (accelerometer, magnetometer with or without gyroscope sensors) have become a commonly used measurement tool in team-sport settings to describe player activity profiles. Output measures from these devices have evolved from the examination of speed, distance covered, and high-intensity efforts (movement demands) to the quantification of collisions and body loads (physical loads). To date, collision forces associated with the physical demands of collision sports have been reported in rugby league,1–4 rugby union5–7 and Australian football.8 An understanding of player collisions and the physical demands in team contact sports is important for monitoring training, performance, and injury prevention to effectively manage and prepare players for the rigors of competitive match play. The validity and reliability of accelerometer data from GPS and integrated accelerometry technology in the sporting environment have recently been collected for instantaneous velocity during accelerations and decelerations9 and foot strikes.10 One commercially available GPS unit (Catapult MinimaxX) enables automated tackle and collision detection using algorithms that incorporate changes in unit orientation and spikes in accelerometer data. Although validated for quantifying the number and intensity of collisions in rugby league,11 these units have not been effective in detecting tackles in Australian football.12 The validity of the automatic collision detection of another GPS unit (GPSports) to measure the contact loads and physical collisions in rugby codes The authors are with the Faculty of Health Sciences, University of Sydney, Lidcombe, NSW, Australia. Address author correspondence to Cloe Cummins at [email protected].
732
has not been reported. In rugby union, however, the automatic detection of physical contacts has been explored through the use of static window features and a mathematical learning grid that created algorithms through classifying the accelerometry signals of tackle and nontackle events.13 That study demonstrated that the GPSports accelerometer is capable of detecting collisions in rugby union. In addition, GPSports units (SPI-ProX II) have recently been shown to demonstrate high intra-accelerometer and interaccelerometer reliability.14 Only small to moderate (ES < 0.5) differences were reported between accelerometers, with no significant differences seen within the interaccelerometer reliability. This high reliability indicates that these units are sufficient in providing consistent acceleration measures. SPI-ProX II accelerometers also consistently measured at a lower magnitude of acceleration than a criterion reference accelerometer, suggesting the potential for errors in the magnitude of accelerometer measures. The authors of that study posit that this difference may be confounded by the accelerometer hardware or data-processing procedures.14 Rugby league players require the ability to perform and withstand physical collisions during both defensive and attacking phases of play.11 The defensive play of tackling is integral in rugby league, where the ability to effectively perform a winning tackle may prove crucial in determining match outcome.15 The attacking play of hitups and the ability to tolerate physical collisions is important for gaining distance and field position. Not only is the total number of collisions sustained in a match important, but also the collision type and magnitude are vital considerations for match outcome. The use of GPS and integrated triaxial accelerometers to quantify forces associated with collision events is an emerging area of research.16 In rugby league, the total number and intensity
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
Collisions in Elite Rugby League Matches 733
of collisions experienced by players during high-velocity collisions in match play1–4 and training1 have been noted. The difference in collisions experienced during attacking (hit-up) and defensive (tackle) match play within a team as a whole4 and between forward playing positions only3 have also been reported. Forwards sustained an average of 55 collisions (39 tackles, 16 hit-ups) compared with 29 collisions (16 tackles, 13 hit-ups) experienced by backs.17,18 Furthermore, hit-up and wide-running forwards recorded a higher number of collisions than both adjustables and outside backs during the course of a match.1,17,19 However, little is known about the collision magnitude and collision type for both back and forward players in elite rugby league. A greater understanding of collisions experienced by players and a quantification of the forces associated with these events will enable position-specific collision profiles to be determined, assisting to inform the most appropriate collisiontraining regimens. To our knowledge, no study has explored the differential collision forces between the 4 positional groups in a team. Furthermore, the total number of collisions, collision type, and collision magnitude are yet to be fully elucidated for different playing groups. Therefore, the objective of this study was twofold: to investigate the forces of collision events during both attacking and defensive play in elite rugby league match play and to compare the collision profiles between forward and back playing positions using GPS technology.
Methods Subjects Twenty-six elite male rugby league players, (mean ± SD age 24.2 ± 2.8 y, height 184.7 ± 6.5 cm, and body mass 99.0 ± 9.5 kg) representing 1 National Rugby League (NRL) team participated in this study. The study was approved by the University of Sydney Human Research Ethics Committee (Protocol number 14020), in accordance with the Declaration of Helsinki. All participants gave written informed consent before participating.
Methodology Players were categorized into 1 of 4 positional groups representing the outside backs (centers and wingers), adjustables (hookers, halfbacks, five-eighths, and fullbacks), wide-running forwards (second rowers and locks), and hit-up forwards (props).1,17,20,21 Publicly available television footage from each NRL match was used to determine attacking and defensive match-play events. A tackle was defined as occurring when the ball carrier was held by 1 or more opposing players and either the ball or hand of the arm holding the ball made contact with the ground or the ball carrier could not make any further progress.22,23 Tackles also included ineffective tackles (defending player made contact with the ball carrier and failed to prevent the attacking player from offloading the ball)17 and those occurring from support or decoy runs. Each tackle included the subsequent identification of a hit-up (ie, player with the ball) and tackler (ie, player contacting the ball carrier). These methods differ from those of Gabbett et al,17 in that all attacking collisions were collectively called hit-ups and not separated into subcategories. Match exposure was calculated as the time the player spent on the field of play, including time off allocated to injury or video referee decisions. Therefore, individual match playing times may exceed the 80 minutes of match play. Times off the field due
to injury or interchange periods were excluded from data analysis. Tackles, hit-ups, and total collisions were calculated per minute of match play to provide information about the collision frequency, as indicators of player work rate. Match video was coded (League Analyst, version 4.07.280) via notational analysis of the team’s attacking and defensive performances. Each identified tackle or hit-up was time-coded at the point of contact. An experienced coder and 1 author (C.C.) independently analyzed the attacking and defensive performances of 328 collision events. The interrater reliability (using Cohen kappa statistic) for the coding of collision events between the coder and author (C.C.) was κ = 0.909.24 Player movements and collisions were assessed using GPS devices (SPI-Pro X II, GPSports, Canberra, Australia) worn in a pocket inside the playing jersey, located on the upper back between the shoulder blades.8 The GPS receiver sampled at 15 Hz (interpolated data) and contained an integrated triaxial accelerometer (sampling at 100 Hz) that measured acceleration in gravitational force (g) over 3 planes (x, y, and z) of movement (anteroposterior, vertical, and mediolateral). The accelerometer (with a maximum of 8 g acceleration in each axis) was used to establish collision data (intensity and force distribution). Instantaneous player-collision forces (g) were calculated as the resultant acceleration (square root of the sum of the square of x, y, and z axes), whereby 12.8 g was the maximum accelerometry output. The player-collision count was recorded, with each collision categorized into 1 of 6 collision bands (Team AMS classification, GPSports), corresponding to low (zone 1, 12.0 g). Previous collision research using GPSports units2 used 6 classification zones that were derived from methods used in rugby union. As this approach may potentially fail to adequately reflect the demands of rugby league, the 6 zones in this study are based on slightly modified GPSports recommendations. Each video-time-coded collision event (tackle or hit-up) was allocated to 1 of the 6 classification zones based on the acceleration g-force characteristic recorded by the triaxial accelerometer. Collisions automatically recorded via the GPS devices were compared with those coded from video recordings of the tackle and hit-up events. Only events resulting in a collision as confirmed by video recordings were coded. Match data included 359 GPS files from outside backs (n = 78 files, n = 5 players), adjustables (n = 97 files, n = 5 players), wide-running forwards (n = 136 files, n = 9 players), and hit-up forwards (n = 48 files, n = 4 players).
Statistical Analysis All statistical analyses were performed using the Statistical Package for Social Sciences (SPSS, version 20; SPSS, Inc, Chicago, IL, USA). Data were inspected for normality before analysis. Differences in the physical demands among positional groups were compared using a 1-way analysis of variance. The source of any significant difference was examined with a Tukey post hoc test. The level of significance was accepted at P £ .05 and all data are reported as mean and 95% confidence intervals (CI), unless otherwise stated. To adopt a practical approach to the study, differences between playing positions were also analyzed using Cohen effect size (ES) statistic and 95% confidence interval (CI). Effect sizes were categorized as trivial (2.0).25
IJSPP Vol. 10, No. 6, 2015
734 Cummins and Orr
Results
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
Hit-up forwards sustained more collisions during match play than all other positions (Figure 1). Hit-up forwards experienced greater total collisions per minute of match play than the outside backs (ES = 6.00; 95% CI [2.54, 8.12]) and adjustables (ES = 1.82; 95% CI [–3.13, –0.11]), while the wide-running forwards experienced greater total collisions than outside backs (ES= 2.07; 95% CI [0.54, 3.32]). Hit-up forwards performed 0.5 tackles per minute of match play, 5 times that of outside backs (ES = 1.90; 95% CI [0.26, 3.16]) (Figure 2), and 0.2 hit-ups per minute of match play, twice as many as adjustables (Figure 3).
Outside backs recorded significantly greater match exposure than hit-up forwards (ES = –5.17; 95% CI [–7.09, –2.10]) (Table 1); however, hit-up forwards averaged a collision every 1.25 minutes of match play compared with 1 every 5 minutes for outside backs. Hit-up and wide-running forwards were involved in significantly more moderate- to high-collision tackles (≥8 g) than the outside backs and adjustables. Tackle performance differences were also apparent within the collision bands. Outside backs performed significantly fewer tackles per minute at the higher collision forces than all other positional groups. Outside backs recorded no tackles in the highest collision band (>12 g). Notably, hit-up forwards and adjustables performed more tackles per minute at moderate collision
Figure 1 — Collisions per game and collisions per minute of match play for rugby league positional groups during match play, mean ± SD. Collisions are composed of both hit-ups and tackles. Outside backs = centers and wingers; adjustables = hookers, halfbacks, five-eighths, and fullbacks; widerunning forwards = second rowers and locks; hit-up forwards = props. *Significantly different (P < .05) from hit-up forwards. †Significantly different (P < .05) from wide-running forwards. ‡Significantly different (P < .05) from adjustables.
Figure 2 — Tackles per game and tackles per minute of match play for 4 positional groups during rugby league match play, mean ± SD. *Significantly different (P < .05) from hit-up forwards. Outside backs = centers and wingers; adjustables = hookers, halfbacks, five-eighths, and fullbacks; widerunning forwards = second rowers and locks; hit-up forwards = props. †Significantly different (P < .05) from wide-running forwards. ‡Significantly different (P < .05) from adjustables. IJSPP Vol. 10, No. 6, 2015
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
Collisions in Elite Rugby League Matches 735
Figure 3 — Hit-ups per game and hit-ups per minute of match play for 4 positional groups during rugby league match play, mean ± SD. Collisions are composed of both hit-ups and tackles. Outside backs = centers and wingers; adjustables = hookers, halfbacks, five-eighths, and fullbacks; wide-running forwards = second rowers and locks; hit-up forwards = props. *Significantly different (P < .05) from hit-up forwards. †Significantly different (P < .05) from wide-running forwards. ‡Significantly different (P < .05) from adjustables.
forces in zone 4 (ES = 16.89; 95% CI [7.95,22.05], ES = 2.30; 95% CI [0.54,3.62], respectively) than outside backs. Hit-up forwards performed significantly greater total collisions per minute across zones 2 to 6 (P = .000–.031) than outside backs (Table 2) and greater than both adjustables and wide-running forwards in zone 5 (P = .007, P = .010, respectively). Hit-up forwards also sustained significantly greater high-intensity collisions per minute in zones 5 and 6 than the wide-running forwards (ES = 1.70; 95% CI [0.26, 2.91]), adjustables (ES = 2.00; 95% CI [0.23, 3.33]), and outside backs (ES = 4.43; 95% CI [1.69, 6.18]). Similarly, wide-running forwards experienced more high-intensity collisions per minute than outside backs (ES = 1.47; 95% CI [0.16, 2.57]). Hit-up forwards experienced a collision in the highest classification of force (zones 5 and 6) every 2.5 minutes of match play compared with 1 every 5 and 9 minutes by the adjustables and outside backs, respectively.
Discussion To our knowledge, this is the first study to quantify the collision frequency and forces associated with tackling and hit-up events between forward and back playing positions in professional rugby league match play using GPS and integrated triaxial accelerometer technology. The current findings demonstrate that throughout a match, players are exposed to a significant number of collision events, with positional differences between the forward and back playing positions. Hit-up forwards produced 33% to 300% more collisions per minute of match play than all other positional groups (wide-running forwards 33.3%, adjustables 60%, and outside backs 300%). This confirms previous literature by Gissane and White,18 in which forward positions experienced more total collisions per match (29) than back positions (20). By contrast, Gabbett et al1 reported that wide-running forwards experienced the greatest number of collisions per match. Compared with the current study, those authors report more collisions in all positional groups: outside backs (28 vs 18.3), adjustables (34 vs 21.7), hit-up forwards (42 vs 30.5), and wide-running forwards (45 vs 29.8).1 The observed inconsistency may be attributed to various factors; namely, different
GPS devices used between studies, collision events automatically detected by GPS units not being matched to video footage, and possible dissimilarity in the physical qualities or playing styles of the teams analyzed. Hit-up forwards produced a greater number of high-impact (zone 5 and 6) tackles and hit-ups per minute of match play than wide-running forwards (43%), adjustables (100%), and outside backs (263.6%). Compared with outside backs, wide-running forwards produced 200% more collisions per minute of match play and experienced a collision at a rate of every 1.75 minutes compared with a collision every 5 minutes, or 7 times more than outside backs. Wide-running forwards also experienced more collisions per minute of match play in zones 5 and 6 (154.5%) while producing more tackles per minute of match play (300%) than outside backs. These characteristics are similar to previous literature suggesting that forwards are predominantly involved in a greater number of tackles and physical collisions, while backs have a greater emphasis on running.3,26 Based on these data, small-sided games replicating the minimal space between attacking and defensive sides may be an efficient method of training the capacity to complete repeated tackle and hit-up collision events in forward positional groups. Despite the significance of collisions, a balance exists between the minimum number of training collisions needed to maintain skills and conditioning and the number of collisions in which the risk of injury, fatigue, and microtrauma increases.15 In agreement with previous findings,3,19 apart from outside backs, whose total collision count was composed of more collisions in attack than defense per match, all positional groups experienced more collision events in defense rather than attack. The distributional difference of collisions of the outside backs may be indicative of an individual team’s defensive play strategies, whereby an increased reliance on attacking play through wide-running plays, kick chase plays, and dummy half runs is seen. In rugby league match play, moderate- to high-impact collisions between players are associated with forces in in the 3 highest collision zones (>8 g). During attack no significant difference was found between the numbers of hit-ups per match in each collision zone between playing positions. However, during defensive play,
IJSPP Vol. 10, No. 6, 2015
736 Cummins and Orr
Table 1 Match Exposure, Collision Frequency, and Player Hit-Up and Tackle Frequency by Impact Zone, Mean (95% CI) Hit-up forwards (n = 4)
Wide-running forwards (n = 9)
Adjustables (n = 5)
Match exposure (min)
40.9 (20.5,61.3)
57.9 (40.0,75.8)
Collision frequency
1 every 1.25 min
zone 1
Outside backs (n = 5)
P < .05
57.2 (9.9,104.5)
84.8 (82.2,26.6)
.040*
1 every 1.75 min
1 every 2 min
1 every 5 min
0.2 (0.04,0.36)
0.1 (0.02,0.18)
0.1 (0.09,0.11)
0.5 (0.49,1.50)
zone 2
0.2 (0.04,0.36)
0.1 (0.05,0.25)
0.1 (0.09,0.11)
0.2 (0.30,0.70)
zone 3
0.2 (0.04,0.36)
0.2 (0.05,0.35)
0.2 (0.08,0.32)
0.5 (0.00,1.0)
zone 4
1.5 (0.55,2.45)
1.9 (1.2,2.6)
1.5 (0.11,3.11)
2.8 (1.70,3.92)
zone 5
4.6 (2.05,7.15)
4.6 (3.37,5.83)
2.6 (0.38,5.58)
5.0 (3.14,6.86)
zone 6
2.3 (0.73,5.32)
1.6 (1.06,2.14)
0.6 (0.15,1.34)
2.2 (0.95,3.44)
zone 1
2.4 (0.33,4.47)
1.8 (0.26,2.34)
1.1 (0.27,2.47)
1.2 (0.17,2.57)
zone 2
1.7 (1.54,8.56)
1.4 (0.86,1.93)
1.13 (0.07,2.67)
0.5 (0.12,1.12)
zone 3
2.6 (1.01,4.19)
2.4 (1.71,3.09)
1.9 (0.4,3.39)
1.1 (0.27,2.47)
zone 4
7.2 (2.27,12.13)
7.2 (5.28,9.12)
7.3 (1.59,13.01)
2.5 (0.85,5.85)†
.026
zone 5
6.4 (2.49,10.45)
5.6 (4.6,6.60)
4.5 (1.21,10.21)
1.5 (0.88,2.12)*†
.044* .049†
zone 6
1.3 (0.35,2.25)
1.7 (0.85,2.55)
1.2 (0.46,1.94)
0.2 (0.05,0.45)†
.011
zone 1
0.0 (0.0,0.0)
0.0 (0.0,0.0)
0.0 (0.0,0.0)
0.01 (0.01,0.01)
zone 2
0.01 (0.01,0.02)
0.0 (0.0,0.0)
0.0 (0.0,0.0)
0.0 (0.0,0.0)
zone 3
0.0 (0.0,0.0)
0.01 (0.01,0.01)
0.0 (0.0,0.0)
0.01 (0.01,0.01)
zone 4
0.04 (0.04,0.04)
0.04 (0.04,0.04)
0.02 (0.02,0.02)
0.03 (0.012,0.04)
zone 5
0.1 (0.01,0.01)
0.1 (0.01,0.01)
0.05 (0.05,0.05)*
.002
5.4 (3.56,7.3)*
zone 6
0.1 (0.01,0.01)
0.04 (0.04,0.04)
0.01 (0.01,0.01)*
.003
0.03 (0.02,0.04)
zone 1
0.1 (0.01,0.03)
0.04 (0.04,0.04)
0.0 (0.0,0.0)
0.01 (0.01,0.01)
zone 2
0.0 (0.0,0.0)
0.02 (0.02,0.02)
0.02 (0.02,0.02)
0.0 (0.0,0.0)*
.001
zone 3
0.1 (0.1,0.1)
0.1 (0.01,0.01)
0.05 (0.05,0.05)
0.01 (0.010,0.02)*
.037
zone 4
1.2 (1.18,1.22)
0.1 (0.01,0.01)
0.2 (0.08,0.32)
0.03 (0.01,0.07)*‡
.021* .015‡
zone 5
0.2 (0.02,0.02)
0.1 (0.02,0.18)
0.1 (0.02,0.22)
0.02 (0.02,0.02)*†‡
.001* .024† .020‡
zone 6
0.03 (0.03,0.03)
0.03 (0.03,0.03)
0.04 (0.04,0.04)
0.0 (0.0,0.0)†‡
.041† .034‡
P < .05
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
Hit-ups/match
Tackles/match
Hit-ups/min
.007
Tackles/min
Note: Zone 1, 12 g. Total collisions comprise both hit-ups and tackles. *Significantly different (P < .05) from hit-up forwards. †Significantly different (P < .05) from wide-running forwards. ‡Significantly different (P < .05) from adjustables.
the outside backs produced significantly fewer tackles per match than wide-running forwards in zones 5 (21.4% vs 27.9%) and 6 (2.9% vs 8.5%) and the hit-up forwards in zone 5 (21.4% vs 29.6%). By contrast, McLellan et al2 observed no significant difference between the number of entries into each collision zone between
forwards and backs during both offensive and defensive match play. Direct comparisons of the number of collisions in each zone were not possible due to the different collision criteria adopted. McLellan et al2 used total collisions provided by manufacturer software, which included all forces (running, directional changes,
IJSPP Vol. 10, No. 6, 2015
Collisions in Elite Rugby League Matches 737
Table 2 Total Collision (Tackles and Hit-Ups) Frequency by Impact Zone, Mean (95% CI) Hit-up forwards (n = 4)
Wide-running forwards (n = 9)
zone 1
2.6 (0.53,4.67)
1.8 (1.34,2.26)
1.3 (0.9,2.79)
1.7 (0.09,3.31)
zone 2
1.9 (1.58,2.22)
1.6 (1.06,2.14)
1.3 (0.9,2.79)
0.7 (0.17,1.57)
zone 3
2.6 (0.70,4.51)
2.7 (1.85,3.55)
2.1 (0.61,3.59)
1.6 (0.14,3.34)
zone 4
8.7 (3.29,14.11)
9.13 (6.67,11.59)
8.1 (2.51,13.69)
5.4 (1.80,9.00)
zone 5
11.0 (4.65,17.36)
10.33 (8.71,11.94)
7.2 (1.6,12.79)
6.5 (4.27,8.73)
zone 6
3.5 (0.48,7.48)
3.33 (8.72,11.94)
1.8 (1.18,2.42)
2.4 (1.28,3.51)
zone 1
0.1 (0.06,0.26)
0.04 (0.04,0.04)
0.02 (0.02,0.02)
0.02 (0.02,0.02)
zone 2
0.1 (0.01,0.01)
0.03 (0.03,0.03)
0.02 (0.02,0.02)
0.01 (0.01,0.01)*
.002
zone 3
0.1 (0.01,0.01)
0.05 (0.05,0.05)
0.05 (0.05,0.05)
0.02 (0.02,0.02)*
.031
zone 4
0.2 (0.02,0.02)
0.17 (0.017,0.17)
0.19 (0.06,0.31)
0.06 (0.06,0.06)*†‡
.010* .038† .023‡
zone 5
0.3 (0.03,0.03)
0.21 (0.13,0.29)*
0.08 (0.08,0.08)*†
.000* .009†
zone 6
0.1 (0.1,0.1)
0.07 (0.07,0.07)
0.04 (0.04,0.04)
0.03 (0.03,0.03)*
.010
Total high collisions (zones 5 and 6)
14.5 (4.48,24.52)
13.9 (11.13,16.67)
9.0 (2.79,15.2)
8.9 (7.53,10.27)
Total collisions/min (zones 5 and 6)
0.4 (0.24,0.56)
0.28 (0.13,0.29)*
High collision frequency (zones 5 and 6)
1 every 2.5 min
1 every 4.3 min
P < .05
Adjustables (n = 5)
P < .05
Outside backs (n = 5)
P < .05
Collisions/match
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
Collisions/min
.010
.018
0.16 (0.04,0.28)*
0.2 (0.08,0.32)*
1 every 5 min
.007
.005
0.11 (0.11,0.11)*†
.000* .010†
1 every 9 min
Note: Zone 1, 12 g. *Significantly different (P < .05) from hit-up forwards. †Significantly different (P < .05) from wide-running forwards. ‡Significantly different (P < .05) from adjustables.
and player-to-player collisions) experienced by a player during match play. The heavily weighted vertical forces experienced during running and walking, as well as impacts from directional changes, were not differentiated from the physical collisions of hit-ups and tackles and may therefore explain the similar collision profiles seen between forward and back playing positions. Even when directional changes and minor collisions (zones 1 and 2)2 were removed from the data, an average of 464 impacts per player were recorded.15 This is substantially greater than the collision number previously reported (600 per match, 300 per team),17 suggesting caution on inspecting this data.15 The rate of collisions (number of collisions per minute of match play) differed between playing positions for both offensive and defensive play in each collision zone. Adjustables were exposed to significantly fewer high-collision (zones 5 and 6) hit-ups per minute of match play than hit-up forwards. This difference may be attributed to the adjustables directing play and being the intermediary between attack and defense. As expected, outside backs performed fewer tackles per match than all other positions (67% less than hit-up forwards). The most notable differences were observed between backs and hit-up forwards across zones 2 to 5, with hit-up forwards executing 10 times the number of tackles per minute at moderate and high collision bands (zones 3 and 5) than outside backs. The
hit-up forwards experienced almost 1 collision per minute, with 1 collision every 2.5 minutes sustained in the highest recorded collision forces of zones 5 and 6. In addition, in zones 5 and 6 they performed 1 tackle every 2 minutes throughout the match compared with 1 tackle every 9 minutes for outside backs. The repeated physical collisions and higher-frequency collisions by hit-up and wide-running forwards underscores the need to tailor training toward contact and collision conditioning in these playing groups. Physical conditioning programs based on the rate of collisions as opposed to mean collision values will likely result in players being better prepared for intense periods of match play and may reduce injury incidence. The risk of muscle damage is increased in this group, where moderate collisions have been shown to cause significant skeletal-muscle damage,2 and exposure to repetitive subconcussive collisions may increase the risk of traumatic brain injury.27 To date, no comparative data quantifying attacking- and defensive-play collisions by zone categories are available. However, a similar study1 broadly defined collisions as mild, moderate, and heavy without assigning g-forces to each category. Comparatively fewer collisions in every zone were observed in the current study. Although relative collisions (per minute of match play) for mild and heavy collisions across positions were similar between studies, fewer relative collisions across all positions for moderate collisions
IJSPP Vol. 10, No. 6, 2015
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
738 Cummins and Orr
(zones 3 and 4)1 were observed in this study: hit-up forwards (0.3 vs 0.58), wide-running forwards (0.22 vs 0.39), adjustables (0.24 vs 0.33), and outside backs (0.08 vs 0.17). The disparity in findings likely reflects inherent differences between the playing styles or physical caliber of the individual teams or the inclusion of incidental collision events. Inconsistency of collision-zone classifications in the literature1–3,16 hinders direct comparisons with the current study. The 6 zone classifications used by McLellan et at2 were modeled from collision classifications designed for rugby union.5 Due to the inherent differences in the 2 rugby codes, these zones may be inadequate in reflecting the collision demands of rugby league. A recent study reported on the collisions experienced in attacking and defensive play by forward playing positions but attributed g-force classification zones (mild, 1–2 g; moderate, 2.1–4 g; and heavy, >4 g) without providing a rationale.3 Clarity and consistency of collision zones and definitions in rugby league would facilitate precise comparison and analysis of performance between individual players, teams, levels of competition, and seasons.16 In rugby league, collisions and tackles are commonly regarded as the most demanding aspect of match play,28 making measurement of collisions important from both an injury-prevention and a physical-conditioning perspective. An understanding of the frequency and forces associated with collision events provides the opportunity to develop position-specific collision and tackle conditioning programs for elite rugby league players. This study is not without limitations. In using player GPS files across a season, there is a pseudo replication in reporting multiple measures from individual players in the same positional group. In addition, the GPS devices used in this study record accelerations up to 8 g in each axis, giving a peak resultant acceleration of up to 12.8 g. This capability may be insufficient in differentiating moderate- and high-collision events due to the attenuation of the accelerometer signal. Facilitated by the inclusion of a gyroscope, the Catapult microtechnology units are the only devices that have validated automatic collision detection and quantified the contact load in collision sports. Future research investigating the validity of automatic collision detection in GPSports devices is warranted.
Conclusion In conclusion, this is the first study, to our knowledge, to quantify the intensity of collisions in attacking and defensive play between forward and back playing positions in elite rugby league match play. The findings demonstrate that during the course of a match, players are exposed to a significant number of collision events. Hit-up and wide-running forwards experience greater collision events than adjustables and outside backs, highlighting the presence of positional-play differences. Although these results may be unique to the individual team’s defensive and attacking play strategies, they are indicative of the significant collision profiles in professional rugby league.
Practical Implications • Wide-running and hit-up forwards experienced greater collisions per match in defensive rather than attacking play. Forward players should undertake specific collision training to tolerate the heavy contact demands associated with defensive play. • Attacking and defensive play in rugby league are executed with considerable collision forces and add to the cumulative physical load experienced by players during match play.
• Positional-play collision-profile differences highlight the need for position-specific collision training to prepare players for the physical demands of competition. Acknowledgments This manuscript contributes to C.C.’s PhD qualification. The authors wish to thank the players and staff of the participating rugby league club for their support of this project. No funding has been received for the preparation of this manuscript. The authors declare that there are no conflicts of interest that are directly relevant to the context of this review.
References 1. Gabbett TJ, Jenkins D, Abernethy B. Physical demands of professional rugby league training and competition using microtechnology. J Sci Med Sport. 2012;15(1):80–86. PubMed doi:10.1016/j. jsams.2011.07.004 2. McLellan CP, Lovell D, Gass G. Biochemical and endocrine responses to impact and collision during elite rugby league match play. J Strength Cond Res. 2011;25(6):1553–1562. PubMed doi:10.1519/ JSC.0b013e3181db9bdd 3. Gabbett TJ, Polley C, Dwyer D, Kearney S, Corvo A. Influence of field position and phase of play on the physical demands of match-play in professional rugby league forwards. J Sci Med Sport. 2014;17(5):556– 561. PubMed doi:10.1016/j.jsams.2013.08.002 4. Kempton T, Sirotic A, Coutts A. An integrated analysis of match-related fatigue in professional rugby league. J Sports Sci. 2015;33(1):39–47. PubMed doi:10.1080/02640414.2014.921832 5. Cunniffe B, Proctor W, Baker J, Davies B. An evaluation of the physiological demands of elite rugby union using global positioning system tracking software. J Strength Cond Res. 2009;23(4):1195–1203. PubMed doi:10.1519/JSC.0b013e3181a3928b 6. Suarez-Arrones L, Portillo L, Gonzalez-Rave J, Munoz V, Sanchez F. Match running performance in Spanish elite male rugby union using global positioning system. Isokinet Exerc Sci. 2012;20(2):77–83. 7. Venter R, Opperman E, Opperman S. The use of global positioning system (GPS) tracking devices to assess movement demands and impacts in under-19 rugby union match play. Afr J Phys Health Educ Recr Dance. 2011;17:1–8. 8. Gastin PB, McLean O, Spittle M, Breed R. Quantification of tackling demands in professional Australian football using integrated wearable athlete tracking technology. J Sci Med Sport. 2013;16(6):589–593. PubMed doi:10.1016/j.jsams.2013.01.007 9. Varley MC, Fairweather I, Aughey R. Validity and reliability of GPS for measuring instantaneous velocity during acceleration, deceleration, and constant motion. J Sports Sci. 2012;30(2):121–127. PubMed doi :10.1080/02640414.2011.627941 10. Wundersitz DW, Netto KJ, Aisbett B, Gastin PB. Validity of an upperbody-mounted accelerometer to measure peak vertical and resultant force during running and change-of-direction tasks. Sports Biomech. 2013;12(4):403–412. PubMed 11. Gabbett T, Jenkins D, Abernethy B. Physical collisions and injury during professional rugby league skills training. J Sci Med Sport. 2010;13(6):578–583. PubMed doi:10.1016/j.jsams.2010.03.007 12. Gastin PB, Mclean O, Breed R, Spittle M. Tackle and impact detection in elite Australian football using wearable microsensor technology. J Sports Sci. 2014;32(10):947–953. PubMed doi:10.1080/02640414.2 013.868920 13. Kelly D, Coughlan G, Green B, Caulfield B. Automatic detection of collisions in elite level rugby union using a wearable sensing device. Sports Eng. 2012;15(2):81–92. doi:10.1007/s12283-012-0088-5
IJSPP Vol. 10, No. 6, 2015
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
Collisions in Elite Rugby League Matches 739
14. Kelly SJ, Murphy A, Watsford M, Austin D, Rennie M. Reliability and validity of sports accelerometers during static and dynamic testing. Int J Sports Physiol Perform. 2015;10(1):106–111. PubMed doi:10.1123/ ijspp.2013-0408 15. Gabbett TJ. Quantifying the physical demands of collision sports: does microsensor technology measure what it claims to measure? J Strength Cond Res. 2013;27(8):2319–2322. PubMed doi:10.1519/ JSC.0b013e318277fd21 16. Cummins C, Orr R, O’Connor H, West C. Global positioning systems (GPS) and microtechnology sensors in team sports: a systematic review. Sports Med. 2013;43(10):1025–1042. PubMed doi:10.1007/ s40279-013-0069-2 17. Gabbett TJ, Jenkins D, Abernethy B. Physical collisions and injury in professional rugby league match-play. J Sci Med Sport. 2011;14(3):210–215. PubMed doi:10.1016/j.jsams.2011.01.002 18. Gissane C, White J. Physical collisions in professional Super League rugby, the demands on different player positions. Cleve Med J. 2001 4:137–146. 19. Dave S, Craig T, Shayne H, Ceri N, Kevin L. Semi-automated timemotion analysis of senior elite rugby league. Int J Perform Anal Sport. 2009;9(1):47–59. 20. Gabbett TJ. Effects of physical, technical, and tactical factors on final ladder position in semiprofessional rugby league. Int J Sports Physiol Perform. 2014;9(4):680–688. PubMed doi:10.1123/IJSPP.2013-0253 21. Kempton T, Sirotic A, Rampinini E, Coutts A. Metabolic power demands of rugby league match play. Int J Sports Physiol Perform. 2015;10(1):23–28. PubMed doi:10.1123/ijspp.2013-0540
22. National Rugby League. Rugby League Laws of the Game International Level With Notes on the Laws and NRL Telstra Premiership Interpretations. National Rugby League; 2013. 23. King D, Hume P, Clark T. Nature of tackles that result in injury in professional rugby league. Res Sports Med. 2012;20(2):86–104. PubMed 24. Viera AJ, Garrett J. Understanding interobserver agreement: the kappa statistic. Fam Med. 2005;37(5):360–363. PubMed 25. Hopkins WG, Marshall S, Batterham A, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc. 2009;41(1):3–13. PubMed doi:10.1249/ MSS.0b013e31818cb278 26. McLellan CP, Lovell D, Gass G. Performance analysis of elite rugby league match play using global positioning systems. J Strength Cond Res. 2011;25(6):1703–1710. PubMed doi:10.1519/ JSC.0b013e3181ddf678 27. Breedlove EL, Robinson M, Talavage T, et al. Biomechanical correlates of symptomatic and asymptomatic neurophysiological impairment in high school football. J Biomech. 2012;45(7):1265–1272. PubMed doi:10.1016/j.jbiomech.2012.01.034 28. Brewer J, Davis J. Applied physiology of rugby league. Sports Med. 1995;20(3):129–135. PubMed doi:10.2165/00007256-19952003000001
IJSPP Vol. 10, No. 6, 2015
Original Investigation
Analysis of Physical Collisions in Elite National Rugby League Match Play Cloe Cummins and Rhonda Orr Objective: To investigate the impact forces of collision events during both attack and defense in elite rugby league match play and to compare the collision profiles between playing positions. Participants: 26 elite rugby league players. Methods: Player collisions were recorded using an integrated accelerometer in global positioning system units (SPI-Pro X, GPSports). Impact forces of collisions in attack (hit-ups) and defense (tackles) were analyzed from 359 files from outside backs (n = 78), adjustables (n = 97), wide-running forwards (n = 136), and hit-up forwards (n = 48) over 1 National Rugby League season. Results: Hit-up forwards were involved in 0.8 collisions/min, significantly more than all other positional groups (wide-running forwards P = .050, adjustables P = .042, and outside backs P = .000). Outside backs experienced 25% fewer collisions per minute than hit-up forwards. Hit-up forwards experienced a collision within the 2 highest classifications of force (≥10 g) every 2.5 min of match play compared with 1 every 5 and 9 min for adjustables and outside backs, respectively. Hit-up forwards performed 0.5 tackles per minute of match play, 5 times that of outside backs (ES = 1.90; 95% CI [0.26,3.16]), and 0.2 hit-ups per minute of match play, twice as many as adjustables. Conclusions: During a rugby league match, players are exposed to a significant number of collision events. Positional differences exist, with hit-up and wide-running forwards experiencing greater collision events than adjustables and outside backs. Although these results may be unique to the individual team’s defensive- and attacking-play strategies, they are indicative of the significant collision profiles in professional rugby league. Keywords: global positioning system, GPS, impact force, hit-up, tackle Portable global positioning system (GPS) devices coupled with integrated microsensor technology (accelerometer, magnetometer with or without gyroscope sensors) have become a commonly used measurement tool in team-sport settings to describe player activity profiles. Output measures from these devices have evolved from the examination of speed, distance covered, and high-intensity efforts (movement demands) to the quantification of collisions and body loads (physical loads). To date, collision forces associated with the physical demands of collision sports have been reported in rugby league,1–4 rugby union5–7 and Australian football.8 An understanding of player collisions and the physical demands in team contact sports is important for monitoring training, performance, and injury prevention to effectively manage and prepare players for the rigors of competitive match play. The validity and reliability of accelerometer data from GPS and integrated accelerometry technology in the sporting environment have recently been collected for instantaneous velocity during accelerations and decelerations9 and foot strikes.10 One commercially available GPS unit (Catapult MinimaxX) enables automated tackle and collision detection using algorithms that incorporate changes in unit orientation and spikes in accelerometer data. Although validated for quantifying the number and intensity of collisions in rugby league,11 these units have not been effective in detecting tackles in Australian football.12 The validity of the automatic collision detection of another GPS unit (GPSports) to measure the contact loads and physical collisions in rugby codes The authors are with the Faculty of Health Sciences, University of Sydney, Lidcombe, NSW, Australia. Address author correspondence to Cloe Cummins at [email protected].
732
has not been reported. In rugby union, however, the automatic detection of physical contacts has been explored through the use of static window features and a mathematical learning grid that created algorithms through classifying the accelerometry signals of tackle and nontackle events.13 That study demonstrated that the GPSports accelerometer is capable of detecting collisions in rugby union. In addition, GPSports units (SPI-ProX II) have recently been shown to demonstrate high intra-accelerometer and interaccelerometer reliability.14 Only small to moderate (ES < 0.5) differences were reported between accelerometers, with no significant differences seen within the interaccelerometer reliability. This high reliability indicates that these units are sufficient in providing consistent acceleration measures. SPI-ProX II accelerometers also consistently measured at a lower magnitude of acceleration than a criterion reference accelerometer, suggesting the potential for errors in the magnitude of accelerometer measures. The authors of that study posit that this difference may be confounded by the accelerometer hardware or data-processing procedures.14 Rugby league players require the ability to perform and withstand physical collisions during both defensive and attacking phases of play.11 The defensive play of tackling is integral in rugby league, where the ability to effectively perform a winning tackle may prove crucial in determining match outcome.15 The attacking play of hitups and the ability to tolerate physical collisions is important for gaining distance and field position. Not only is the total number of collisions sustained in a match important, but also the collision type and magnitude are vital considerations for match outcome. The use of GPS and integrated triaxial accelerometers to quantify forces associated with collision events is an emerging area of research.16 In rugby league, the total number and intensity
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
Collisions in Elite Rugby League Matches 733
of collisions experienced by players during high-velocity collisions in match play1–4 and training1 have been noted. The difference in collisions experienced during attacking (hit-up) and defensive (tackle) match play within a team as a whole4 and between forward playing positions only3 have also been reported. Forwards sustained an average of 55 collisions (39 tackles, 16 hit-ups) compared with 29 collisions (16 tackles, 13 hit-ups) experienced by backs.17,18 Furthermore, hit-up and wide-running forwards recorded a higher number of collisions than both adjustables and outside backs during the course of a match.1,17,19 However, little is known about the collision magnitude and collision type for both back and forward players in elite rugby league. A greater understanding of collisions experienced by players and a quantification of the forces associated with these events will enable position-specific collision profiles to be determined, assisting to inform the most appropriate collisiontraining regimens. To our knowledge, no study has explored the differential collision forces between the 4 positional groups in a team. Furthermore, the total number of collisions, collision type, and collision magnitude are yet to be fully elucidated for different playing groups. Therefore, the objective of this study was twofold: to investigate the forces of collision events during both attacking and defensive play in elite rugby league match play and to compare the collision profiles between forward and back playing positions using GPS technology.
Methods Subjects Twenty-six elite male rugby league players, (mean ± SD age 24.2 ± 2.8 y, height 184.7 ± 6.5 cm, and body mass 99.0 ± 9.5 kg) representing 1 National Rugby League (NRL) team participated in this study. The study was approved by the University of Sydney Human Research Ethics Committee (Protocol number 14020), in accordance with the Declaration of Helsinki. All participants gave written informed consent before participating.
Methodology Players were categorized into 1 of 4 positional groups representing the outside backs (centers and wingers), adjustables (hookers, halfbacks, five-eighths, and fullbacks), wide-running forwards (second rowers and locks), and hit-up forwards (props).1,17,20,21 Publicly available television footage from each NRL match was used to determine attacking and defensive match-play events. A tackle was defined as occurring when the ball carrier was held by 1 or more opposing players and either the ball or hand of the arm holding the ball made contact with the ground or the ball carrier could not make any further progress.22,23 Tackles also included ineffective tackles (defending player made contact with the ball carrier and failed to prevent the attacking player from offloading the ball)17 and those occurring from support or decoy runs. Each tackle included the subsequent identification of a hit-up (ie, player with the ball) and tackler (ie, player contacting the ball carrier). These methods differ from those of Gabbett et al,17 in that all attacking collisions were collectively called hit-ups and not separated into subcategories. Match exposure was calculated as the time the player spent on the field of play, including time off allocated to injury or video referee decisions. Therefore, individual match playing times may exceed the 80 minutes of match play. Times off the field due
to injury or interchange periods were excluded from data analysis. Tackles, hit-ups, and total collisions were calculated per minute of match play to provide information about the collision frequency, as indicators of player work rate. Match video was coded (League Analyst, version 4.07.280) via notational analysis of the team’s attacking and defensive performances. Each identified tackle or hit-up was time-coded at the point of contact. An experienced coder and 1 author (C.C.) independently analyzed the attacking and defensive performances of 328 collision events. The interrater reliability (using Cohen kappa statistic) for the coding of collision events between the coder and author (C.C.) was κ = 0.909.24 Player movements and collisions were assessed using GPS devices (SPI-Pro X II, GPSports, Canberra, Australia) worn in a pocket inside the playing jersey, located on the upper back between the shoulder blades.8 The GPS receiver sampled at 15 Hz (interpolated data) and contained an integrated triaxial accelerometer (sampling at 100 Hz) that measured acceleration in gravitational force (g) over 3 planes (x, y, and z) of movement (anteroposterior, vertical, and mediolateral). The accelerometer (with a maximum of 8 g acceleration in each axis) was used to establish collision data (intensity and force distribution). Instantaneous player-collision forces (g) were calculated as the resultant acceleration (square root of the sum of the square of x, y, and z axes), whereby 12.8 g was the maximum accelerometry output. The player-collision count was recorded, with each collision categorized into 1 of 6 collision bands (Team AMS classification, GPSports), corresponding to low (zone 1, 12.0 g). Previous collision research using GPSports units2 used 6 classification zones that were derived from methods used in rugby union. As this approach may potentially fail to adequately reflect the demands of rugby league, the 6 zones in this study are based on slightly modified GPSports recommendations. Each video-time-coded collision event (tackle or hit-up) was allocated to 1 of the 6 classification zones based on the acceleration g-force characteristic recorded by the triaxial accelerometer. Collisions automatically recorded via the GPS devices were compared with those coded from video recordings of the tackle and hit-up events. Only events resulting in a collision as confirmed by video recordings were coded. Match data included 359 GPS files from outside backs (n = 78 files, n = 5 players), adjustables (n = 97 files, n = 5 players), wide-running forwards (n = 136 files, n = 9 players), and hit-up forwards (n = 48 files, n = 4 players).
Statistical Analysis All statistical analyses were performed using the Statistical Package for Social Sciences (SPSS, version 20; SPSS, Inc, Chicago, IL, USA). Data were inspected for normality before analysis. Differences in the physical demands among positional groups were compared using a 1-way analysis of variance. The source of any significant difference was examined with a Tukey post hoc test. The level of significance was accepted at P £ .05 and all data are reported as mean and 95% confidence intervals (CI), unless otherwise stated. To adopt a practical approach to the study, differences between playing positions were also analyzed using Cohen effect size (ES) statistic and 95% confidence interval (CI). Effect sizes were categorized as trivial (2.0).25
IJSPP Vol. 10, No. 6, 2015
734 Cummins and Orr
Results
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
Hit-up forwards sustained more collisions during match play than all other positions (Figure 1). Hit-up forwards experienced greater total collisions per minute of match play than the outside backs (ES = 6.00; 95% CI [2.54, 8.12]) and adjustables (ES = 1.82; 95% CI [–3.13, –0.11]), while the wide-running forwards experienced greater total collisions than outside backs (ES= 2.07; 95% CI [0.54, 3.32]). Hit-up forwards performed 0.5 tackles per minute of match play, 5 times that of outside backs (ES = 1.90; 95% CI [0.26, 3.16]) (Figure 2), and 0.2 hit-ups per minute of match play, twice as many as adjustables (Figure 3).
Outside backs recorded significantly greater match exposure than hit-up forwards (ES = –5.17; 95% CI [–7.09, –2.10]) (Table 1); however, hit-up forwards averaged a collision every 1.25 minutes of match play compared with 1 every 5 minutes for outside backs. Hit-up and wide-running forwards were involved in significantly more moderate- to high-collision tackles (≥8 g) than the outside backs and adjustables. Tackle performance differences were also apparent within the collision bands. Outside backs performed significantly fewer tackles per minute at the higher collision forces than all other positional groups. Outside backs recorded no tackles in the highest collision band (>12 g). Notably, hit-up forwards and adjustables performed more tackles per minute at moderate collision
Figure 1 — Collisions per game and collisions per minute of match play for rugby league positional groups during match play, mean ± SD. Collisions are composed of both hit-ups and tackles. Outside backs = centers and wingers; adjustables = hookers, halfbacks, five-eighths, and fullbacks; widerunning forwards = second rowers and locks; hit-up forwards = props. *Significantly different (P < .05) from hit-up forwards. †Significantly different (P < .05) from wide-running forwards. ‡Significantly different (P < .05) from adjustables.
Figure 2 — Tackles per game and tackles per minute of match play for 4 positional groups during rugby league match play, mean ± SD. *Significantly different (P < .05) from hit-up forwards. Outside backs = centers and wingers; adjustables = hookers, halfbacks, five-eighths, and fullbacks; widerunning forwards = second rowers and locks; hit-up forwards = props. †Significantly different (P < .05) from wide-running forwards. ‡Significantly different (P < .05) from adjustables. IJSPP Vol. 10, No. 6, 2015
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
Collisions in Elite Rugby League Matches 735
Figure 3 — Hit-ups per game and hit-ups per minute of match play for 4 positional groups during rugby league match play, mean ± SD. Collisions are composed of both hit-ups and tackles. Outside backs = centers and wingers; adjustables = hookers, halfbacks, five-eighths, and fullbacks; wide-running forwards = second rowers and locks; hit-up forwards = props. *Significantly different (P < .05) from hit-up forwards. †Significantly different (P < .05) from wide-running forwards. ‡Significantly different (P < .05) from adjustables.
forces in zone 4 (ES = 16.89; 95% CI [7.95,22.05], ES = 2.30; 95% CI [0.54,3.62], respectively) than outside backs. Hit-up forwards performed significantly greater total collisions per minute across zones 2 to 6 (P = .000–.031) than outside backs (Table 2) and greater than both adjustables and wide-running forwards in zone 5 (P = .007, P = .010, respectively). Hit-up forwards also sustained significantly greater high-intensity collisions per minute in zones 5 and 6 than the wide-running forwards (ES = 1.70; 95% CI [0.26, 2.91]), adjustables (ES = 2.00; 95% CI [0.23, 3.33]), and outside backs (ES = 4.43; 95% CI [1.69, 6.18]). Similarly, wide-running forwards experienced more high-intensity collisions per minute than outside backs (ES = 1.47; 95% CI [0.16, 2.57]). Hit-up forwards experienced a collision in the highest classification of force (zones 5 and 6) every 2.5 minutes of match play compared with 1 every 5 and 9 minutes by the adjustables and outside backs, respectively.
Discussion To our knowledge, this is the first study to quantify the collision frequency and forces associated with tackling and hit-up events between forward and back playing positions in professional rugby league match play using GPS and integrated triaxial accelerometer technology. The current findings demonstrate that throughout a match, players are exposed to a significant number of collision events, with positional differences between the forward and back playing positions. Hit-up forwards produced 33% to 300% more collisions per minute of match play than all other positional groups (wide-running forwards 33.3%, adjustables 60%, and outside backs 300%). This confirms previous literature by Gissane and White,18 in which forward positions experienced more total collisions per match (29) than back positions (20). By contrast, Gabbett et al1 reported that wide-running forwards experienced the greatest number of collisions per match. Compared with the current study, those authors report more collisions in all positional groups: outside backs (28 vs 18.3), adjustables (34 vs 21.7), hit-up forwards (42 vs 30.5), and wide-running forwards (45 vs 29.8).1 The observed inconsistency may be attributed to various factors; namely, different
GPS devices used between studies, collision events automatically detected by GPS units not being matched to video footage, and possible dissimilarity in the physical qualities or playing styles of the teams analyzed. Hit-up forwards produced a greater number of high-impact (zone 5 and 6) tackles and hit-ups per minute of match play than wide-running forwards (43%), adjustables (100%), and outside backs (263.6%). Compared with outside backs, wide-running forwards produced 200% more collisions per minute of match play and experienced a collision at a rate of every 1.75 minutes compared with a collision every 5 minutes, or 7 times more than outside backs. Wide-running forwards also experienced more collisions per minute of match play in zones 5 and 6 (154.5%) while producing more tackles per minute of match play (300%) than outside backs. These characteristics are similar to previous literature suggesting that forwards are predominantly involved in a greater number of tackles and physical collisions, while backs have a greater emphasis on running.3,26 Based on these data, small-sided games replicating the minimal space between attacking and defensive sides may be an efficient method of training the capacity to complete repeated tackle and hit-up collision events in forward positional groups. Despite the significance of collisions, a balance exists between the minimum number of training collisions needed to maintain skills and conditioning and the number of collisions in which the risk of injury, fatigue, and microtrauma increases.15 In agreement with previous findings,3,19 apart from outside backs, whose total collision count was composed of more collisions in attack than defense per match, all positional groups experienced more collision events in defense rather than attack. The distributional difference of collisions of the outside backs may be indicative of an individual team’s defensive play strategies, whereby an increased reliance on attacking play through wide-running plays, kick chase plays, and dummy half runs is seen. In rugby league match play, moderate- to high-impact collisions between players are associated with forces in in the 3 highest collision zones (>8 g). During attack no significant difference was found between the numbers of hit-ups per match in each collision zone between playing positions. However, during defensive play,
IJSPP Vol. 10, No. 6, 2015
736 Cummins and Orr
Table 1 Match Exposure, Collision Frequency, and Player Hit-Up and Tackle Frequency by Impact Zone, Mean (95% CI) Hit-up forwards (n = 4)
Wide-running forwards (n = 9)
Adjustables (n = 5)
Match exposure (min)
40.9 (20.5,61.3)
57.9 (40.0,75.8)
Collision frequency
1 every 1.25 min
zone 1
Outside backs (n = 5)
P < .05
57.2 (9.9,104.5)
84.8 (82.2,26.6)
.040*
1 every 1.75 min
1 every 2 min
1 every 5 min
0.2 (0.04,0.36)
0.1 (0.02,0.18)
0.1 (0.09,0.11)
0.5 (0.49,1.50)
zone 2
0.2 (0.04,0.36)
0.1 (0.05,0.25)
0.1 (0.09,0.11)
0.2 (0.30,0.70)
zone 3
0.2 (0.04,0.36)
0.2 (0.05,0.35)
0.2 (0.08,0.32)
0.5 (0.00,1.0)
zone 4
1.5 (0.55,2.45)
1.9 (1.2,2.6)
1.5 (0.11,3.11)
2.8 (1.70,3.92)
zone 5
4.6 (2.05,7.15)
4.6 (3.37,5.83)
2.6 (0.38,5.58)
5.0 (3.14,6.86)
zone 6
2.3 (0.73,5.32)
1.6 (1.06,2.14)
0.6 (0.15,1.34)
2.2 (0.95,3.44)
zone 1
2.4 (0.33,4.47)
1.8 (0.26,2.34)
1.1 (0.27,2.47)
1.2 (0.17,2.57)
zone 2
1.7 (1.54,8.56)
1.4 (0.86,1.93)
1.13 (0.07,2.67)
0.5 (0.12,1.12)
zone 3
2.6 (1.01,4.19)
2.4 (1.71,3.09)
1.9 (0.4,3.39)
1.1 (0.27,2.47)
zone 4
7.2 (2.27,12.13)
7.2 (5.28,9.12)
7.3 (1.59,13.01)
2.5 (0.85,5.85)†
.026
zone 5
6.4 (2.49,10.45)
5.6 (4.6,6.60)
4.5 (1.21,10.21)
1.5 (0.88,2.12)*†
.044* .049†
zone 6
1.3 (0.35,2.25)
1.7 (0.85,2.55)
1.2 (0.46,1.94)
0.2 (0.05,0.45)†
.011
zone 1
0.0 (0.0,0.0)
0.0 (0.0,0.0)
0.0 (0.0,0.0)
0.01 (0.01,0.01)
zone 2
0.01 (0.01,0.02)
0.0 (0.0,0.0)
0.0 (0.0,0.0)
0.0 (0.0,0.0)
zone 3
0.0 (0.0,0.0)
0.01 (0.01,0.01)
0.0 (0.0,0.0)
0.01 (0.01,0.01)
zone 4
0.04 (0.04,0.04)
0.04 (0.04,0.04)
0.02 (0.02,0.02)
0.03 (0.012,0.04)
zone 5
0.1 (0.01,0.01)
0.1 (0.01,0.01)
0.05 (0.05,0.05)*
.002
5.4 (3.56,7.3)*
zone 6
0.1 (0.01,0.01)
0.04 (0.04,0.04)
0.01 (0.01,0.01)*
.003
0.03 (0.02,0.04)
zone 1
0.1 (0.01,0.03)
0.04 (0.04,0.04)
0.0 (0.0,0.0)
0.01 (0.01,0.01)
zone 2
0.0 (0.0,0.0)
0.02 (0.02,0.02)
0.02 (0.02,0.02)
0.0 (0.0,0.0)*
.001
zone 3
0.1 (0.1,0.1)
0.1 (0.01,0.01)
0.05 (0.05,0.05)
0.01 (0.010,0.02)*
.037
zone 4
1.2 (1.18,1.22)
0.1 (0.01,0.01)
0.2 (0.08,0.32)
0.03 (0.01,0.07)*‡
.021* .015‡
zone 5
0.2 (0.02,0.02)
0.1 (0.02,0.18)
0.1 (0.02,0.22)
0.02 (0.02,0.02)*†‡
.001* .024† .020‡
zone 6
0.03 (0.03,0.03)
0.03 (0.03,0.03)
0.04 (0.04,0.04)
0.0 (0.0,0.0)†‡
.041† .034‡
P < .05
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
Hit-ups/match
Tackles/match
Hit-ups/min
.007
Tackles/min
Note: Zone 1, 12 g. Total collisions comprise both hit-ups and tackles. *Significantly different (P < .05) from hit-up forwards. †Significantly different (P < .05) from wide-running forwards. ‡Significantly different (P < .05) from adjustables.
the outside backs produced significantly fewer tackles per match than wide-running forwards in zones 5 (21.4% vs 27.9%) and 6 (2.9% vs 8.5%) and the hit-up forwards in zone 5 (21.4% vs 29.6%). By contrast, McLellan et al2 observed no significant difference between the number of entries into each collision zone between
forwards and backs during both offensive and defensive match play. Direct comparisons of the number of collisions in each zone were not possible due to the different collision criteria adopted. McLellan et al2 used total collisions provided by manufacturer software, which included all forces (running, directional changes,
IJSPP Vol. 10, No. 6, 2015
Collisions in Elite Rugby League Matches 737
Table 2 Total Collision (Tackles and Hit-Ups) Frequency by Impact Zone, Mean (95% CI) Hit-up forwards (n = 4)
Wide-running forwards (n = 9)
zone 1
2.6 (0.53,4.67)
1.8 (1.34,2.26)
1.3 (0.9,2.79)
1.7 (0.09,3.31)
zone 2
1.9 (1.58,2.22)
1.6 (1.06,2.14)
1.3 (0.9,2.79)
0.7 (0.17,1.57)
zone 3
2.6 (0.70,4.51)
2.7 (1.85,3.55)
2.1 (0.61,3.59)
1.6 (0.14,3.34)
zone 4
8.7 (3.29,14.11)
9.13 (6.67,11.59)
8.1 (2.51,13.69)
5.4 (1.80,9.00)
zone 5
11.0 (4.65,17.36)
10.33 (8.71,11.94)
7.2 (1.6,12.79)
6.5 (4.27,8.73)
zone 6
3.5 (0.48,7.48)
3.33 (8.72,11.94)
1.8 (1.18,2.42)
2.4 (1.28,3.51)
zone 1
0.1 (0.06,0.26)
0.04 (0.04,0.04)
0.02 (0.02,0.02)
0.02 (0.02,0.02)
zone 2
0.1 (0.01,0.01)
0.03 (0.03,0.03)
0.02 (0.02,0.02)
0.01 (0.01,0.01)*
.002
zone 3
0.1 (0.01,0.01)
0.05 (0.05,0.05)
0.05 (0.05,0.05)
0.02 (0.02,0.02)*
.031
zone 4
0.2 (0.02,0.02)
0.17 (0.017,0.17)
0.19 (0.06,0.31)
0.06 (0.06,0.06)*†‡
.010* .038† .023‡
zone 5
0.3 (0.03,0.03)
0.21 (0.13,0.29)*
0.08 (0.08,0.08)*†
.000* .009†
zone 6
0.1 (0.1,0.1)
0.07 (0.07,0.07)
0.04 (0.04,0.04)
0.03 (0.03,0.03)*
.010
Total high collisions (zones 5 and 6)
14.5 (4.48,24.52)
13.9 (11.13,16.67)
9.0 (2.79,15.2)
8.9 (7.53,10.27)
Total collisions/min (zones 5 and 6)
0.4 (0.24,0.56)
0.28 (0.13,0.29)*
High collision frequency (zones 5 and 6)
1 every 2.5 min
1 every 4.3 min
P < .05
Adjustables (n = 5)
P < .05
Outside backs (n = 5)
P < .05
Collisions/match
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
Collisions/min
.010
.018
0.16 (0.04,0.28)*
0.2 (0.08,0.32)*
1 every 5 min
.007
.005
0.11 (0.11,0.11)*†
.000* .010†
1 every 9 min
Note: Zone 1, 12 g. *Significantly different (P < .05) from hit-up forwards. †Significantly different (P < .05) from wide-running forwards. ‡Significantly different (P < .05) from adjustables.
and player-to-player collisions) experienced by a player during match play. The heavily weighted vertical forces experienced during running and walking, as well as impacts from directional changes, were not differentiated from the physical collisions of hit-ups and tackles and may therefore explain the similar collision profiles seen between forward and back playing positions. Even when directional changes and minor collisions (zones 1 and 2)2 were removed from the data, an average of 464 impacts per player were recorded.15 This is substantially greater than the collision number previously reported (600 per match, 300 per team),17 suggesting caution on inspecting this data.15 The rate of collisions (number of collisions per minute of match play) differed between playing positions for both offensive and defensive play in each collision zone. Adjustables were exposed to significantly fewer high-collision (zones 5 and 6) hit-ups per minute of match play than hit-up forwards. This difference may be attributed to the adjustables directing play and being the intermediary between attack and defense. As expected, outside backs performed fewer tackles per match than all other positions (67% less than hit-up forwards). The most notable differences were observed between backs and hit-up forwards across zones 2 to 5, with hit-up forwards executing 10 times the number of tackles per minute at moderate and high collision bands (zones 3 and 5) than outside backs. The
hit-up forwards experienced almost 1 collision per minute, with 1 collision every 2.5 minutes sustained in the highest recorded collision forces of zones 5 and 6. In addition, in zones 5 and 6 they performed 1 tackle every 2 minutes throughout the match compared with 1 tackle every 9 minutes for outside backs. The repeated physical collisions and higher-frequency collisions by hit-up and wide-running forwards underscores the need to tailor training toward contact and collision conditioning in these playing groups. Physical conditioning programs based on the rate of collisions as opposed to mean collision values will likely result in players being better prepared for intense periods of match play and may reduce injury incidence. The risk of muscle damage is increased in this group, where moderate collisions have been shown to cause significant skeletal-muscle damage,2 and exposure to repetitive subconcussive collisions may increase the risk of traumatic brain injury.27 To date, no comparative data quantifying attacking- and defensive-play collisions by zone categories are available. However, a similar study1 broadly defined collisions as mild, moderate, and heavy without assigning g-forces to each category. Comparatively fewer collisions in every zone were observed in the current study. Although relative collisions (per minute of match play) for mild and heavy collisions across positions were similar between studies, fewer relative collisions across all positions for moderate collisions
IJSPP Vol. 10, No. 6, 2015
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
738 Cummins and Orr
(zones 3 and 4)1 were observed in this study: hit-up forwards (0.3 vs 0.58), wide-running forwards (0.22 vs 0.39), adjustables (0.24 vs 0.33), and outside backs (0.08 vs 0.17). The disparity in findings likely reflects inherent differences between the playing styles or physical caliber of the individual teams or the inclusion of incidental collision events. Inconsistency of collision-zone classifications in the literature1–3,16 hinders direct comparisons with the current study. The 6 zone classifications used by McLellan et at2 were modeled from collision classifications designed for rugby union.5 Due to the inherent differences in the 2 rugby codes, these zones may be inadequate in reflecting the collision demands of rugby league. A recent study reported on the collisions experienced in attacking and defensive play by forward playing positions but attributed g-force classification zones (mild, 1–2 g; moderate, 2.1–4 g; and heavy, >4 g) without providing a rationale.3 Clarity and consistency of collision zones and definitions in rugby league would facilitate precise comparison and analysis of performance between individual players, teams, levels of competition, and seasons.16 In rugby league, collisions and tackles are commonly regarded as the most demanding aspect of match play,28 making measurement of collisions important from both an injury-prevention and a physical-conditioning perspective. An understanding of the frequency and forces associated with collision events provides the opportunity to develop position-specific collision and tackle conditioning programs for elite rugby league players. This study is not without limitations. In using player GPS files across a season, there is a pseudo replication in reporting multiple measures from individual players in the same positional group. In addition, the GPS devices used in this study record accelerations up to 8 g in each axis, giving a peak resultant acceleration of up to 12.8 g. This capability may be insufficient in differentiating moderate- and high-collision events due to the attenuation of the accelerometer signal. Facilitated by the inclusion of a gyroscope, the Catapult microtechnology units are the only devices that have validated automatic collision detection and quantified the contact load in collision sports. Future research investigating the validity of automatic collision detection in GPSports devices is warranted.
Conclusion In conclusion, this is the first study, to our knowledge, to quantify the intensity of collisions in attacking and defensive play between forward and back playing positions in elite rugby league match play. The findings demonstrate that during the course of a match, players are exposed to a significant number of collision events. Hit-up and wide-running forwards experience greater collision events than adjustables and outside backs, highlighting the presence of positional-play differences. Although these results may be unique to the individual team’s defensive and attacking play strategies, they are indicative of the significant collision profiles in professional rugby league.
Practical Implications • Wide-running and hit-up forwards experienced greater collisions per match in defensive rather than attacking play. Forward players should undertake specific collision training to tolerate the heavy contact demands associated with defensive play. • Attacking and defensive play in rugby league are executed with considerable collision forces and add to the cumulative physical load experienced by players during match play.
• Positional-play collision-profile differences highlight the need for position-specific collision training to prepare players for the physical demands of competition. Acknowledgments This manuscript contributes to C.C.’s PhD qualification. The authors wish to thank the players and staff of the participating rugby league club for their support of this project. No funding has been received for the preparation of this manuscript. The authors declare that there are no conflicts of interest that are directly relevant to the context of this review.
References 1. Gabbett TJ, Jenkins D, Abernethy B. Physical demands of professional rugby league training and competition using microtechnology. J Sci Med Sport. 2012;15(1):80–86. PubMed doi:10.1016/j. jsams.2011.07.004 2. McLellan CP, Lovell D, Gass G. Biochemical and endocrine responses to impact and collision during elite rugby league match play. J Strength Cond Res. 2011;25(6):1553–1562. PubMed doi:10.1519/ JSC.0b013e3181db9bdd 3. Gabbett TJ, Polley C, Dwyer D, Kearney S, Corvo A. Influence of field position and phase of play on the physical demands of match-play in professional rugby league forwards. J Sci Med Sport. 2014;17(5):556– 561. PubMed doi:10.1016/j.jsams.2013.08.002 4. Kempton T, Sirotic A, Coutts A. An integrated analysis of match-related fatigue in professional rugby league. J Sports Sci. 2015;33(1):39–47. PubMed doi:10.1080/02640414.2014.921832 5. Cunniffe B, Proctor W, Baker J, Davies B. An evaluation of the physiological demands of elite rugby union using global positioning system tracking software. J Strength Cond Res. 2009;23(4):1195–1203. PubMed doi:10.1519/JSC.0b013e3181a3928b 6. Suarez-Arrones L, Portillo L, Gonzalez-Rave J, Munoz V, Sanchez F. Match running performance in Spanish elite male rugby union using global positioning system. Isokinet Exerc Sci. 2012;20(2):77–83. 7. Venter R, Opperman E, Opperman S. The use of global positioning system (GPS) tracking devices to assess movement demands and impacts in under-19 rugby union match play. Afr J Phys Health Educ Recr Dance. 2011;17:1–8. 8. Gastin PB, McLean O, Spittle M, Breed R. Quantification of tackling demands in professional Australian football using integrated wearable athlete tracking technology. J Sci Med Sport. 2013;16(6):589–593. PubMed doi:10.1016/j.jsams.2013.01.007 9. Varley MC, Fairweather I, Aughey R. Validity and reliability of GPS for measuring instantaneous velocity during acceleration, deceleration, and constant motion. J Sports Sci. 2012;30(2):121–127. PubMed doi :10.1080/02640414.2011.627941 10. Wundersitz DW, Netto KJ, Aisbett B, Gastin PB. Validity of an upperbody-mounted accelerometer to measure peak vertical and resultant force during running and change-of-direction tasks. Sports Biomech. 2013;12(4):403–412. PubMed 11. Gabbett T, Jenkins D, Abernethy B. Physical collisions and injury during professional rugby league skills training. J Sci Med Sport. 2010;13(6):578–583. PubMed doi:10.1016/j.jsams.2010.03.007 12. Gastin PB, Mclean O, Breed R, Spittle M. Tackle and impact detection in elite Australian football using wearable microsensor technology. J Sports Sci. 2014;32(10):947–953. PubMed doi:10.1080/02640414.2 013.868920 13. Kelly D, Coughlan G, Green B, Caulfield B. Automatic detection of collisions in elite level rugby union using a wearable sensing device. Sports Eng. 2012;15(2):81–92. doi:10.1007/s12283-012-0088-5
IJSPP Vol. 10, No. 6, 2015
Downloaded by The University of Calgary on 09/16/16, Volume 10, Article Number 6
Collisions in Elite Rugby League Matches 739
14. Kelly SJ, Murphy A, Watsford M, Austin D, Rennie M. Reliability and validity of sports accelerometers during static and dynamic testing. Int J Sports Physiol Perform. 2015;10(1):106–111. PubMed doi:10.1123/ ijspp.2013-0408 15. Gabbett TJ. Quantifying the physical demands of collision sports: does microsensor technology measure what it claims to measure? J Strength Cond Res. 2013;27(8):2319–2322. PubMed doi:10.1519/ JSC.0b013e318277fd21 16. Cummins C, Orr R, O’Connor H, West C. Global positioning systems (GPS) and microtechnology sensors in team sports: a systematic review. Sports Med. 2013;43(10):1025–1042. PubMed doi:10.1007/ s40279-013-0069-2 17. Gabbett TJ, Jenkins D, Abernethy B. Physical collisions and injury in professional rugby league match-play. J Sci Med Sport. 2011;14(3):210–215. PubMed doi:10.1016/j.jsams.2011.01.002 18. Gissane C, White J. Physical collisions in professional Super League rugby, the demands on different player positions. Cleve Med J. 2001 4:137–146. 19. Dave S, Craig T, Shayne H, Ceri N, Kevin L. Semi-automated timemotion analysis of senior elite rugby league. Int J Perform Anal Sport. 2009;9(1):47–59. 20. Gabbett TJ. Effects of physical, technical, and tactical factors on final ladder position in semiprofessional rugby league. Int J Sports Physiol Perform. 2014;9(4):680–688. PubMed doi:10.1123/IJSPP.2013-0253 21. Kempton T, Sirotic A, Rampinini E, Coutts A. Metabolic power demands of rugby league match play. Int J Sports Physiol Perform. 2015;10(1):23–28. PubMed doi:10.1123/ijspp.2013-0540
22. National Rugby League. Rugby League Laws of the Game International Level With Notes on the Laws and NRL Telstra Premiership Interpretations. National Rugby League; 2013. 23. King D, Hume P, Clark T. Nature of tackles that result in injury in professional rugby league. Res Sports Med. 2012;20(2):86–104. PubMed 24. Viera AJ, Garrett J. Understanding interobserver agreement: the kappa statistic. Fam Med. 2005;37(5):360–363. PubMed 25. Hopkins WG, Marshall S, Batterham A, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc. 2009;41(1):3–13. PubMed doi:10.1249/ MSS.0b013e31818cb278 26. McLellan CP, Lovell D, Gass G. Performance analysis of elite rugby league match play using global positioning systems. J Strength Cond Res. 2011;25(6):1703–1710. PubMed doi:10.1519/ JSC.0b013e3181ddf678 27. Breedlove EL, Robinson M, Talavage T, et al. Biomechanical correlates of symptomatic and asymptomatic neurophysiological impairment in high school football. J Biomech. 2012;45(7):1265–1272. PubMed doi:10.1016/j.jbiomech.2012.01.034 28. Brewer J, Davis J. Applied physiology of rugby league. Sports Med. 1995;20(3):129–135. PubMed doi:10.2165/00007256-19952003000001
IJSPP Vol. 10, No. 6, 2015
