Training & Testing
485
Activity Profiles of Successful and Less-successful Semi-elite Rugby League Teams
Authors
B. T. Hulin1, 2, T. J. Gabbett1, 3
Affiliations
1
Key words ▶ time-motion analysis ● ▶ performance ● ▶ team sport ●
Abstract
School of Exercise Science, Australian Catholic University, Brisbane, Australia School of Medicine, The University of Wollongong, Australia 3 School of Human Movement Studies, The University of Queensland, Brisbane, Australia
▼
This study investigated whether match intensities during predefined periods differed among successful and less-successful rugby league teams. 4 semi-elite rugby league teams were split into ‘high-success’ and ‘low-success’ groups based on their success rates. Movement was recorded using a global positioning system (10 Hz) during 20 rugby league matches. Following the peak ball-in-play time period, the highsuccess group was able to maintain ball-in-play time that was: (1) 22 % greater than the low-success group (P = 0.01) and (2) greater than their mean period of match-play (P = 0.01). In the peak and mean periods of match play, hit-up forwards from the high-success group covered less total distance (P = 0.02; P = 0.01), less high-intensity
Introduction
▼ accepted after revision November 27, 2014 Bibliography DOI http://dx.doi.org/ 10.1055/s-0034-1398532 Published online: March 3, 2015 Int J Sports Med 2015; 36: 485–489 © Georg Thieme Verlag KG Stuttgart · New York ISSN 0172-4622 Correspondence Dr. Tim J Gabbett School of Exercise Science Australian Catholic University 1100 Nudgee Road Brisbane 4014 Australia Tel.: + 61/7/3263 7589 Fax: + 61/7/3263 7589 [email protected]
There is conflicting evidence relating to differences in the intensity of match-play between successful and less successful rugby league teams. In a study of one elite team, it was shown that competitive success was associated with greater total distance (TD) covered [8]; these findings were confirmed by others that investigated match intensities between successful and less-successful semi-elite rugby league teams [3]. However, more recent evidence comparing separate elite teams with high- and low-success rates demonstrated that greater running workloads were not indicative of success in rugby league match-play, and that teams with higher success rates were involved in a greater number of collisions [16]. The finding that less successful teams cover greater TD [16] may be expected given that defending has been associated with greater TD than attacking [12]. Previous investigations [16] quantified activity profiles in 5-min periods in order to examine if the intensity of activity dur-
running distance (P = 0.01; P = 0.01) and were involved in a greater number of collisions (P = 0.03; P = 0.01) than hit-up forwards from the low-success group. These results demonstrate that greater amounts of high-intensity running and total distance are not related to competitive success in semi-elite rugby league. Rather, competitive success is associated with involvement of hit-up forwards in a greater number of collisions and the ability of high-success teams to maintain a higher ball-in-play time following the peak period. Strength and conditioning programs that: (1) emphasize high-intensity running and neglect to combine these running demands with collisions, and (2) do not offer exposure to match specific ball-in-play time demands, may not provide sufficient physiological preparation for teams to be successful in rugby league.
ing predefined periods of match-play differed between successful and less-successful teams. Interestingly, the TD covered in the mean period of match play (67–69 m · min − 1, depending on position) by the team with a greater success rate [16] may be considered ‘low’ in comparison with the locomotive rates demonstrated by others, i. e. 93–101 m · min − 1 [10, 22, 23]. These differences in running data may lead to circumspect interpretation of the findings of Hulin et al. [16] that more successful teams cover less TD, as the result may indeed be due to outlying data, suggesting that the findings were potentially a one-off ‘coincidence’. Similar methods of segmental global positioning system (GPS) data analysis have been used by others [21] to examine the effect of the most intense 5-min period of match-play on physical and technical performance in elite senior and junior ‘adjustable’ positions (i. e. hooker, halfback, five-eighth, and fullback) within rugby league teams having low success rates (elite, 36 % of matches won; junior elite, 13 % of matches
Hulin BT, Gabbett TJ. Activity Profiles of Successful … Int J Sports Med 2015; 36: 485–489
Downloaded by: Chinese University of Hong Kong. Copyrighted material.
2
486 Training & Testing
Material & Methods
▼
Participants
77 players (mean ± SD age; 23.9 ± 3.2 years) from 4 semi-elite rugby league teams competing in the Queensland Cup Rugby League competition participated in this study. The institutional review board for human investigation approved all experimental procedures, and informed consent was obtained from all participants. All experimental procedures were in accordance with international standards in sport and exercise science research [14].
Procedures
Global positioning system (GPS) analysis was completed during 20 semi-elite rugby league matches. 4 rugby league teams were split into ‘high-success’ and ‘low-success’ groups based on the percentage of matches won throughout the team’s season. The high-success group consisted of teams that finished 1st and 2nd on the competition ladder, and that won 75 % of matches played, whereas the low-success group consisted of teams that finished 8th and 11th and had a combined winning percentage of 32 %. The 20 matches analyzed (11 for high-success group, 9 for low-success group) equated to 200 GPS files. Positional groups were categorized into adjustables (hooker, halfback, five-eighth, and fullback), hit-up forwards (props, second-rowers, and locks), and outside backs (centre and wing) [2]. First and second half GPS, and ball-in-play data for each match were separated into 8 equal periods (mean ± SD; 319 ± 24 s). Thus, each match was separated into 16 periods of approximately 5-min, based on the duration of each half. A customized Microsoft Excel (Microsoft, Redmond, USA) spreadsheet was used to identify periods which contained either the maximum: (1) ball-in-play time, (2) TD, (3) high-intensity running (HIR) distance, (4) number of collisions, or (5) number of RHIE bouts. These periods were referred to as the ‘peak period’ for each variable. Additionally, the period following the peak period for each
variable was identified and referred to as the ‘subsequent period’, and the average of the remaining 14 periods was also calculated and referred to as the ‘mean period’. In order to ensure any variances in the activity profiles of the 2 groups were not due to differences in phases of play (i. e. time in attack and defense), the percentage of time spent in possession during each match was also calculated.
Global positioning system analysis
Movement was recorded using a minimaxX GPS unit (Catapult Innovations, Melbourne, Australia) sampling at 10 Hz, and subsequently analyzed using the manufacturer’s software (Catapult Sprint, 5.1.0.1). Whilst the GPS signal provided information on velocity, distance, and acceleration, the GPS unit includes a triaxial accelerometer and gyroscope sampling at 100 Hz, which provided information on physical collisions and repeated highintensity effort (RHIE) bouts. Collisions were defined as a spike in instantaneous ‘player load’ shortly before a change in orientation of the GPS unit [9]. Instantaneous player load was calculated as the square root of the sum of the tri-axial accelerometer [9]. A RHIE bout was defined as 3 or more high acceleration ( ≥ 2.79 m.s − 2) [1], high velocity ( > 5 m.s − 1), or contact efforts with less than 21 s recovery between efforts [10]. The unit was worn in a small vest on the upper back of the players. Data were categorized into: (1) movement velocity bands corresponding to low (0–5 m.s − 1) and high ( > 5 m.s − 1) intensities, with TD covered represented by the summation of distances covered at these 2 intensity bands, (2) number of collisions; and (3) RHIE bouts. The 10 Hz minimaxX units have been shown to have acceptable validity and reliability for measuring sprinting velocities, accelerations [20], collisions [9], and RHIE bouts [7] commonly observed in rugby league.
Statistical analysis
Independent t-tests were used to compare means of high- and low-success groups for the percentage of time spent in possession. Factorial analysis of variance (ANOVA) tests were used to identify differences between the peak, subsequent, and mean periods for the same team and to identify differences between peak, subsequent, and mean periods of high- and low-success teams. A practical approach determining the magnitude of difference between high- and low-success teams was also used. Specifically, Cohen’s effect size (ES) statistic and 90 % confidence intervals (CI) were used to determine the magnitude of any differences [5]. The magnitude of the ES was classified as trivial ( ≤ 0.2), small (0.21–0.6), moderate (0.61–1.2), and large (1.21– 2.0) [15]. Statistical significance was set at P
485
Activity Profiles of Successful and Less-successful Semi-elite Rugby League Teams
Authors
B. T. Hulin1, 2, T. J. Gabbett1, 3
Affiliations
1
Key words ▶ time-motion analysis ● ▶ performance ● ▶ team sport ●
Abstract
School of Exercise Science, Australian Catholic University, Brisbane, Australia School of Medicine, The University of Wollongong, Australia 3 School of Human Movement Studies, The University of Queensland, Brisbane, Australia
▼
This study investigated whether match intensities during predefined periods differed among successful and less-successful rugby league teams. 4 semi-elite rugby league teams were split into ‘high-success’ and ‘low-success’ groups based on their success rates. Movement was recorded using a global positioning system (10 Hz) during 20 rugby league matches. Following the peak ball-in-play time period, the highsuccess group was able to maintain ball-in-play time that was: (1) 22 % greater than the low-success group (P = 0.01) and (2) greater than their mean period of match-play (P = 0.01). In the peak and mean periods of match play, hit-up forwards from the high-success group covered less total distance (P = 0.02; P = 0.01), less high-intensity
Introduction
▼ accepted after revision November 27, 2014 Bibliography DOI http://dx.doi.org/ 10.1055/s-0034-1398532 Published online: March 3, 2015 Int J Sports Med 2015; 36: 485–489 © Georg Thieme Verlag KG Stuttgart · New York ISSN 0172-4622 Correspondence Dr. Tim J Gabbett School of Exercise Science Australian Catholic University 1100 Nudgee Road Brisbane 4014 Australia Tel.: + 61/7/3263 7589 Fax: + 61/7/3263 7589 [email protected]
There is conflicting evidence relating to differences in the intensity of match-play between successful and less successful rugby league teams. In a study of one elite team, it was shown that competitive success was associated with greater total distance (TD) covered [8]; these findings were confirmed by others that investigated match intensities between successful and less-successful semi-elite rugby league teams [3]. However, more recent evidence comparing separate elite teams with high- and low-success rates demonstrated that greater running workloads were not indicative of success in rugby league match-play, and that teams with higher success rates were involved in a greater number of collisions [16]. The finding that less successful teams cover greater TD [16] may be expected given that defending has been associated with greater TD than attacking [12]. Previous investigations [16] quantified activity profiles in 5-min periods in order to examine if the intensity of activity dur-
running distance (P = 0.01; P = 0.01) and were involved in a greater number of collisions (P = 0.03; P = 0.01) than hit-up forwards from the low-success group. These results demonstrate that greater amounts of high-intensity running and total distance are not related to competitive success in semi-elite rugby league. Rather, competitive success is associated with involvement of hit-up forwards in a greater number of collisions and the ability of high-success teams to maintain a higher ball-in-play time following the peak period. Strength and conditioning programs that: (1) emphasize high-intensity running and neglect to combine these running demands with collisions, and (2) do not offer exposure to match specific ball-in-play time demands, may not provide sufficient physiological preparation for teams to be successful in rugby league.
ing predefined periods of match-play differed between successful and less-successful teams. Interestingly, the TD covered in the mean period of match play (67–69 m · min − 1, depending on position) by the team with a greater success rate [16] may be considered ‘low’ in comparison with the locomotive rates demonstrated by others, i. e. 93–101 m · min − 1 [10, 22, 23]. These differences in running data may lead to circumspect interpretation of the findings of Hulin et al. [16] that more successful teams cover less TD, as the result may indeed be due to outlying data, suggesting that the findings were potentially a one-off ‘coincidence’. Similar methods of segmental global positioning system (GPS) data analysis have been used by others [21] to examine the effect of the most intense 5-min period of match-play on physical and technical performance in elite senior and junior ‘adjustable’ positions (i. e. hooker, halfback, five-eighth, and fullback) within rugby league teams having low success rates (elite, 36 % of matches won; junior elite, 13 % of matches
Hulin BT, Gabbett TJ. Activity Profiles of Successful … Int J Sports Med 2015; 36: 485–489
Downloaded by: Chinese University of Hong Kong. Copyrighted material.
2
486 Training & Testing
Material & Methods
▼
Participants
77 players (mean ± SD age; 23.9 ± 3.2 years) from 4 semi-elite rugby league teams competing in the Queensland Cup Rugby League competition participated in this study. The institutional review board for human investigation approved all experimental procedures, and informed consent was obtained from all participants. All experimental procedures were in accordance with international standards in sport and exercise science research [14].
Procedures
Global positioning system (GPS) analysis was completed during 20 semi-elite rugby league matches. 4 rugby league teams were split into ‘high-success’ and ‘low-success’ groups based on the percentage of matches won throughout the team’s season. The high-success group consisted of teams that finished 1st and 2nd on the competition ladder, and that won 75 % of matches played, whereas the low-success group consisted of teams that finished 8th and 11th and had a combined winning percentage of 32 %. The 20 matches analyzed (11 for high-success group, 9 for low-success group) equated to 200 GPS files. Positional groups were categorized into adjustables (hooker, halfback, five-eighth, and fullback), hit-up forwards (props, second-rowers, and locks), and outside backs (centre and wing) [2]. First and second half GPS, and ball-in-play data for each match were separated into 8 equal periods (mean ± SD; 319 ± 24 s). Thus, each match was separated into 16 periods of approximately 5-min, based on the duration of each half. A customized Microsoft Excel (Microsoft, Redmond, USA) spreadsheet was used to identify periods which contained either the maximum: (1) ball-in-play time, (2) TD, (3) high-intensity running (HIR) distance, (4) number of collisions, or (5) number of RHIE bouts. These periods were referred to as the ‘peak period’ for each variable. Additionally, the period following the peak period for each
variable was identified and referred to as the ‘subsequent period’, and the average of the remaining 14 periods was also calculated and referred to as the ‘mean period’. In order to ensure any variances in the activity profiles of the 2 groups were not due to differences in phases of play (i. e. time in attack and defense), the percentage of time spent in possession during each match was also calculated.
Global positioning system analysis
Movement was recorded using a minimaxX GPS unit (Catapult Innovations, Melbourne, Australia) sampling at 10 Hz, and subsequently analyzed using the manufacturer’s software (Catapult Sprint, 5.1.0.1). Whilst the GPS signal provided information on velocity, distance, and acceleration, the GPS unit includes a triaxial accelerometer and gyroscope sampling at 100 Hz, which provided information on physical collisions and repeated highintensity effort (RHIE) bouts. Collisions were defined as a spike in instantaneous ‘player load’ shortly before a change in orientation of the GPS unit [9]. Instantaneous player load was calculated as the square root of the sum of the tri-axial accelerometer [9]. A RHIE bout was defined as 3 or more high acceleration ( ≥ 2.79 m.s − 2) [1], high velocity ( > 5 m.s − 1), or contact efforts with less than 21 s recovery between efforts [10]. The unit was worn in a small vest on the upper back of the players. Data were categorized into: (1) movement velocity bands corresponding to low (0–5 m.s − 1) and high ( > 5 m.s − 1) intensities, with TD covered represented by the summation of distances covered at these 2 intensity bands, (2) number of collisions; and (3) RHIE bouts. The 10 Hz minimaxX units have been shown to have acceptable validity and reliability for measuring sprinting velocities, accelerations [20], collisions [9], and RHIE bouts [7] commonly observed in rugby league.
Statistical analysis
Independent t-tests were used to compare means of high- and low-success groups for the percentage of time spent in possession. Factorial analysis of variance (ANOVA) tests were used to identify differences between the peak, subsequent, and mean periods for the same team and to identify differences between peak, subsequent, and mean periods of high- and low-success teams. A practical approach determining the magnitude of difference between high- and low-success teams was also used. Specifically, Cohen’s effect size (ES) statistic and 90 % confidence intervals (CI) were used to determine the magnitude of any differences [5]. The magnitude of the ES was classified as trivial ( ≤ 0.2), small (0.21–0.6), moderate (0.61–1.2), and large (1.21– 2.0) [15]. Statistical significance was set at P
