Pacing commonly occurs in self-paced endurance events (e.g., cycling and running) in an attempt to prevent fatigue. Studies have shown that the pacing strategy employed is a major determinant of performance during continuous exercise lasting longer than 30 seconds (1). It has also been found that the anticipation of exercise-induced stress influences performance during continuous exercise (1). However, it is important to note that not all exercise is continuous; many sports involve short bouts of high-intensity efforts interspersed with varying recovery periods (24).
It has been proposed that pacing strategies are both consciously and subconsciously regulated (10). The knowledge of the end point of exercise (also referred to as “anticipatory pacing”) has been demonstrated to be a major factor in the allocation of physiological resources during exercise (5). Recently, researchers studied the influence of the knowledge of sprint number on pacing strategies employed during repeated sprint activity (5). The findings of this study demonstrated that pacing does, in fact, occur during repeated high-intensity sprint activity and that this pacing is influenced by the knowledge of sprint number and anticipation of the exercise end point (5). Although subconscious pacing occurs based on the athlete's knowledge of the impending activity, during exercise, afferent information is also delivered from the muscles and cardiorespiratory systems to the brain; the brain interprets this information against expected outcomes and perceived effort and makes a conscious decision to either increase or decrease exercise intensity (10). In this respect, pacing strategies are employed to not only protect physiological systems and uphold homeostasis by decreasing intensity but also improve the likelihood of competitive success by increasing physical effort (4,10).
It has also recently been shown that pacing may occur in prolonged high-intensity intermittent team sports (8,15). Research has shown that team sport athletes reduce the amount of low-speed activity performed in competition to adequately perform the repeated high-intensity components of match play (8,15). Edwards and Noakes (10) found that during the second half of a soccer match there was a decrease in the amount of physical efforts performed, suggesting that either players were affected by cumulative fatigue or that players employed a subconscious pacing strategy to ensure they reached the end of the game. The more physically demanding periods of play are usually followed by extended periods of low-speed activity (20); therefore, players appear to temporarily experience symptoms of fatigue during a game and attempt to maintain homeostasis by subconsciously pacing (10). This may explain the ability of whole-game players to complete the full duration of a match regardless of varying fitness levels and remain in a reasonable physical condition (10).
There is currently limited information on whether pacing occurs during rugby league match play and whether players from different positions employ similar pacing strategies. Waldron et al. (27) recently studied the pacing strategies of a single team of rugby league players who either played the entire match or were interchanged throughout the match. Consistent with previous findings in rugby league (23,25), the authors found that transient fatigue occurred throughout a match. However, in contrast to players who played an entire match, interchanged players set a higher pacing strategy and also exhibited an “end-spurt” in the later stages of the match (27). Although these authors provided important information on the different pacing strategies employed by whole-game and interchanged players, there were some limitations of this study that restrict its application. First, forwards (the interchanged players) were compared against backs (players who played the entire game). Given the differences in physical qualities (19) and demands of competition (14,25,26) between forwards and backs, any comparisons between interchanged and “whole game” players should be made between players of similar positions. Second, and equally importantly, data were collected on a limited sample (n = 35 individual appearances) and from only one team. To gain a greater understanding of pacing within rugby league match play, these limitations should be addressed.
In addition to different pacing strategies between whole-game and interchanged players, it is highly likely that pacing strategies may differ between successful and less successful teams. Research has suggested that the competitive advantage of successful elite rugby league teams is closely linked to their ability to maintain a higher playing intensity than their less successful opposition (11). However, to date, no research has investigated whether pacing strategies differ between winning and losing teams.
Based on the current evidence regarding pacing in rugby league, the present study compared the pacing strategies of whole-game and interchanged rugby league players drawn from a sample of 11 teams competing in the semi-elite Queensland Cup competition. We extend on the work of Waldron et al. (27) by comparing players from the same position. Furthermore, the pacing strategies of winning and losing teams were investigated. We hypothesized that interchanged players and those from winning teams would set a higher pacing strategy than players who played the entire match, or competed in losing teams.
Experimental Approach to the Problem
The pacing strategies of rugby league players were analyzed by comparing teams from an entire rugby league competition. First and second half data were divided into quartiles so that each player had 8 quartiles by the end of the match. Data were divided into winning and losing teams, and whole-game and interchanged players. Differences in the physical demands and pacing strategies of whole-game and interchanged players, and winning and losing teams were compared using magnitude-based inferences (Cohen's effect size [ES] statistic and 90% confidence intervals [CIs]), employing a practical approach based on the real-world relevance of the results.
At the beginning of the 2012 rugby league season, the 12 teams competing in the Queensland Cup rugby league competition were invited to participate in a study of the physical demands of semi-elite rugby league match play. All teams but 1 agreed to participate in the study. The final sample included 52 male rugby league players (mean ± SD age; 24 ± 3 yr). Players were participants from 1 of the 11 remaining teams competing in the competition. All participants received a clear explanation of the study, including information on the risks and benefits, and written consent was obtained. All experimental procedures were approved by the Institutional Review Board for Human Investigation.
Global positioning system (GPS) analysis was completed during 26 matches (totaling 90 appearances). Players were selected from 3 positional groups representing the second rowers, locks, and hookers. As backs are rarely interchanged throughout a match, and props rarely play an entire match, these positional groups were excluded from the analysis. Performances were divided into 4 distinct performance bouts. These were whole-game first half, whole-game second half, interchange bout 1 (players either replacing or being replaced by another player for their first bout on the field), and interchange bout 2 (players replacing others or being replaced by another player for a second bout on the field). All bout 1 interchanges occurred during the first half and all bout 2 interchanges took place in the second half. Each playing half or bout was further divided into match quartiles. Therefore, each player had a total of 8 “quartiles” by the end of the match. Whole match players consisted of 39 players (n = 9 hookers; n = 4 locks; n = 26 second rowers) and 22 interchanged players (n = 5 hookers; n = 7 locks; n = 10 second rowers). There were 32 winning files analyzed (n = 20 whole match winning files; n = 12 interchanged winning files) and 54 losing files analyzed (n = 35 losing whole match files; n = 19 losing interchanged files).
Movement was recorded by a minimaxX GPS unit (Catapult Innovations, Melbourne, Australia) sampling at 10 Hz. The GPS signal provided information on speed, distance, position, and acceleration. The GPS unit also included tri-axial accelerometers and gyroscopes sampling at 100 Hz to provide information on physical collisions and repeated high-intensity efforts. The unit was worn in a small vest, on the upper back of the players.
Although recommendations for reporting GPS data have been presented (9), there is generally a wide range of reporting methods employed in the scientific literature (14,27). To date, there are no standardized methods for reporting velocity “zones,” and several definitions of what constitutes an “effort” (2,9,14). The majority of previous studies of team sport demands have employed 5 Hz technology (14,27) and as such have employed broad movement velocity bands because of the large measurement error associated with lower sampling frequencies. To allow comparisons with other researchers (14), the present data were categorized into (a) movement speed bands, corresponding to low (0–5 m·s−1) and high (>5 m·s−1) speeds (2); (b) collisions; and (c) repeated high-intensity effort bouts. A repeated high-intensity effort bout was defined as 3 or more high accelerations (≥2.79 m·s−2) (2), high speed, or contact efforts with less than 21-seconds recovery between efforts (14). The 10 Hz minimaxX units have been shown to have acceptable validity and reliability for estimating longer distances at walking through to striding speeds (21) and good reliability and accuracy for measuring sprint efforts commonly performed by team sport athletes (6). The minimaxX units have also been shown to offer a valid measurement of tackles and repeated efforts commonly observed in collision sports (13).
Data from the 11 teams were dichotomized into winning and losing teams. Differences in the physical demands and pacing strategies of whole-game and interchanged players, and winning and losing teams were compared using Cohen's ES statistic (7) and 90% CIs. Effect sizes of <0.2, 0.2–0.6, 0.6–1.2, 1.2–2.0, and >2.0 were considered trivial, small, moderate, large, and very large, respectively (16). Magnitudes of differences between the two groups were classified as substantially greater or lesser when there was a ≥75% likelihood of the effect being equal to or greater than the smallest worthwhile change estimated as 0.2× between-subject SD (small ES). Effects with less certainty were classified as trivial and where the ±90% CI of the ES crossed the boundaries of ES −0.2 and 0.2, the effect was reported as unclear. All data are reported as mean ± SD.
Whole-Game vs. Interchanged Players
The interchanged players covered likely probable greater total distance across all quartiles of the match (91 ± 6 vs. 87 ± 4 m·min−1, ES = 0.66 ± 0.87, 83%). The total distance covered by interchanged players was very likely (ES = 0.67 ± 0.35, 100%) higher than whole-game players in quartile 1. Interchanged players also covered greater (ES = 0.44 ± 0.43, 93%) total distance in quartile 8 than whole-game players. The total distance covered by interchanged players was achieved almost certainly through greater low-speed distances in quartile 1 (ES = 0.74 ± 0.35, 100%) and quartile 8 (ES = 0.47 ± 0.47, 94%). The low-speed distance covered across all quartiles of the match was greater (ES = 0.86 ± 0.82, 92%) for the interchanged players compared with whole-game players. Interchanged players performed an almost certainly (ES = 1.86 ± 0.25, 100%) greater frequency of repeated high-intensity effort bouts across all quartiles of the match (1 every 4.8 ± 0.3 minutes vs. 1 every 7.3 ± 0.5 minutes) (Figure 1).
Winning vs. Losing Teams
The total distance covered across all quartiles of the match was very likely (ES = 1.03 ± 0.77, 96%) greater in the whole-game players from winning teams than the whole-game players from losing teams. The whole-game players from winning teams covered a likely probable (ES = 0.55 ± 0.55, 93%) greater total distance than the whole-game players from losing teams in quartile 4. The greater total distance covered in quartile 4 by whole-game players from the winning teams was achieved through likely probable (ES = 0.51 ± 0.54, 83%) greater low-speed distances compared with whole-game players from the losing teams. Whole-game players from winning teams covered very likely (ES = 1.05 ± 0.77, 96%) greater low-speed distance over the course of the match compared with whole-game players from losing teams. Whole-game players from winning teams covered more high-speed distance in quartile 7 (ES = 0.70 ± 0.50, 82%) than the whole-game players from losing teams. There was a likely probable (ES = 0.81 ± 0.79, 90%) greater frequency of repeated high-intensity effort bouts for whole-game players from winning teams (1 every 5.7 ± 3.1 minutes) than whole-game players from losing teams (1 every 8.2 ± 2.8 minutes) in quartile 5 (Figure 2).
The interchanged players from losing teams covered a higher total distance in quartile 8 (ES = 0.60 ± 0.52, 96%) compared with the interchanged players from winning teams. The greater total distance covered by interchanged players from losing teams was achieved through likely probable (ES = 0.56 ± 0.52, 95%) greater low-speed distances compared with interchanged players from winning teams. The interchanged players from losing teams possibly (ES = 0.76 ± 0.99, 75%) engaged in a greater frequency of repeated high-intensity effort bouts than interchanged players from winning teams during quartile 4. No other meaningful differences were found in physical demands between whole-game and interchanged players from winning and losing teams (Figure 3).
The interchanged hookers covered greater total distance than the whole-game hookers in quartiles 1 (ES = 1.26 ± 0.77, 99%) and 8 (ES = 0.68 ± 1.58, 78%). The total distance covered in quartiles 5 (ES = 0.69 ± 1.20, 83%) and 6 (ES = 0.68 ± 1.11, 84%) was greater for the whole-game hookers than the interchanged hookers. The greater total distance covered by interchanged hookers in quartiles 1 (ES = 1.25 ± 0.79, 99%) and 8 (ES = 0.70 ± 1.60, 78%) was achieved through greater distances at low-speeds than whole-game hookers. The interchanged hookers covered greater (ES = 0.73 ± 1.32, 78%) high-speed distances (7.0 ± 6.0 m·min−1) during quartile 5 compared with whole-game hookers (3.0 ± 4.0 m·min−1). The interchanged hookers performed a greater frequency (ES = 1.68 ± 0.44, 100%) of repeated high-intensity effort bouts across all quartiles of the match than the whole-game hookers (Figure 4).
Interchanged locks covered greater total distance in quartiles 1 (ES = 0.70 ± 1.07, 100%) and 2 (ES = 0.64 ± 1.68, 84%) than whole-game locks. Whole-game locks covered greater total distance in quartiles 3 (ES = 1.18 ± 0.59, 100%) and 8 (ES = 0.52 ± 0.86, 83%) than interchanged locks. The greater distances covered by interchanged locks in quartiles 1 and 2 and whole-game locks in quartiles 3 and 8 were achieved through greater distances at low speeds. The high-speed distances covered in quartiles 5 (ES = 0.57 ± 0.84, 78%) and 7 (ES = 0.73 ± 0.67, 84%) were greater for whole-game locks than interchanged locks. Interchanged locks performed a likely probable (ES = 0.96 ± 0.82, 83%) greater frequency of repeated high-intensity effort bouts across all quartiles of the match than whole-game locks (Figure 5).
Interchanged second rowers covered greater (ES = 0.56 ± 0.70, 88%) total distance in quartile 8 than whole-game second rowers, which was achieved through greater (ES = 0.65 ± 0.66, 92%) distances at low speeds. However, the high-speed distances covered during quartiles 1 (ES = 0.59 ± 0.50, 84%), 2 (ES = 0.80 ± 0.46, 96%), and 5 (ES = 0.97 ± 0.42, 100%) were greater in whole-game second rowers than interchanged second rowers. Interchanged second rowers almost certainly (ES = 1.64 ± 0.49, 100%) performed a greater frequency of repeated high-intensity effort bouts across all quartiles of the match than whole-game second rowers (Figure 6).
This study investigated the pacing strategies employed by a large sample of semi-professional rugby league teams. A novel aspect of this study was the comparison between whole-game and interchanged players, winning and losing teams and similar positional groups. The total distance and low-speed distance covered across all quartiles of the match, but specifically quartiles 1 and 8, were greater for interchanged players than whole-game players. The match outcome differentially affected the pacing strategies of whole-game and interchanged players. Whole-game players from winning teams set a higher pacing strategy than whole-game players from losing teams, whereas interchanged players from losing teams had a greater “end-spurt” than interchanged players from winning teams. The pacing strategies of interchanged players were higher than whole-game players, irrespective of playing position. The results of this study suggest that pacing strategies differ between interchanged and whole-game rugby league players. Furthermore, our results demonstrate a different pacing strategy between whole-game and interchanged players from winning and losing teams. These findings suggest that the physical preparation for rugby league matches, and recovery from these matches, should be individualized for whole-game and interchanged players.
A novel finding of the present study was the differences in pacing strategies between winning and losing whole-game players, as well as winning and losing interchanged players. As hypothesized, whole-game players from winning teams set a higher pacing strategy than the losing teams, with a greater total distance and low-speed distance covered across all quartiles of the match. Pacing in team sports is closely linked to recovery, and the intensity players believe they can withstand to complete the game without failure of any single physiological system (10). Indeed, during high-intensity intermittent team sport competition, players have been shown to reduce low-intensity activity to preserve the repeated high-intensity effort demands of match play (12). A key difference in the pacing strategies of the players from winning and losing teams in the present study was that the winning whole-game players recovered at a higher intensity, covering greater low-speed distances, than the losing whole-game players. Consistent with previous findings from soccer (22), more successful teams covered greater total distances throughout the entire match. These results suggest that the match outcome of whole-game players is closely related to their ability to maintain a higher playing intensity (11) over all quartiles of the match, and that this increased intensity is mediated through greater intensity during recovery periods. From a practical perspective, our findings suggest that successful teams set a pacing strategy designed to win the match, while the pacing strategies of unsuccessful teams are established based on “survival.”
The interchanged players from losing teams covered greater total distances and performed more low-speed activity during the final quartile of the match. These findings may indicate that the interchanged players from losing teams were interchanged at a time where the match could have potentially been salvaged. It has been demonstrated that defending is more physically demanding than attacking, as more low-speed distances are covered when defending because of decreased recovery periods (25). Given the findings of Sykes et al. (25), the present results may suggest that the losing interchanged players spend more time defending compared with the winning interchanged players, therefore covering greater low-speed distances. However, there were no meaningful differences found between interchanged players from winning and losing teams for the amount of high-speed running performed. These results are generally consistent with recent findings from elite rugby league match play, which have shown that the major differences between winning and losing teams are the recovery intensity between high-intensity bouts rather than the amount of high-speed running performed (11).
The interchanged losing players engaged in a greater frequency of repeated high-intensity effort bouts compared with interchanged winning players. Previous researchers have reported that the majority of tries scored or conceded in rugby league occur in close proximity to a repeated high-intensity effort bout (3), suggesting that the ability (or inability) to perform repeated efforts may prove critical to the outcome of the game (11,14). A possible explanation for differences in repeated high-intensity effort activity between winning and losing teams may lie in the nature of these efforts, with most repeated high-intensity effort bouts reported to occur while either attacking or defending the try-line (18). It is likely that players from losing teams performed the majority of repeated high-intensity effort bouts defending their try-line to avoid conceding a try. These results highlight the importance of repeated high-intensity effort ability to playing performance.
An interesting finding from the present study was the differences in pacing strategies between whole-game and interchanged hookers and second rowers. The interchanged hookers covered greater total distances and low-speed distances in quartiles 1 and 8, whereas the whole-game hookers covered greater total distances in quartiles 5 and 6. Although the whole-game hookers covered greater total distances during quartile 5 (at the start of the second half), the interchanged hookers covered greater high-speed distances. These results demonstrate the differences in pacing strategies employed by whole-game and interchanged players. Both whole-game and interchanged second rowers set a similar pacing strategy to the hookers. The interchanged second rowers exhibited an “end-spurt” during the last quartile of match play demonstrating their importance in the final stages of the match. In contrast to these findings, the interchanged locks covered greater total distance during quartiles 1 and 2, whereas whole-game locks covered greater total distances during quartiles 3 and 8. Whole-game locks also performed greater amounts of high-speed running than interchanged locks. These results are in contrast to our findings in whole-game and interchanged hookers and second rowers, as well as the findings of others (27) comparing whole-game and interchanged players. The present findings raise an interesting question about the pacing strategy adopted by interchanged locks and also their underlying purpose during a match. It is likely that differences in tactical strategies could influence the different pacing strategies of hookers, second rowers, and locks. Indeed, although some of the pacing strategy employed by players will be self-paced, other components of match play are externally paced by the opposition. For example, given the fatiguing nature of repeated high-intensity effort exercise (17), attacking players often target weaker defensive players to increase their tackle count and frequency of repeated high-intensity effort bouts. It is likely that the higher playing intensity and greater frequency of repeated high-intensity effort bouts in interchanged players reflect at least in part a tactical strategy on behalf of these players. Future studies linking the physical demands of competition with game-specific events may provide further insight into the different pacing strategies of different playing positions.
The present study is the first to investigate the difference between whole-game and interchanged players, winning and losing teams, and similar positional groups. Whole-game players from winning teams set a higher pacing strategy than whole-game players from losing teams, whereas interchanged players from losing teams had a greater “end-spurt” than interchanged players from winning teams. Our results also demonstrate that interchanged second rowers and hookers set a higher pacing strategy in the initial periods of play compared with whole-game second rowers and hookers. The interchanged second rowers and hookers conserve their physical exertion for later stages of the match as evidenced by an “end-spurt” in the final quartile of play. In contrast, the whole-game locks set a higher pacing strategy than interchanged locks, emphasizing the importance of considering playing position when implementing interchanges. Future research should consider the presence of different pacing strategies between whole-game and interchanged players during attack and defense, and the effect that changes in the evolving scoreline has on the pacing strategies of rugby league players.
This study described the unique pacing strategies of whole-game and interchanged rugby league players, as well as differences in pacing strategies between winning and losing teams. The interchanged hookers and second rowers both employed a pacing strategy that preserved energy expenditure to perform at a higher intensity in the final quartile of play. Knowledge of the likely duration of an interchange bout may assist the pacing strategies of these players. In contrast, the whole-game locks set a higher pacing strategy than interchanged locks. These results suggest that the interchanged locks may require substitution at an earlier stage of the match to allow them to make an “impact” on the match outcome.
In the final stages of match play, the interchanged players from losing teams set a higher pacing strategy than the interchanged players from winning teams, covering greater total distances. Although the influence of team strategies should be taken into account, these results suggest that these players may benefit from lower-intensity training sessions following matches, rather than greater amounts of physical conditioning, to facilitate recovery for the next match. These results emphasize the importance of interchanged players and also highlight the importance of individualized training following different match outcomes.
Given the importance of the brain in the regulation of exercise intensity (10), these findings have important implications for the applied sport scientist and strength and conditioning coach. Although training programs may result in some obvious physical transformations (e.g., increased lean body mass), it is important to recognize that strength and conditioning practitioners should consider the brain and it's regulatory role on exercise performance when training individual physiological systems. Appropriately designed strength and conditioning programs also provide an opportunity to “train the brain” to tolerate the fatigue and intensity of competition. Exposing the brain to physically intense training on a regular basis improves the body's ability to cope with fatigue (4). Equally, by restricting athletes from performing physically intense training, it is likely that they will set a lower pacing strategy during match play (4). Performing physically intense training (coupled with adequate recovery periods) on a regular basis develops physical qualities necessary for competition but is also likely to increase the mental durability of players.
1. Ansley L, Robson PJ, St Clair Gibson A, Noakes TD. Anticipatory pacing strategies during supramaximal exercise lasting longer than 30 s. Med Sci Sports Exerc 36: 309–314, 2004.
2. Aughey R. Australian football player work rate: Evidence of fatigue and pacing? Int J Sports Physiol Perform 5: 394–405, 2010.
3. Austin D, Gabbett T, Jenkins D. Repeated high-intensity exercise in a professional rugby league. J Strength Cond Res 25: 1898–1904, 2011.
4. Baron B, Moullan F, Deruelle F, Noakes TD. The role of emotions on pacing strategies and performance in middle and long duration sport events. Br J Sports Med 45: 511–517, 2011.
5. Billaut F, Bishop D, Scharez S, Noakes T. Influence of knowledge of sprint number on pacing during repeated-sprint exercise. Med Sci Sports Exerc 43: 665–672, 2011.
6. Castellano J, Casamichana D, Calleja-Gonzalez J, San Roman J, Ostojic SM. Reliability and accuracy of 10 Hz GPS devices for short-distance exercise. J Sports Sci Med 10: 233–234, 2011.
7. Cohen J. Statistical Power Analysis for the Behavioural Sciences (2nd ed.). New York, NY: Academic Press, 1988.
8. Duffield R, Coutts A, Quinn J. Core temperature responses and match running performance during intermittent-sprint exercise competition in warm condition. J Strength Cond Res 23: 1238–1244, 2009.
9. Dwyer D, Gabbett T. Global positioning system
data analysis: Velocity ranges and a new definition of sprinting for field sport athletes. J Strength Cond Res 26: 818–824, 2012.
10. Edwards A, Noakes T. Dehydration: Cause of fatigue or sign of pacing in elite soccer? Sports Med 39: 1–13, 2009.
11. Gabbett T. Influence of the opposing team on the physical demands of elite rugby league match play. J Strength Cond Res 27: 1629–1635, 2013.
12. Gabbett T. Influence of playing standard on the physical demands of professional rugby league. J Sports Sci 31: 1125–1138, 2013.
13. Gabbett T, Jenkins D, Abernethy B. Physical collisions and injury during professional rugby league skills training. J Sci Med Sport 13: 578–583, 2010.
14. Gabbett T, Jenkins D, Abernethy B. Physical demands of professional rugby league training and competition using microtechnology. J Sci Med Sport 15: 80–86, 2012.
15. Gabbett T, Wiig H, Spencer H. Repeated high-intensity running and sprinting in elite women's soccer competition. Int J Sports Physiol Perform 8: 130–138, 2013.
16. Hopkins WG. A spreadsheet for analysis of straightforward controlled trials. Sportscience 7:sportsci.org/jour/03/wghtrials.htm, 2003.
17. Johnston RD, Gabbett TJ. Repeated-sprint and effort ability in rugby league players. J Strength Cond Res 25: 2789–2795, 2011.
18. King T, Jenkins D, Gabbett T. A time–motion analysis of professional rugby league match-play. J Sports Sci 27: 213–219, 2009.
19. Meir R, Newton R, Curtis E, Fardell M, Butler B. Physical fitness qualities of professional rugby league players: Determination of positional differences. J Strength Cond Res 15: 450–458, 2001.
20. Mohr M, Krustrup P, Bangsbo J. Match performance of high-standard soccer players with special reference to development of fatigue. J Sports Sci 21: 519–528, 2003.
21. Petersen C, Pyne D, Portus M, Dawson B. Validity and reliability of GPS units to monitor cricket-specific movement patterns. Int J Sports Physiol Perform 4: 381–393, 2009.
22. Rampinini E, Coutts AJ, Castagna C, Sassi R, Impellizzeri FM. Variation in top level soccer match performance. Int J Sports Med 28: 1018–1024, 2007.
23. Sirotic A, Coutts A, Knowles H, Catterick C. A comparison of match demands
between elite and semi-elite rugby league competition. J Sports Sci 27: 203–211, 2009.
24. Spencer M, Bishop D, Dawson B, Goodman C. Physiological and metabolic responses of repeated-sprint activities: Specific to field-based team sports
. Sports Med 35: 1025–1044, 2005.
25. Sykes D, Twist C, Hall S, Lamb K, Nicholas C. Semi-automated time-motion analysis of senior elite rugby league. Int J Perform Anal Sport 9: 47–59, 2009.
26. Twist C, Waldron M, Highton J, Burt D, Daniels M. Neuromuscular, biochemical and perceptual post-match fatigue in professional rugby league forwards and backs. J Sports Sci 30: 359–367, 2011.
27. Waldron M, Highton J, Daniels M, Twist C. Preliminary evidence of transient fatigue and pacing during interchanges in rugby league. Int J Sports Physiol Perform 8: 157–164, 2013.