Over the past decade, computer-based time motion analysis and global positioning system (GPS) devices have become increasingly popular in an attempt to quantify player movements during competitive sports. Global positioning system technology obtains real-time objective data on the physical work of a player during sport competition. These data can be used by a variety of sports professionals to enhance the practices and conditioning sessions of their respective sports. For example, coaches can use this information to assess their players' performance (i.e., speed profiles and work to rest ratios), whereas strength and conditioning coaches can make modifications to the design of workouts based on these movement patterns. In addition, health care professionals such as certified athletic trainers can monitor an athlete's safety by quantifying the physiological demands during activity.
The GPS receivers that are worn by players during training and competition use information from the earth's orbiting satellites to calculate and store data (14). When worn during competition, the GPS unit records player information such as the distance covered (DC), velocity, accelerations, and heart rate (HR). The accuracy of this device has been shown to be between 1 and 5% based on reliability and validity of speed and accuracy (3,7,17).
Global positioning system technology has previously been used to quantify the physical demands of sports such as rugby (4,16), soccer (2,11,13), Australian rules football (6,17), and tennis (8). These data have produced information on the specific game demands and associated physiological responses endured by athletes. Furthermore, these data have been used to identify performance characteristics to mimic specific demands of sport through training program specificity. More specifically, this information can be used to acknowledge the differences in the physical demands of different playing positions, different skill levels (i.e., starters [S] vs. backups), and different drills. For example, the studies conducted by Cunniffe et al. (4), Wisbey et al. (17), and Burgess et al. (2) all showed differences in the movement profiles of different playing positions within the same training session. Similarly, Sirotic et al. (16) found that elite rugby players performed more high-intensity activity than did semielite players who participated in the same game format. Associations between these physical characteristics and physiological variables such as HR and core body temperature have also been studied (1,4-6,9-11,13).
American football is an intermittent high-intensity sport characterized by quick, intense bursts of work and short rest periods between plays (12). These high-intensity movements coupled with stressful environmental conditions, hyperthermia, dehydration, and other stresses that are inherent of preseason football practice can produce considerable physical and physiological strain that can result in a diminished exercise performance. Iosia and Bishop (12) reported that the average duration of a football play during collegiate competition was 5.23 seconds, whereas the average duration of rest between plays was 36.1 seconds. Furthermore, it was reported that the average duration of a pass play was significantly longer than that of a run play (12). Rhea et al. (15) reported similar findings when identifying play characteristics of high school, collegiate, and professional football games.
The purpose of this study was to identify the physical demands, as recorded by a GPS device, imposed on Division I football players during preseason training in hot conditions. Furthermore, comparisons of these demands were made between the players of different playing positions (nonlinemen [NL] vs. linemen [L]) and playing statuses (S vs. nonstarters [NS]). It was hypothesized that NL would cover more distance and obtain higher velocities than L would; however, L would have higher HRs because of the isometric nature of their drills. It was also hypothesized that S would cover more distance, obtain higher velocities, and have higher HRs than NS would because of the higher volume of repetitions that S participate in during practice.
Experimental Approach to the Problem
Although the movement patterns of other sports have been extensively studied via GPS devices, no current literature exists on the physical demands of American football players during preseason practices. The demands of this sport have previously been difficult to quantify because of the spontaneous and unpredictable movement patterns of the sport. An observational field study was used to acquire data regarding the physical demands imposed during a preseason football practice via a GPS device worn by the subjects. Because there is a lack of current knowledge with respect to American football, these data will provide a preliminary approach to be able to quantify movement patterns and work output of American football players.
Forty-nine male football players participated in this study. All the players were members of the same National Collegiate Athletic Association (NCAA) Division I, Bowl Championship Series (BCS) conference football team in the Southeast United States. In addition, all the participants were heat acclimatized and had engaged in conditioning sessions in the weeks leading up to data collection. These conditioning sessions involved lifting in the team weight room and outdoor conditioning drills. The participants included all playing positions except for place kickers and punters because of the limited physical activity endured by these positions during practice. The participants were divided into 2 groups: L and NL, and S and NS. Linemen included the positions of defensive end, defensive tackle, offensive tackle, offensive guard, center, and tight end. Nonlinemen included the positions of cornerback, free safety, strong safety, outside linebacker, middle linebacker, wide receiver, quarterback, full back, and running back. These groups were determined primarily on the types of drills they engaged in during practice, and body mass differences among these groups. Demographic information by group can be seen in Table 1. Inclusion criteria included being a member of the football team while participating in the first 8 days of preseason practices. All the participants completed a medical history questionnaire and a preseason risk assessment questionnaire before data collection and were cleared to participate by the team physician. All the participants read and signed an informed consent before the commencement of this study and the University's Institutional Review Board approved the study.
Each of the 49 subjects' data were recorded for 1 practice session during the first 8 days of preseason football practice on the University's campus (days 1–4) and off-campus at a training facility (days 5–8). On a few occasions, some subjects' data were recorded on multiple days because of equipment malfunctions (i.e., HR monitor did not record, GPS device turned off, etc.); however, each of the 49 subjects' data was only used once for statistical analysis. Data were collected over 8 consecutive days of outdoor football practice in the beginning of August in hot conditions (Table 2). Each of the 8 days involved single practice sessions, except for day 6, which involved 2 practice sessions. Each practice session began with a self-directed warm-up (i.e., jogging) and team stretch (static stretches), was followed by a segment of “position drills,” and ended with a segment of “team drills.” Position drills involved players being grouped based on their playing position and performing position specific activities within those groups. For example, offensive L would engage in “stance and start” drills in which they would practice “firing out” from their stance or wide receivers would practice running routes while catching passes. Team drills occurred in a similar fashion, however, and involved combining similar position groups and performing drills against one another (i.e., offensive L vs. defensive L would compete in blocking drills, whereas wide receivers vs. defensive backs would compete in passing drills). In addition, the practice session on day 5 and the evening practice on day 6 ended with a team scrimmage. Each segment of practice immediately proceeded into the next, except on days 3, 5, and 8 on which a short break was given between the position drills and team drills.
Global positioning system: Before each practice session, the participants were provided a GPS device (MinimaxX 2.5; Catapult innovations, Melbourne, Australia). The GPS unit was turned on and placed inside the pocket of a protective garment (Figure 1). The unit was positioned on the upper back just above the shoulder blades, and the garment was worn underneath the players' regular protective equipment. An HR monitor (Polar Electro Inc., Lake Success, NY, USA) was placed inside the chest strap of the garment, and the HR was recorded synchronously with the GPS data. The GPS and HR data were continuously recorded for the duration of each practice session. The GPS device recorded data at 5 Hz and stored physical components of player movements such as DC, velocity, and HR. After each practice session, the participants returned the GPS device, and the data were uploaded to a personal computer and stored for analysis via the manufacturer's software (LoganPlus 4.4.0, Catapult Innovations, Melbourne, Australia).
Core body temperature: Body temperature (TGI) was recorded via ingestible thermistor (HQ Inc. Palmetto, FL, USA). Sensors were distributed approximately 6 hours before the beginning of practice to ensure passage into the small intestine. Sensors were checked before the start of each practice to ensure accurate readings. If an accurate reading was not displayed, that participant's data were not included. During practice, researchers were stationed at each group (i.e., L, receivers, etc.) and recorded TGI approximately every 5 minutes via a handheld recorder (HQ). Because this study was a part of a larger study, TGI was used here primarily as a descriptive factor, because statistical comparison of this variable is beyond the scope of this article.
Hydration status: Before and immediately after each practice, body mass was recorded on a calibrated scale (Tanita Corp., Tokyo, Japan), with the subjects wearing only shorts to determine percent body mass loss (%BML). Because this was an observational study, fluid consumption was not controlled for. However, it is important to note that throughout the entirety of all practice sessions, the players were given ample opportunity to hydrate and were permitted to drink ad libitum via water bottles and coolers stationed with each group.
After the GPS data were uploaded after practice, each subject's data were saved to an individual file. Analysis included practice time (minutes), DC (meters), velocity (kilometers per hour), maximal HR (HRmax), and mean HR (HRavg). Specific measurements of velocity were analyzed as a percentage of total DC (% DC) and a percentage of total time (% T) spent in specific velocity zones: zone 1 (standing: 0–1.0 km·h−1), zone 2 (walking: 1.1–6.0 km·h−1), zone 3 (jogging: 6.1–12.0 km·h−1), zone 4 (running: 12.1–16.0 km·h−1), and zone 5 (sprinting: >16 km·h−1). We chose these velocity ranges based on similar studies (2,4,6,11,13,16,17). These variables were statistically analyzed for total practice time, the segment of practice that corresponded to position drills, and the segment of practice that corresponded to team drills. Although the total practice time included stretching, warm-ups, breaks, position drills, and team drills, we chose to only partition out position drills and team drills for further analysis. These segments were individually viewed by setting time specific periods to the entire practice data file.
Descriptive data (mean ± SD) were calculated for all variables. The participants were grouped into L vs. NL and S vs. NS, for statistical analysis. Independent samples T-tests were used to determine the differences in HR, DC, V, and % BML between groups. These comparisons were made for total practice time, the time that corresponded to position drills, and the time that corresponded to team drills. Statistical significance was set at p ≤ 0.05. All statistical analyses were performed using SPSS version 18.0.
Linemen Vs. Nonlinemen
Total DC during practice was significantly higher in NL than in L (3,532 ± 943 vs. 2,573 ± 489 m; p < 0.001). In addition, NL covered greater distances during position drills (1,673 ± 420 vs. 1,231 ± 189 m; p < 0.001) and team drills (1,262 ± 626 vs. 915 ± 383; p = 0.026), respectively compared with that of L (Figure 2A).
Starters Vs. Nonstarters
Total DC was significantly higher during team drills in S vs. that in NS (1,222 ± 508 vs. 850 ± 525 m; p = 0.018); however, no significant differences were found during position drills (1,432 ± 423 vs. 1,475 ± 336 m; p = 0.714) or total practice time (3,116 ± 925 vs. 2,915 ± 817 m; p = 0.448) between S and NS, respectively (Figure 2B).
Linemen Vs. Nonlinemen
The NL spent a higher % DC in zone 4 for practice drills (p = 0.002), team drills (p < 0.001), and total practice time (p < 0.001) and in zone 5 for position drills (p < 0.001), team drills (p < 0.001), and total practice time (p < 0.001) than did NL. Conversely, L spent a higher % DC in zone 1 for position drills (p = 0.001) and total practice time (p = 0.008) and in zone 2 for practice drills (p < 0.001), team drills (p = 0.005), and total practice time (p < 0.001). No significant differences were found between L and NL for zone 3 (Table 3).
Similar to the % DC, NL spent a greater % T in zone 4 for practice drills (p = 0.002), team drills (p = 0.002) and total practice time (p = 0.001) and in zone 5 for position drills (p < 0.001), team drills (p < 0.001), and total practice time (p < 0.001). In addition, NL spent a greater % T in zone 3 during position drills (p = 0.019) and total practice time (p = 0.018). No significant differences were observed in % T in zones 1 or 2 between L and NL during any segment of practice, except for total practice time in which L spent more time in zone 1 than NL did (p = 0.034) (Table 3).
Starters Vs. Nonstarters
No significant differences were observed in % DC between S and NS for any of the velocity zones during any segment of practice (Table 4). However, significant differences occurred between S and NS in % T spent in velocity zones 1 and 2 for team drills (p = 0.008, p = 0.005) and total practice time (p = 0.031, p = 0.019), respectively. No other differences were noted between S and NS, except during position drills in which S spent more time in zone 4 than NS did (p = 0.046) (Table 4).
Linemen Vs. Nonlinemen
The NL reached a higher HRmax than L did during position drills (201 ± 9 vs. 194 ± 11 b·min−1; p = 0.025) and total practice time (203 ± 8 vs. 197 ± 9 b·min−1; p = 0.013); however, no significant differences occurred in HRmax during team drills (189 ± 12 vs. 192 ± 12; p = 0.343) between L and NL, respectively (Figure 3A). In addition, no significant differences in HRavg were seen between L and NL for position drills (143 ± 8 vs. 142 ± 11; p = 0.715), team drills (134 ± 11 vs. 135 ± 12; p = 0.698), or total practice time (136 ± 7 vs. 135 ± 11; p = 0.580) (Figure 3B).
Starters Vs. Nonstarters
No significant differences were observed in the HRmax during position drills (197 ± 12 vs. 198 ± 8 b·min−1; p = 0.885), team drills (192 ± 13 vs. 189 ± 10 b·min−1; p = 0.385) or total practice time (200 ± 10 vs. 199 ± 8 b·min−1; p = 0.771) between S and NS, respectively (Figure 3C). Starters had a higher HRavg than NS did during team drills (138 ± 11 vs. 129 ± 11 b·min−1; p = 0.008); however, no significant differences were observed between S and NS for the HRavg during position drills (143 ± 9 vs. 142 ± 11 b·min−1; p = 0.747) or total practice time (138 ± 8 vs. 133 ± 11; p = 0.073) (Figure 3D).
All the players had a similar %BML during the duration of each practice (L = 1.6 ± 0.8%BML, NL = 1.8 ± 0.9%BML; p = 0.590). Similarly, no significant difference in %BML was observed between S and NS (1.8 ± 0.9%BML and 1.6 ± 0.7%BML, respectively; p = 0.656). This relatively mild state of dehydration is caused by the ad libitum drinking that was permitted throughout each practice session. No significant differences were observed for Tgi between L and NL (101.00 ± 0.56° F vs. 101.11 ± 0.65° F; p = 0.662) or S and NS (101.05 ± 0.63 vs. 100.88 ± 0.53; p = 0.523) for the duration of the practice.
The purpose of this study was to identify the physical demands imposed on NCAA Division I football players during preseason training in hot conditions. Furthermore, it was of primary interest to compare these demands between players of different positions and playing statuses. The main findings were that NL covered more distance and obtained higher velocities than did L while also reaching a greater HRmax. These differences were not as apparent when comparing S vs. NS, with the most notable differences occurring during team drills.
The subjects covered 2,938 ± 794 m during practice. This distance is considerably less than that reported in other sports, such as rugby (6,953 m) (4) and Australian rules football (10,100–12,000 m) (2,17), during match play. In addition, these distances are likely much greater than those covered during a game of football. However, similar to our findings, these sports exhibit differences between playing position and the total amount of DC. The NL covered significantly more distance than L did during all practice segments (Figure 3A). These results are consistent with those of Burgess et al. (2) and Wisbey et al. (17) who showed that Australian rules football defenders cover less distance than do other positions, while also performing at slower speeds, also consistent with our findings.
The difference in DC between L and NL in our study is likely because of the difference in the requirements of specific positions. For example, the primary responsibility of L is to block, which almost always occurs within a few yards of the line of scrimmage. On the other hand, NL (such as wide receivers, cornerbacks, and running backs) are often required to cover far more distance during both games and practices, likewise because of the specific demands of their position. In this case, it is not uncommon for these players to run 30 yd to catch a pass or to cover a receiver. Therefore, the majority of practice drills designated to players of different positions tend to mimic their respective game responsibilities, ultimately requiring various distances to be covered. This notion is further supported by the results exhibited by Rhea et al. (15) showing that NL such as wide receivers and defensive backs most often jogged back to the line of scrimmage, often 20–40 yd away, after a play, whereas lineman walked shorter distances during recovery.
The difference in DC between S and NS was not as dramatic as that found between L and NL, with the only significant difference occurring during the team drills (Figure 3B). However, these results are not surprising given the nature of team drills. More specifically, it is common for those players who are considered to be S (i.e., based on the playing status of the previous season) to participate in the majority of the repetitions that occur during these drills. Often times, those “back-up” players only participate in the drills when the S need a break. During position drills, all the players, regardless of playing status, rotate through the drills an equal number of times. Thus there are minimal differences between S and NS in DC during that part of practice (Table 4, Figures 2 and 3).
Several studies have examined the difference in velocity obtained by players of different positions (2,4,17) and of different playing statuses (13,16). For example, Cunniffe et al. (4) found that rugby backs participated in a greater amount of higher intensity work when compared with forwards. Similarly, Burgess et al. (2) showed that Australian rules football backs performed at slower speeds than those of midfielders and forwards. Similar to these findings, this study showed considerable differences in the speed at which L vs. NL performed (Table 3). Of particular interest, L and NL both obtained similar distances via jogging (zone 3), accounting for approximately 30% of their total distance. Therefore, of the remaining 70% of the DC during practice, NL covered significantly more distance while running and sprinting (zones 4 and 5), whereas L covered significantly more distance at slower speeds (zones 1 and 2). The difference in % T spent in these velocity zones showed similar results as % DC, with the most drastic differences once again occurring in zones 4 and 5 (Table 3). Although the total time was shorter (83 vs. 142 minutes) and total distance was longer (6,953 vs. 2,938 m), Cunniffe et al. (4) showed similar results in rugby players regarding % T spent in each velocity zone (zone 1: 72 vs. 76%; zone 2: 19 vs. 18%; zone 3: 3 vs. 4%; zone 4: 4 vs. 1%; zone 5: 1 vs. 1%). The difference in total time and DC is most likely because our data were collected during practices, where players are often standing or walking between repetitions and drills, instead of constantly moving during a rugby match.
Previous studies have also examined the difference in velocity among players of different physical statuses (13,16), showcasing that elite players tend to perform at higher intensities than those of semielite players (16). Furthermore, Krustrup et al. found a positive correlation between O2max and high-intensity running performance among female soccer players (13). In this study, although fitness status was not directly measured for each subject, minimal significance was found in % D and % T spent in the velocity zones between S and NS (Table 4). Similar to the % D spent jogging (zone 3) with L vs. NL, S and NS also spent about 30% of their total distance at this speed. However, in contrast to L vs. NL, no differences occurred between S and NS in the other velocity zones during any part of the practice. Although % T spent in zone 1 was greater in NS vs. that in S during team drills and total practice time, % T spent in zone 2 was greater in S vs. that in NS. Furthermore, no other clinically significant differences occurred between these 2 groups. Therefore, because velocities were similar between S and NS, it appears that the effort put forth was similar between groups. Because each group worked equally hard, it seems as though playing status may be more determinant on the skill of a player, and not on effort.
Our subjects maintained an HRavg of 136 b·min−1 and obtained an HRmax of 200 b·min−1. Interestingly, our results showed quite diverse findings when comparing HR measurements of L vs. NL with those results of S vs. NS. More specifically, no significant differences were found for HRavg between L and NL in any of the practice segments. However, differences were observed for HRmax among these 2 position groups for both position drills and total practice time, because NL elicited a greater HRmax during both segments. Conversely, S and NS did not differ in HRmax in any of the practice segments, but S had a higher HRavg than NS did during team drills.
Although several studies have examined HR as a means of exercise intensity (4,5,11,13), albeit within different sporting events, the results are mixed as to whether HR is an indicator of performance level. For example, Krustrup et al. (13) showed no relationship between percent of HRmax and various performance measurements in female soccer players during match play. Conversely, Hill-Haas et al. (11) demonstrated that soccer games involving fewer players and on a smaller field elicit a greater HR response and a greater time spent at higher HRs than do those games with more players playing in a larger area. In addition, the increased HR found with the smaller-sided games was accompanied by higher lactate levels and rating of perceived exertion, implying a diminished performance.
To further support the idea that exercise intensity can influence performance, a study by Dellal et al. (5) found that soccer players participating in various intermittent running drills elicited different HR responses. Most notably, when completing the same running drill, percent HR reserve was significantly higher when the drill was coupled with an active recovery compared with a passive recovery. This notion may explain why HRmax was greater in NL compared with that in L. As previously mentioned, it has been shown that even when a play is over, NL are typically required to jog or even run back to the line of scrimmage to get ready for the next play (15). On the other hand, because L do not commonly travel far from the line of scrimmage during a play, they are able to simply walk back to the line of scrimmage and can even stand when waiting for their teammates to return to the huddle. Therefore, it can be viewed that L undergo more of a passive recovery in between plays, whereas NL undergo an active recovery. As a result, the HR of NL is likely elevated more so than that of L when returning to the line of scrimmage and also at the start of the next play.
The HRavg was similar between L and NL during all the practice segments. This result was somewhat unexpected given that HRmax was greater in NL during position drills and total practice time. However, although NL elicited a greater HRmax, covered more distance, and had shorter absolute rest (as denoted by spending less time standing and walking), L likely engaged in more anaerobic work compared with NL. Therefore, it may be true that the HRavg during the active portion of the drills (from the time the ball is snapped until the time the play is whistled dead) may actually be greater in L vs. that in NL. However, because the results presented in this article contain values irrespective of time between plays, this analysis could not be verified.
Our findings for the HR responses between S and NS are not surprising, especially when considering the difference in DC during team drills between these 2 groups. Starters covered more distance during the team drills portion of the practice, likely because S complete more repetitions than do NS. As expected, S also maintained a greater HRavg during this part of the practice, likely a direct result of completing more repetitions. In other words, during team drills, S spent more time engaged in drills, whereas NS spent more time standing. This is further supported by the fact that NS spent significantly more time in velocity zone 1 (standing) during this part of the practice (Table 2). Lastly, the lack of difference in HRmax between these 2 groups shows that when NS did get an opportunity to compete in drills, they did so at the same intensity as did S.
Data acquired by GPS devices can accurately assess specific components of athletic performance. This information can be used to assess an athlete's workouts, improve conditioning sessions to mimic specific sport demands, and enhance performance. More specifically, coaches can establish an individual player's baseline measurements by assessing the data recorded by the GPS and evaluate a player's improvement by recording GPS data at different time points during the season. Furthermore, strength and conditioning coaches can collect data on player demands during games to determine suitable exercises to mimic the volume and speed appropriate for conditioning sessions. Because there is no current literature on these activities for the sport of American football, this information is helpful for coaches and strength and conditioning personnel alike.
Future research should aim to associate performance data collected by the GPS with that of injury risk. More specifically, the amount of DC and velocities obtained can be correlated with injuries that are related to excessive volume and overuse injuries endured by these players. Furthermore, these performance measures can also be correlated to physiological data such as core body temperature, skin temperature, HR, and dehydration markers to assess the relationship of these variables to the prevalence of exertional heat illnesses, so athletes can be safely monitored during these stressful activities.
The authors would like to give special thanks to Ollie Jay and his research team for their assistance and efforts in this study.
1. Black, W and Roundy, W. Comparisons of size, strength, speed, and power in NCAA division 1-A football players. J Strength Cond Res
8: 80–85, 1994.
2. Burgess, DJ, Naughton, G, and Norton, KI. Profile of movement demands of national football players in Australia. J Sci Med Sport
9: 334–341, 2006.
3. Coutts, AJ and Duffield, R. Validity and reliability of GPS devices for measuring movement demands in team sports. J Sci Med Sport
13: 133–135, 2010.
4. Cunniffe, B, Proctor, W, Baker, JS, and Davies, B. An evaluation of the physiological demands of elite rugby union using global positioning system tracking software. J Strength Cond Res
23: 1195–1203, 2009.
5. Dellal, A, Chamari, K, and Pintus, A. Heart rate responses during small-sided games and short intermittent running training in elite soccer players: A comparative study. J Strength Cond Res
22: 1449–1457, 2008.
6. Duffield, R, Coutts, AJ, and Quinn, J. Core temperature responses and match running performance during intermittent-sprint exercise competition in warm conditions. J Strength Cond Res
23: 1238–1244, 2009.
7. Edgecomb, SJ and Norton, KI. Comparison of global positioning and computer-based tracking systems for measuring player movement distance during Australian Football. J Sci Med Sport
9: 25–32, 2006.
8. Fernandez, JF, Villanueva, AM, Garcia, BF, and Terrandos, N. Match activity and physiological responses during a junior female singles tennis tournament. Br J Sports Med
. 41: 711–716, 2007.
9. Fowkes-Godek, S, Godek, JJ, and Bartolozzi, AR. Thermal responses in football and cross-country athletes during their respective practices in a hot environment. J Athl Train
39: 235–240, 2004.
10. Fry, A and Kraemer, W. Physical performance characteristics of American collegiate football players. J Appl Sport Sci Res
5: 126–138, 1991.
11. Hill-Haas, SV, Dawson, BT, Coutts, AJ, and Rowsell, GJ. Physiological responses and time-motion characteristics of various small-sided soccer games in youth players. J Sport Sci
27: 1–8, 2009.
12. Iosia, MF and Bishop, PA. Analysis of exercise-to-rest ratios during division IA televised football competition. J Strength Cond Res
22: 332–340, 2008.
13. Krustrup, P, Mohr, M, Ellingsgaard, H, and Bangsbo, J. Physical demands during an elite female soccer game: Importance of training status. Med Sci Sports Exerc
37: 1242–1248, 2005.
14. Larsson, P. Global positioning system and sport-specific testing. Sports Med
33: 1093–1101, 2003.
15. Rhea, MR, Hunter, RL, and Hunter, TJ. Competition modeling of American football: Observational data and implications for high school, collegiate, and professional player conditioning. J Strength Cond Res
20: 58–61, 2006.
16. Sirotic, AC, Coutts, AJ, Knowles, H, and Catterick, C. A comparison of match demands between elite and semi-elite rugby league competition. J Sport Sci
27: 203–211, 2009.
17. Wisbey, B, Montgomery, PG, Pyne, DB, and Rattray, B. Quantifying movement demands of AFL football using GPS tracking
. J Sci Med Sport
13: 531–536, 2010.