Time-motion analysis is considered a useful tool to systematically determine the physiological demands of a sport. Specifically, time-motion analysis has been used to determine the time spent performing various activities with (2,6) or without (1,4,11) accompanying physiological measurements in a variety of sports including hockey, rugby, water polo, synchronized swimming, wheelchair basketball, and soccer. Time-motion analysis can be completed through observation (4); however, gold standard analysis involves video capture and recording of the sport being played in its competitive environment. The main dependent variables derived from time-motion analysis include segmenting a game into activities to estimate time spent performing those activities, determine the physiological stress imposed throughout the game, and to measure the frequency of the different types of movements performed. This allows sport scientists to best determine the “characteristics” of the game itself. This type of quantification is extremely important for sport enhancement and has important implications for the development of more appropriate training programs and monitoring to optimize physical preparations for competition.
Canoe polo is an emerging sport and is growing in popularity (2). The International Canoe Federation first sanctioned canoe polo in 1989 and the first World Championships were held in 1994. There were 23 countries competing in the men's division at the most recent International Canoe Federation World Championships in 2012. Canoe polo is a competitive ball sport played on water, in a defined pitch (35 m in length by 25 m wide), between 2 teams of 5 players, each in a kayak. In addition, each team has up to 3 players behind the goal line who can substitute in at any time. The object of the game is to score by throwing a water polo ball into the opponent's net (measuring 1.0 by 1.5 m) that is suspended 2 m above the water. A player acts as a goalie to defend the goal with their paddle. During game play, the ball is passed among the players by hand or with some use of the paddle or thrown out of reach from the player and then regaining possession; however, the player may only have the ball in their possession for a maximum of 5 seconds. The game is 20 minutes in length consisting of two 10-minute halves separated by 3 minutes and begins with a sprint start to gain possession of the ball at the center line (19).
The growth of canoe polo and the introduction of the world championships have led to an increased competitiveness in the game resulting in a need for a scientific evaluation and a measure of the physiological stress imposed on players during competition (2). The quantification (e.g., frequency and duration) of common movements (e.g., sprinting, slow paddling, resting and gliding, backward paddling, turning, and dribbling) and assessment of the physical demands (e.g., heart rates [HRs]) of canoe polo will have important implications for the development of more appropriate and specific training regimes, and allow coaches to better monitor programs through sport-specific fitness assessments to optimize physical preparations for competitions. Recently, Alves et al. (2) examined the anthropometric and physiological characteristics of international canoe polo athletes and evaluated preliminary time-motion analysis data (work-to-rest ratios and sprints times) during a single simulated game. Previous research in soccer has shown differences in time-motion analysis and physiological demands between simulated compared with official international matches (16). Therefore, the purpose of this investigation is to extend previous literature and to (a) evaluate the frequency and the time elite canoe polo players spend performing various activities during official international games; (b) measure individual HR response during these official games; and (c) assess the physiological capabilities of the players. It was hypothesized based on expert opinion of coaches and players that the game of canoe polo would consist primarily of intermittent forward paddling and contesting for position (e.g., challenging for position). In addition, because of the intermittent nature of canoe polo, it was hypothesized that elite canoe polo players would demonstrate high muscular endurance and a well-developed anaerobic and aerobic energy systems.
Experimental Approach to the Problem
A descriptive experimental design was selected to examine activity type and time spent performing each movement using time-motion analysis. Time-motion analysis involves quantification (frequency and duration) of common movements from video downloaded to specialized software (Dartfish 5.0 software, Switzerland). The movements were selected based on expert opinion and the dependent variables were frequency, total time, and % time in each movement category. The additional dependent variable of the HR was measured during the same games and used to indicate the intensity and cardiovascular demands of international competition. As well, standardized physiological fitness assessments were completed. The fitness assessment evaluated muscular strength, muscular endurance, flexibility, and aerobic and anaerobic power of each player.
A convenient sample of 8 male canoe polo athletes, all currently competing at the international level, participated in this study. Data collection for physical assessment occurred after the 2008 World Championships. The mean age, body weight, height, percent body fat, and years experience in the sport were 25 ± 1 years, 81.9 ± 10.9 kg, 1.82 ± 0.04 m, 9.9 ± 1.7%, and 6.2 ± 3.9 years, respectively. All the subjects were required to complete a physical activity readiness questionnaire. Each participant read and signed an informed consent document approved by a University Research Ethics Board.
Three preliminary (e.g., round robin format) games at a World Championships were filmed. A commercially available digital video camera was used from a raised vantage that allowed coverage of the entire pitch. Suunto T6 HR monitors (Finland) were worn by each player (n = 8) during the games, and the HR was recorded to memory and downloaded. Video analysis was performed using Dartfish 5.0 software (Switzerland). To distinguish the number of activity types, 7 different movement categories were selected through observation by the investigators and an expert judge who had extensive playing and international coaching experience in the sport. The following 7 movement categories and definitions were agreed upon:
- Sprinting: Propelling of a player in a kayak across the pitch quickly in a forward direction with a high stroke rate (maximal effort, visible splash).
- Slow/Moderate Forward Paddling: Propelling of a player in a kayak across the pitch in a forward direction with a slow-to-moderate stroke rate.
- Backwards Paddling: Propelling of a player in a kayak across the pitch in a backward direction.
- Turning: Propelling of a player to turn or rotate the kayak or change direction.
- Dribbling: Propelling of a player in a kayak across the pitch while handling the ball.
- Contesting: Defensive or offensive challenging for position while in the kayak (with or without the ball).
- Resting/Gliding: Resting in the kayak with no movement or any movement across the pitch without paddling.
To further assess the physical demands, these movements were separated into 2 intensity ratings; low-to-moderate intensity movements included slow and moderate forward paddling, backward paddling, turning, dribbling, and resting and gliding. High-intensity movements included sprinting and contesting.
Independent to the 7 movement categories, the amount of time spent in the starting sprint, offensive play, defensive play, and transition phases of the game and total number of shots and passes were recorded.
The videotapes were coded for the entire length of the game, excluding the 3-minute half time. Data were analyzed for the total game and separated into halves. The analysis included movements during whistle stops and movements during substitution. The same individual, who had experience playing canoe polo, completed all the time-motion analyses.
After the World Championships, each subject completed a battery of standardized anthropometric and fitness tests, which included height, weight, skinfolds, sit and reach flexibility, hand grip dynamometry, curl-ups, arm crank Wingate 30-second anaerobic power test, and arm crank peak oxygen uptake protocol (peak V[Combining Dot Above]O2). All fitness assessments used standardized published protocols as indicated in each section. During the time of the study, all the subjects were encouraged to undertake their normal training and diet. They were instructed to be adequately hydrated and not to eat 2 hours before the fitness assessment. One subject was injured during the last day of the World Championships and was unable to complete the physiological assessment (n = 7).
Height (meters) was measured with a wall-mounted device (Tanita, Arlington, IL, USA) to the nearest 0.01 m without shoes and with each subject standing erect against a wall. Body mass (kilograms) was measured to the nearest 0.1 kg using a balance beam scale (HealthoMeter, Bridgeview, IL, USA). Percent body fat was calculated by summing the mean of 2 skinfolds (Harpenden, United Kingdom) from 6 different anatomical sites (midtricep, chest, iliac crest, subscapular, abdominal, and front thigh) and entering these values in the following formula: % BF = sum of 6 × 0.097 + 3.64 (31).
Trunk forward flexion (i.e., sit and reach) using a Wells-Dillion (30) flexometer (Fitsystems Inc., Canada) and combined grip strength of the right and left hands using a hand dynamometer (Almedic Dynamometer: 100 kg, Japan) was measured as the highest score of 2 trials according to the Canadian Physical Activity and Lifestyle Appraisal protocol (10). Abdominal muscular endurance was assessed using a modified curl-up protocol that controlled tempo at a rate of 25 curl-ups per minute using a standardized protocol as previously reported by our laboratory (24).
Anaerobic power was determined during a 30-second Wingate test (18) at maximal effort on an arm crank ergometer (Monark, Model 881, Sweden) modified to accommodate calibrated weights to provide resistance to the friction belt around a flywheel. Frictional resistance for the ergometer wheel was set as 0.065 kg·kg−1 of the body weight (6). The ergometer was secured to a table 76 cm in height. During the test, the seat height was adjusted to ensure that the joint center of rotation for the shoulder was at the same height as the axis of rotation for the crank arm. The distance between the axis of the shoulder and crank was adjusted to ensure full arm extension. The participants cranked the ergometer and were allowed to move their whole upper body to ensure maximal performance during the test. Each subject was allocated a 5-minute warm-up on the arm crank ergometer at 60 rpm with 1 kg of resistance (∼25 W). Five seconds before data collection, they were instructed to increase the crank rate to a maximum to ensure the highest power and force production at the beginning of the test and to continue cranking at a maximal rate for the duration of the 30-second test. The subjects were consistently encouraged verbally throughout the test. Peak power output was averaged for the highest 5-second interval, and average power throughout the 30-second test was determined by videotaping the test and counting the number of revolutions over the appropriate time interval. Power output was determined from the following formula:
A continuous, incremental arm crank protocol (13,22) using a calibrated metabolic measurement system (True One 2400, ParvoMedics, UT, USA) was used to determine peak oxygen consumption (peak V[Combining Dot Above]O2). The arm crank protocol consisted of a crank rate of 70 rpm, initial loading of 35 W, and subsequent increases of 35 W every 2 minutes until volitional exhaustion (13,22). The HR was recorded continuously during the test using an HR monitor (Polar T31, Finland). The main criterion for achieving a peak V[Combining Dot Above]O2 was a plateau in the oxygen consumption (a change of <100 ml·min−1) with an increase in power output accompanied by either a respiratory exchange ratio of >1.10 or an age predicted or known maximum HR during arm cranking, or volitional exhaustion at cessation of exercise (3). The ventilatory threshold (VT) was determined as the point at which CO2 production and minute ventilation (VE) deviated from linearity as compared with the rate of rise in oxygen uptake as the work rate was increased (29). The VT was identified on a respiratory gas exchange record by one the investigators and agreement was reached with a second independent observer. Test-retest reliability has been previously established for this latter protocol (intraclass correlation coefficient = 0.97 for absolute peak V[Combining Dot Above]O2 and 0.91 for relative peak V[Combining Dot Above]O2) (13).
All the results are expressed as mean ± SD. To assess the reliability, the movement patterns of 8 players were analyzed by an investigator for one-half of a game on 2 separate occasions (intrarater) and by an independent investigator (interrater). The typical error of measurement (TE; within subject SD) was calculated from the trials for duration and relative percentage variables as described by Hopkins (17). In addition, the differences and the relationship of the movement times between the trials were assessed with a paired t-test and a Pearson correlation coefficient, respectively. Statistical significance was set at p ≤ 0.05.
Test-retest reliability was assessed on separate days (∼7 days apart). The intrarater and interrater TEs for total movement times of the video analysis were 5.7 and 2.1 seconds, respectively. The intrarater TE for movement time ranged from 1.9 to 8.0 seconds and 0.3–1.0% for the relative percentages for each movement category (sprinting: 2.4 seconds and 0.4%; backward paddling: 2.0 seconds and 0.3%; contesting: 5.0 seconds and 0.7%; slow and moderate forward paddling: 2.4 seconds and 0.7%; resting and gliding: 8.0 seconds and 1.0%; turning: 1.9 seconds and 0.3%; and dribbling 2.0 seconds and 0.3%). The interrater TE for movement time ranged from 2.6 to 12.7 seconds and 0.3–2.9% (sprinting: 3.8 seconds and 0.7%; backward paddling: 2.6 seconds and 2.0%; contesting: 10.8 seconds and 1.5%; slow and moderate forward paddling: 5.0 seconds and 2.9%; resting and gliding: 12.7 seconds and 1.0%; turning: 5.5 seconds and 0.3%; and dribbling 5.9 seconds and 0.3%). Furthermore, there was no significant difference (p > 0.05) between trials for any movement time (p > 0.05), and the Pearson correlation coefficients ranged from r = 0.88 to 0.99 for intrarater and r = 0.74–0.98 for interrater assessments of all movements.
Time-motion analysis of 3 games indicated that the players spent the majority of their time forward paddling, contesting, and resting and gliding, as shown in Table 1. In addition, the majority of the time was spent in low-to-moderate (70 ± 4%) intensity movements, whereas 30 ± 3% was spent performing high-intensity activities. Fifty percent (±15) of the game time was spent in offense, 30 ± 11% in defense, 19 ± 5% in transition, and 1 ± 0.2% performing the sprint start in the first half of the games and 29 ± 12% in offense, 51 ± 8% in defense, 20 ± 6% in transition activities, and 1.0 ± 0.2% of the game performing the sprint start in the second half of the games. The total number of whistle stops and time spent for each whistle stop in the first and second halves was 10 ± 2 stops, 14 ± 10 seconds, and 18 ± 6 stops, 19 ± 19 seconds, respectively. The number of passes and shots were 45 ± 22 passes and 4 ± 2 shots in the first half and 23 ± 10 passes and 4 ± 1 shots in the second half of the games.
Heart Rate Response
The mean HRs during the first half and second half of the games were 157 ± 16 and 159 ± 14 b·min−1. Sixty-nine (±22)% of the first half and 68 ± 29% of the second half of the games were played at a mean HR intensity that was above the HR that occurred at VT.
The individual fitness scores for each player is shown in Table 2. The VT occurred at a V[Combining Dot Above]O2 of 2.2 ± 0.3 L·min−1 and was 69.1 ± 6.8% of peak V[Combining Dot Above]O2. Curls-ups were 31 ± 15 repetitions.
Canoe polo was initially sanctioned in 1989 by the International Canoe Federation, and the first World Championships were held in 1994. Currently, there has only been 1 study examining the demands of canoe polo in a simulated game; however, there has been no systematic analysis of specific movements in official international canoe polo games. This study has quantified a range of physical and technical variables to examine the physical demands of canoe polo at the elite level. This information provides a foundation for development of more specific fitness assessments and training programs.
The major purpose of this article was to identify the amount of time spent during various movements in canoe polo. This information has implications for more specific training and development of sport-specific fitness assessments. The time-motion analysis data clearly indicate that international level canoe polo consists of high-intensity, intermittent exercise imposing both anaerobic and aerobic demands. The major findings of this study demonstrated that the predominant movements were slow-to-moderate forward paddling (29%), contesting for position (28%), and resting and gliding (27%), while sprinting, backward paddling, turning, and dribbling contributed to the remaining 16%. Sprinting and contesting are associated with higher intensity anaerobic work and were interspersed with less intensive activities such as resting and gliding. The proportion of most of these movements was similar between the first and second halves of the games, although there was less resting and gliding (25 vs. 28%) and more forward paddling (30 vs. 28%) in the first half compared with that in the second half. With respect to time, the players rested and glided 30 seconds more and forward paddled 2 seconds less in the second half of the games. These differences are relatively small and may translate into 1 less set play in the second half compared with the first half. This is consistent with previous literature in other team sports that commonly report similar percent differences between first and second halves (20,21,25). However, in the case of a soccer match where halves are 45 minutes, a 3% difference in high-intensity forward running (25) translates into a 1-minute and 21-second difference between halves. The sport most related to canoe polo is water polo and descriptive information for intensity of swimming per quarter played revealed a 3% decrease in moderate intensity movement and a 5.5% decrease in high-intensity movement from the first to fourth quarter (28) and has been implicated to be because of accumulated fatigue. As a result, it is possible that players may be less apt to make a high-intensity move to improve position later in a game because of fatigue. In addition, understanding fatigue during competition has important implications for tactical and strategic changes during a competition and may have an impact on the outcome of a game.
In addition, the amount of time spent in the offensive and defensive zone varied depending on the competition and competitiveness of the team being observed, the style and strategy used in the game and the interaction of these factors in consideration of the opposing teams during game play. Interestingly, the amount of time spent in the offense and defense zones between the first and second halves demonstrated a greater amount of time spent in the offense zone in the first half and subsequently a greater amount of time spent in the defensive zone in the second half of the games. These data are supported by the difference in the number of passes made in the first and second halves (45 vs. 23); however, there were equal numbers of shots in both halves of the 3 games analyzed. This data may further suggest accumulated fatigue during a game or a required change in tactical strategy between the first and second halves.
As indicated by the time-motion analysis, canoe polo is an intermittent sport and providing HR responses during game play provides a valid and useful global measure of physiological strain (12). The high HR response in all the players across each game suggests that elite canoe polo game play is associated with a very high level of physical exertion. Alves et al. (2) found a mean HR of 146 ± 11 b·min−1 during a simulated game; however, in this study during official games at the world championships, the average HR was approximately 12 b·min−1 higher. The intensity was at a level that elicited a HR that exceeded their VT HR for a considerable portion of the game (69 ± 20% of the game). Bloxham et al. (6) examined the HR response during an international wheelchair basketball game in paraplegic athletes and found that 24% of the first half and only 18% of the second half was played above the HR that corresponded to the VT. Lythe and Kilding (20) examined elite field hockey players and showed that players were >75% of the HRmax for 90% of the match. Similarly, water polo players achieved 86–95% of the HRmax during game play (27). Interestingly, in this study, all the players exceeded their peak HR achieved the arm crank V[Combining Dot Above]O2 protocol. This finding can be explained by the idea that on-water paddling involves a greater amount of active lower body musculature compared with that exercise on an arm crank ergometer (7,9,13). As a result, it is possible that the greater lower body muscle involvement during on-water paddling elicited a higher peak HR response and previous research comparing the arm crank ergometer to kayak ergometry has demonstrated a significantly higher peak V[Combining Dot Above]O2 and HR on the kayak ergometer in a laboratory setting (13). Furthermore, arm cranking is a small muscle mass upper body activity that reduces total cardiovascular stress (i.e., peak HR and peak V[Combining Dot Above]O2) during maximal exercise compared with upright lower body exercise. In addition, premature fatigue may occur in upper body arm cranking because of the decreased ability to extract and use oxygen compared with the lower body, thus reducing the total cardiovascular stress including the HR response (8). Finally, it is likely that the motivation and game-related effort and stress also contributed to the increased peak HR response from the competition.
Physiological profiling provides further evidence of the demands of the sport (e.g., training adaptations) and knowing the characteristics of elite international players is important to evaluate fitness standards. Forbes and Chilibeck (13) reported a peak V[Combining Dot Above]O2 of 44.2 ml·kg−1·min−1 during arm crank ergometry in trained sprint kayakers, whereas Bloxham et al. (6) reported a peak V[Combining Dot Above]O2 of 37.6 ml·kg−1·min−1 in wheelchair athletes involved in international basketball, which is also a high-intensity intermittent sport. These results are comparable with the peak V[Combining Dot Above]O2 results of 39.2 ml·kg−1·min−1 found in this investigation. These values are notably higher than the mean values reported previously in a systematic review for upper body work (26), suggesting that aerobic fitness of canoe polo players is an integral component to the sport. In addition, the VT may also be used as an indicator of aerobic fitness because it indicates an exercise intensity up to which exercise may be performed without rapid accumulation of blood lactate as a marker of accelerated anaerobic metabolic contribution (5). In this study, VT was achieved at 69% of peak V[Combining Dot Above]O2 and was similar to that in international level wheelchair basketball players (6). In addition, the anaerobic and musculoskeletal assessment demonstrated a high level of overall fitness (6,14); however, grip strength and curl-up results were slightly lower than that for professional hockey players (23).
Canoe polo players may have high levels of aerobic and anaerobic fitness derived from their participation in the sport itself. It is generally accepted that to achieve aerobic and anaerobic adaptations, training intensity must exceed the VT (15). The time-motion analysis and HR data suggest that the demands of participating in and preparing for international canoe polo were sufficient to achieve this latter intensity. Future studies are needed to investigate physiological and fitness profiles of club level and international players to gain a better understanding of the importance of each of these fitness components.
The scientific knowledge and descriptive information pertaining to the movement patterns, physiological demands (e.g., HRs), and physiological characteristics of elite canoe polo athletes are important for talent identification, training monitoring, program designs, and development of sport-specific field testing. The information from this study suggests a highly intermittent nature of international canoe polo games that requires both a well-developed anaerobic and aerobic energy systems. Training should be divided among the movement categories so that forward paddling and contesting comprises in excess of 56% of training time with the remaining time focusing on all the other movements (i.e., sprinting, turning, backwards paddling, and dribbling). This information may also be useful for the development of a sport-specific field test that would be important for training monitoring. For example, this study suggests that a field test should incorporate repeated forward sprints and contesting interspersed with slow easy paddling. Future research is required to examine the reliability and validity of such a field test. Furthermore, training must consist of activities that stress the cardiovascular system, as the average HR response was 158 b·min−1, and 69% of the game was above the VT. These cardiovascular demands were higher than those achieved in a simulated match, suggesting that training beyond simulated matches is required for preparation at an international level. In addition, it is recommended that further research be conducted to evaluate how a difference in playing style or position influences the physiological demand of the sport of canoe polo in players of both international and club levels.
This research was funded by the Sport Science Association of Alberta through the Alberta Sport, Recreation, Parks, and Wildlife Foundation. The authors declared no conflicts of interest. There are no professional and financial relationships between the authors and companies and manufactures. The results of this study do not constitute endorsement by the authors or the National Strength and Conditioning Association.
1. Ali A, Farrally M. A computer-video aided time motion analysis technique for match analysis
. J Sports Med Phys Fitness 31: 83–88, 1991.
2. Alves CRR, Pasqua L, Artioli GG, Roschel H, Solis M, Tobias G, Klansener C, Bertuzzi R, Franchini E, Lancha-Junior AH, Gualano B. Anthropometric, physiological, performance, and nutritional profile of the Brazil National Canoe Polo Team. J Sports Sci 30: 305–311, 2012.
3. American College of Sports Medicine (ACSM). Guidelines for Exercise
Testing and Prescription (4th ed.). Baltimore, MD: Lea and Febiger, 1991.
4. Bell GJ, Snydmiller GD, Game AB. An investigation of the type and frequency of movement patterns of National Hockey League goaltenders. Int J Sports Physiol Perform 3: 80–87, 2008.
5. Bhambhani Y, Singh M. Ventilatory thresholds during a graded exercise
test. Respiration 47: 120–128, 1985.
6. Bloxham LA, Bell GJ, Bhambhani Y, Steadward RD. Time motion analysis and physiological profile of Canadian world cup wheelchair basketball players. Sports Med Train Rehabil 10: 183–198, 2001.
7. Bressel EM, Bressel M, Marquez M, Heise GD. The effect of handgrip position on upper extremity neuromuscular responses to arm cranking exercise
. J Electromyogr Kinesiol 11: 291–298, 2001.
8. Calbet JA, Holmberg HC, Rosdahl H, van Hall G, Jensen-Urstad M, Saltin B. Why do arms extract less oxygen than legs during exercise
? Am J Physiol Regul Integr Comp Physiol 289: 1448–1458, 2005.
9. Campagna PD, Holt LE, Alexander AB, Terauds JA. A kayak
ergometer for dry-land testing and conditioning. In: Biomechanics in Sports III & IV: Proceedings of ISBS. Terauds J., Gowitzke B.A., Holt L.E. eds. Del Mar, CA: Academic Publishers, 1987. pp. 108–114.
10. Canadian Society for Exercise
Physiology. Health and Fitness Program’s Certified Exercise
Physiologist Certification Guide. Ottawa, Canada: Canadian Society for Exercise
11. D’Auria S, Gabbett T. A time-motion analysis of international women’s water polo match play. Int J Sports Physiol Perform 3: 305–319, 2008.
12. Esposito F, Impellizzeri FM, Margonato V, Vanni R, Pizzini G, Veicsteinas A. Validity of heart rate as an indicator of aerobic demand during soccer activities in amateur soccer players. Eur J Appl Physiol 93: 167–172, 2004.
13. Forbes SC, Chilibeck PD. Comparison of a kayaking ergometer protocol with an arm crank protocol for evaluating peak oxygen consumption. J Strength Cond Res 21: 1282–1285, 2007.
14. Forbes SC, Fuller DL, Krentz JR, Little JP, Chilibeck PD. Anthropometric and physiological predictors of flat-water 1000 m kayak
performance in young adolescents and the effectiveness of a high volume training camp. Int J Exerc Sci 2: 106–114, 2009.
15. Fox E, Bowers R, Foss M. The Physiological Basis of Physical Education and Athletics (4th ed.). Dubuque, IA: Wm. C. Brown, 1988.
16. Gabbett TJ, Mulvey MJ. Time-motion analysis of small-sided training games and competitions in elite women soccer players. J Strength Cond Res 22: 543–552, 2008.
17. Hopkins WG. Measures of reliability in sports medicine and science. Sports Med 30: 1–15, 2000.
18. Inbar O, Bar-Or O, Skinner JS. The Wingate Anaerobic Test. Champaigne, IL: Human Kinetics, 1996.
19. International Canoe Federation. Canoe Polo Competitive Rules. Lausanne, Switzerland: Maison du Sport International. Available at: www.canoeicf.com
, 2009. Accessed August 20, 2010.
20. Lythe J, Kilding AE. Physical demands and physiological responses during elite field hockey. Int J Sports Med 32: 523–528, 2011.
21. 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.
22. Price MJ, Campbell IG. Determination of peak oxygen uptake during upper body exercise
. Ergonomics 40: 491–499, 1997.
23. Quinney HA, Dewart R, Game A, Snydmiller G, Warburton D, Bell GA 26 year physiological description of a National hockey league team. Appl Physiol Nutr Metab 33: 753–760, 2008.
24. Quinney HA, Smith D, Wenger HA. A field test for the assessment of abdominal muscular endurance in professional ice hockey players. J Orthop Sports Phys Ther 6: 30–33, 1984.
25. Reilly T, Thomas V. A motional analysis of work-rate in different positional roles in professional football match play. J human Mov Stud 2: 87–97, 1976.
26. Sawka MN. Physiology of upper body exercise
. In: Exercise
and Sport Sciences Reviews. Pandolf K.B., ed. Vol 20. New York, NY: Macmillan, 1986. pp. 175–211.
27. Smith HK. Applied physiology of water polo. Sports Med 26: 317–334, 1998.
28. Tan F, Polglaze T, Dawson B. Activity profiles and physical demands of elite women’s water polo match play. J Sports Sci 27: 1095–1104, 2009.
29. Wasserman K, Whipp BJ, Koyl SN, Beaver WL. Anaerobic threshold and respiratory gas exchange during exercise
. J Appl Physiol 5: 236–243, 1973.
30. Wells KF, Dillon EK. The sit and reach. A test of back and leg flexibility. Res Q 23: 115–118, 1952.
31. Yuhasz MS. Physical Fitness and Sports Appraisal Laboratory Manual. London, Canada: The University of Western Ontario, 1996.