Changes in mood and affective states have frequently been described as consistent, sensitive, and early markers of overreaching and overtraining in competitive athletes (26,39). At least 3 sport-specific tools are available in the literature that attempt to assess an athlete’s ability to cope with the physical demands of training and competition and the daily life demands of performance sport (20,24,36). Several other research groups (22,28) have chosen to use and recommend a combination of available psychometric tools and checklists in their training and performance-monitoring regimes. Self-report measures of mood state responses exhibit a dose–response relationship with training load and have potential for monitoring athletes who undergo intensive physical training (32). In addition, they are efficient, inexpensive, and noninvasive (24).
Monitoring athlete wellness and adaptive responses to training and competition is also of interest to coaches and practitioners, yet in their applied environment, they are generally challenged by the practicalities of incorporating these comprehensive research tools into busy training schedules where concerns relating to compliance (“athletes hate paper work” (3)), the extent of data collection and analysis required, and lack of sport specificity seem to exist (38). As such, practitioners have been encouraged to incorporate the concepts underpinning these psychometric tools into some form of monitoring questionnaire or training diary (4,16,26). In a recent survey of fatigue monitoring in high-performance sport in Australia and New Zealand (50 elite/professional programs from 14 sports), 84% of respondents reported the use of self-report questionnaires (38). Of these, the use of questionnaires frequently used in scientific investigations remained limited, with 80% preferring to use a custom-designed form, typically consisting of 4–12 items measured on a 1–5 or 1–10 Likert point scale. In modifying existing questionnaires, practitioners placed emphasis on ratings of muscle soreness, physical fatigue, and general wellness and collected data on a frequent, often daily, basis. The literature, however, contains little in terms of the usefulness or validity of these increasingly common applied practices, particularly in team sport athletes where training and competition loads are relatively consistent over extended periods of time and the emphasis is on the routine management of fatigue and recovery to perform on a weekly basis.
Australian football is a physically demanding team field sport that includes body contact, repeated high-intensity efforts, and running distances of approximately 12 km per game (40). Changes in neuromuscular performance and endocrine status have been observed up to 72 hours post an elite Australian football game, with mean power in a single countermovement jump decreased and cortisol substantially elevated 24 hours postgame before returning to pregame levels around 72–96 hours postgame (6). Similar disruptions to neuromuscular performance and endocrine and biochemical markers of fatigue have been described in rugby league after match play (25), whereas player perceptions of stress and recovery in rugby league (7) and rugby union (28) have demonstrated sensitivity to manipulations in training load, intensity, and schedule.
Given the apparent sensitivity of self-report ratings to changes in training status and the increasing practical application of customized psychometric tools in high-performance sport, the purpose of this research was to describe how players were coping with the demands of elite-level Australian football over a competitive season using subjective ratings to monitor changes in physical and psychological wellness. The objective was to evaluate the usefulness of an applied practice within a professional sport environment and to begin to assess the ecological validity of such a monitoring approach. Relationships with weekly playing performance, a primary objective of any team and a recommended variable to monitor within the overtraining literature (15), were also assessed.
Experimental Approach to the Problem
Monitoring of senior players from a professional Australian football team using a selection of physical and psychological wellness items related to fatigue, recovery, and well-being was undertaken over a competition season. Subjective feedback ratings were introduced by the club in an attempt to receive feedback on a very regular basis from all players as to how they were coping with the demands of training and competition. The researchers, along with coaching and sport science staff working within the club, were particularly interested in how well the self-reported data reflected individual and team responses to variations in the training and competition program over the season. Consequently, perceptions of wellness were evaluated on different days during the training week (e.g., Do wellness items reflect fatigue and muscle soreness postgame and subsequent recovery prior to the next game?) and weeks across the season (e.g., Do perceptions of wellness change on a weekly basis in response to variations in the periodised program?). Individual player characteristics were used to assess if trends existed in the data based on individual circumstances (e.g., Do age or measures of fitness influence a player’s perception of fatigue or recovery?). Weekly measures of playing performance were also monitored to assess potential relationships with individual wellness items (e.g., Is playing performance adversely affected if wellness ratings are poor leading into a game?).
The statistical approach used employed multilevel linear modelling techniques (33) to predict each wellness item (dependent variable) in 2 separate analyses designed to address research questions related to athlete responses to training and competition demands. The first analyses considered data across the week, pooling data for similar days and modelling the intercept and slope (based on mean data for each player for each day of the week), although at the same time assessing the influence of player characteristics on these coefficients. The second analysis followed a similar approach, yet considered data across the season (based on mean data for each player for each week).
Data were provided by 27 professional players (mean ± SD [range]: age 24.4 ± 2.9 years [19–30]; height 187.4 ± 7.0 cm [175–202]; weight 89.9 ± 8.1 kg [77–108]; playing experience 97 ± 59 senior games [9–213]) from the same club during an entire Australian Football League (AFL) season. The AFL is the highest level of competition played in the sport, and data are presented for both in-season and finals. Player characteristics were recorded at the beginning of the season, whereas the analysis included only data for each week from the 22 competing players (number of AFL games per player during the season = 19 ± 6 [5–25]). Methods for the study were approved by the club and Deakin University Human Research Ethics Committee, and players provided written informed consent.
Player characteristics including measures of running fitness were used as predictor variables in the analysis of wellness data over the season. Speed over 40 m (31.6 ± 1.5 km·h−1 [29–35]) and 3-km time trial running endurance (619.5 ± 39.6 seconds [558–698]) were assessed during mid-January in the precompetition period. Tests were conducted on a well-maintained grass surface, after an appropriate warm-up. Speed was assessed with timing gates (Swift Technology, Lismore, NSW, Australia) to one-hundredth of a second and then converted to km·h−1. The 3-km time trial took place around an oval 460 m in circumference, measured with a calibrated trundle wheel, with markers placed every 20 m. The typical error for these tests in this playing group, expressed as coefficient of variation, was 1.4% for the 40-m sprint (n = 22) and 1.1% for the 3-km time trial (n = 19).
Training and Competition
For this club, the season consisted of 22 in-season games (weeks 1–12, 14–23), 3 finals games (weeks 24, 26, 27), and 2 weeks in which no game was played (weeks 13, 25). Data leading into a week where no game was played were not assessed, whereas data immediately post this week and leading into the next game were included. Weeks 7, 14, 22, and 26 represented periods of reduced training within the periodized training program. Games were played on a weekly basis, with some variation in the number of days between games depending on when the game was played on the weekend (i.e., Friday, Saturday, or Sunday). A consistent pattern of training was prescribed each week with scheduling determined by the day of the upcoming game, allowing for comparisons of data on any given day possible. The days immediately postgame focused on recovery and modified training; the main training session of the week was scheduled 3 days before the game (3 days pre: 72 ± 8 minutes training time excluding breaks, range: 55–85 minutes); light skill and tactical sessions preceded the game (e.g., 1 day pre: 39 ± 3 minutes, range: 33–46 minutes). The weekly training schedule (for a 7-day game–game schedule), including the typical prescribed physical load as a percentage of total weekly load (estimate based on routine Global Positioning System, heart rate, and rating of perceived exertion data), was as follows: 1 day post game: recovery; 2 days post game: skills (6%) and weights (5%); 3 days post game: off; 3 days pre game: skills (17%) and weights (12%); 2 days preskills or cardiovascular fitness (5%) and weights (5%); 1 day pre games: skills (5%); game (45%). Training was adjusted in the middle of the week to accommodate shorter or longer game–game schedules.
The players completed ratings for 9 wellness items, 6 of which were physical in nature (fatigue, general muscle strain, hamstring strain, quadriceps strain, pain/stiffness, power) and 3 psychological or lifestyle-related (sleep quality, stress, well-being). Selection of items was guided by the specific areas of interest of sports science and conditioning staff felt necessary to manage players (which resulted in an emphasis on physical symptoms rather than emotions) and by individual items or subscales frequently appearing in the athlete monitoring and overtraining literature (16,20,24,36). Two additional items were added that focused on common soft tissue injury sites in AFL football (i.e., hamstring strain and quadriceps strain) (30). The number of items was kept small to encourage player compliance and ensure that completion on all training and competition days was achievable over an extended period of time (i.e., 6 months). This monitoring approach was consistent with recent survey data that described fatigue-monitoring practices in high-performance sport in Australia and New Zealand (38).
Data were entered before any scheduled activity, usually in private and at a consistent time in the morning on similar days, with the exception of late afternoon for night matches. Players subjectively rated each item as they arrived at the training or competition venue on a computer touch screen displaying a visual analog Likert scale ranging from 1 (feeling as good as possible) to 5 (feeling as bad as possible), with data recorded to 2 decimal places. Data were recorded directly into commercially available athlete management software (ATHLETRAK, version 8.06; Athlete Logic, Cheshire, United Kingdom; http://athleticlogic.com) that allowed user-defined variable customization and data analysis. The data entry screen displayed all items together, one after the other, with each wellness item listed as a single word or descriptor (e.g., fatigue, sleep quality). Players were instructed to consider their rating as a response to a question that asked about how they currently felt in relation to each item (e.g., How would you rate your current level of fatigue?). The players were familiar with the rating system having completed the process over the preseason period and been instructed in its use by the senior sport scientist at the club.
To assess the reliability of the wellness-monitoring system, data from 2 similar weeks during the preseason competition phase were analyzed. The weeks consisted of a preseason game against an opponent in the same AFL competition with wellness data collected on 5 different days (1 day postgame, 2 days postgame, 2 days pregame, 1 day pregame, and on game day). Measures of test–retest reliability (N = 809), calculated using a customized spreadsheet (17), were as follows: typical error, expressed as a percentage coefficient of variation, % CV = 24.1%; intraclass correlation coefficient, ICC = 0.58; percentage change in the mean = −2.3%. Technical error was greatest on the day immediately after the game (CV = 31.9%, ICC = 0.43) and smallest on the day of the game (CV = 12.4%, ICC = 0.72). These reliability estimates likely reflect variations in player responses from week to week, both in how they felt and in how they rated their feelings, and normal variations in training and game load during the 2 weeks.
For the purpose of this study, data were exported at the conclusion of the season for analysis; however, during the season, the data were considered on a daily and weekly basis by senior sport science and conditioning staff to assist with individual player management and training prescription.
Relationships with weekly playing performance, a recommended variable of interest within the overtraining literature (15) and a primary objective of each individual player, were also assessed. Each player’s performance during competition was recorded as a single-performance score derived from 33 individual game statistics that incorporated all aspects of play (i.e., offensive = 16 statistics, defensive = 9, stoppages = 8) provided by 2 AFL-approved companies (Champion Data, Victoria, Australia, http://www.championdata.com.au/; ProWess Sports, Victoria, Australia, http://www.prowess.com.au/). Champion Data have logged qualitative AFL statistics by computer since 1996 and report a better than 99% accuracy (29). In a procedure similar to that used by Richmond et al. (34), individual statistics were weighted for importance by the coaching staff using a confidential formula agreed at the beginning of the season. The playing performance score was expressed in arbitrary units and calculated on a weekly basis for each player after every game.
Preliminary data analysis, including checks and appropriate actions (e.g., log transformation of wellness variance) to meet assumptions of normality, linearity, and homoscedasticity, along with analysis of bivariate relationships between playing performance and wellness items were completed using the Statistical Package for the Social Sciences for Windows (version 17.0; SPSS, Inc., Chicago, IL, USA). Data are presented as mean ± SD, with statistical significance set at p < 0.05.
Behavioral data, such as that collected over a season in this applied setting, can be hierarchical and commonly have a nested structure as measurement occasions, and the number of repeated observations on each individual are not identical (33). Multilevel linear modelling techniques have been developed to appropriately deal with data structures such as these, with each submodel representing the structural relations and residual variability at that level. In the present study, multilevel models were used (HLM, version 6; Scientific Software International Inc., Lincolnwood, IL, USA) to test the significance of week and day of the week effects on wellness and assess for moderating effects of player characteristics on these relationships. This approach fits a model for each player from repeated observations over time (level 1 predictors) and then models each coefficient in these models as a random effect allowing for differences in player characteristics (level 2 predictors). Characteristics typically used to describe a player were used as level 2 predictors in the analysis as follows: age, height, weight, speed, running endurance, and playing experience. To reduce the impact of multicollinearity, level 1 predictors were player centered and level 2 predictors were grand mean centered (33).
To analyze trends across the week, data for each similar day for each player were used. To analyze trends across the season, the mean and rating variability for each player for each week were used (e.g., the mean and SD of daily ratings for week 1, week 2, etc.). Correlations between the game performance score and wellness scores (mean and SD for the week) were examined to determine what wellness items were likely predictors of performance.
In total 2,583 questionnaires were analyzed from completions on 183 days throughout the season. This represented a mean total of 92 ± 24 completions per player for the season (range: 31–132; compliance 70%), 3.8 ± 0.7 completions per player for each week, and 103 ± 20 completions per week for the entire squad.
Table 1 summarizes significant effects for the absolute player wellness data in player ratings over the week and season. Perceptions of wellness in all 9 items typically had low values (the constant term in Table 1; lower scores being preferable on the 1–5 scale) suggesting players generally coped well with the demands of elite AFL football. Pain/stiffness and sleep quality had the highest average scores (over the entire season) with quadriceps strain, stress, and well-being having the lowest scores.
The slope for days to game is always significant (Table 1 and Figure 1), highlighting the improvement in all wellness items as game day approaches. However, the coefficient for the slope is moderated in several items by maximum speed, indicating that faster players have significantly higher (worse) ratings for muscle strain, hamstring strain, quadriceps strain, and power after a game. Sleep quality is more adversely affected after a game in older players.
Figure 2 presents mean data (±95% confidence intervals) over the season for selected physical (fatigue, hamstring strain) and psychological (sleep quality, well-being) variables, with individual ratings for fatigue, muscle strain, and well-being significantly improving over the course of the season (Table 1). Several items were significantly lower immediately post a week of no competition in week 14 (fatigue, hamstring strain, quadriceps strain, power, sleep) and week 26 (fatigue, pain/stiffness, power).
The effect of weeks on several wellness items is moderated by the individual characteristics of playing experience and maximum speed. For players with greater experience (total AFL games played) ratings of quadriceps strain and power improved to a greater extent over the season, whereas well-being deteriorated. Ratings of fatigue improved more so for players with higher maximum speed as the season progressed.
The log of wellness variance was used to assess variability in the data over time. The slopes for data across the week indicated that variability between players declined significantly (p < 0.001) in all items as game day approached. The greatest decrease in variability occurred for quadriceps strain, fatigue, and power, whereas stress and well-being showed the smallest decrease. Variability also declined significantly over the season for all items except pain/stiffness.
A case study example that provides raw data for a single player over the season is presented in Figure 3. This player suffered a hamstring strain during the game in week 6 and a groin strain in week 17, missing 2 and 3 games, respectively. Fluctuations in ratings are clearly evident throughout each week, whereas fatigue and pain/stiffness appeared elevated in the weeks leading up to both injuries. Feelings of stress were observed in the weeks after the hamstring strain and continued on for a number of weeks despite the player returning to competition.
Analysis of the relationships between wellness and performance formed the second part of this investigation. A total of 359 individual playing performances from 25 competitive games over the season were considered. Playing performance was derived from individual game statistics and was measured in arbitrary units (113.1 ± 51.8 [arbitrary units], range: 8–279).
A few significant but very weak negative correlations with performance were observed for general muscle (r = −0.105, p = 0.042) and hamstring strain (r = −0.110, p = 0.033) and for the SD of quadriceps strain (r = −0.178, p = 0.001) and hamstring strain (r = −0.121, p = 0.022). Stress levels over the week were positively correlated with performance (r = 0.216, p < 0.001).
The purpose of this research was to describe elite-level player perceptions of wellness over an entire season in a professional Australian football team using a customized daily monitoring questionnaire and to begin to evaluate the efficacy of this approach. Trends over the season and within the training week were evident, whereas individual player characteristics such as age, playing experience, and maximum speed moderated responses for a number of wellness items. These results suggest that player self-reported ratings of wellness are sensitive to daily and weekly variations in recovery status and support the use of player self-monitoring within a professional team sport environment.
The time course of changes in neuromuscular performance and endocrine status have been described in a number of field team sports including Australian football (6), soccer (1), and rugby league (25), with a return to pregame levels generally evident around 72–96 hours postgame. Similar fatigue responses and the pattern of change over the training week are evident in players’ perception of physical and psychological wellness in the present study. Poorer ratings of wellness in the days immediately postcompetition suggest considerable fatigue from the game which by the middle of the week were reasonably attenuated to allow participation in the main training session for the week (3 days pregame). Further improvements leading into a game day low suggest players perceived themselves to have recovered and were ready for the upcoming game. A similar pattern in player perceptual responses across the week in a comparable AFL team has recently been observed (2). In monitoring a group of professional rugby union players, Nicholls et al. (28) also found that more stressors were worse than normal the day after a game than on game day, with ratings on training days typically worse than on both rest days and game days.
An important finding in this study was that players with higher maximum speed in preseason testing reported worse ratings for power, muscle strain, hamstring strain, and quadriceps strain in the days after a game and consequently took longer to recover to baseline levels. Exercise-induced muscle damage is influenced by a variety of factors including exercise intensity, the number and velocity of contractions during exercise, work performed, exercised muscle length, and individual differences in fibre-type composition and muscle architecture (5,31). The time course of recovery may also be variable, with Paddon-Jones et al. (31) observing that muscle soreness was evident for a longer period (4 days post vs. 2 days post) after fast velocity eccentric exercise compared with slow velocity eccentric exercise, despite the severity of muscle soreness being similar in both groups. Faster players may therefore be more susceptible to muscle damage due to factors related to how they play the game (27) (e.g., greater speeds, rapid changes in direction and acceleration/deceleration, increased ground impact when running and on body contact) or to inherent factors related to muscle structure and composition. Players who demonstrate good anaerobic running profiles may experience greater muscle damage or be slower to recover from heavy training loads (14). Extensive muscle damage has been reported after sprint running even among trained individuals (12), whereas changes in ratings of physical stress may also reflect neural fatigue (35). Although these and other mechanisms may be involved, this is clearly speculative and warrants further investigation, particularly in collision-based running sports. Measurements of physical load in the actual game along with specific markers of fatigue and muscle damage are required to better assess the potential relationship between speed and recovery. Nonetheless, the finding that faster players experience more symptoms of physical stress after a game has important training and program design implications for individual player management, with increased emphasis on recovery practices and or reduced training load for players with high maximal speed.
Changes in ratings of sleep quality after a game and in the days immediately after are consistent with observations in athletes in heavy training (11), after prolonged vigorous exercise (10), and with suggested links with increases in proinflammatory cytokines (37). Support for further moderation of these changes in older athletes in the present study within such a narrow age range (i.e., 19–30 years) is not available although sleep quality is known to deteriorate with age (18). Older players also report lower average ratings for quadriceps strain and power, whereas those with greater playing experience report improvement in these 2 items across the season. Explanations for these findings in the older players are unclear although an improvement in perception of recovery and a progressive blunted response across the AFL season has been reported (2).
Improvements in ratings of fatigue, muscle strain, and well-being were observed over the course of the season as were decreases in rating variability. These improvements may be a result of a well-managed and administered training program, including effective recovery strategies and training load manipulations (21). They may also reflect structural adaptations to repeated bouts of muscle damage (2) and subtle improvements in players’ ability to cope with the demands of training and competition, although improvements in routine measures of fitness are typically not observed beyond the preseason (13).
The significant improvements in ratings of wellness after a single week of reduced physical load (i.e., no game and reduced training time in weeks 14 and 26) further demonstrates the sensitivity of player self-reports to daily changes in training and competition circumstances. Coutts et al.(8) have demonstrated physiological and performance improvements along with psychological improvements (7) after a short 7-day training taper in semiprofessional rugby players. The current study adds to this work in that it provides a daily time course related to changes in perceived wellness in response to both the routine reduction in training leading into each weekly game and the more marked reduction in training and football involvement associated with a week of no competition. The scheduling of a bye week (or a period of unloading) in the middle of the competitive season clearly has physical and psychological benefits.
The relationship between individual performance and ratings of wellness in this playing group was poor, although a few very weak significant correlations with performance were evident for some physical wellness items. Negative correlations with performance for general muscle and hamstring strain and for variation in ratings over the course of the week in quadriceps and hamstring strain suggest that performance is negatively impacted in players who report higher physical stress in these items in any given week. Although these correlations may only account for 1–3% of the variance in performance, winning and losing in professional sport is often determined by small margins, and any negative impact on performance is considered important.
Monitoring of responses to training and competition load and assessment of fatigue and recovery status is critically important for athlete performance management. Various approaches have been utilized within sport research, including biochemical, hormonal, neuromuscular, cardiovascular, and psychosocial monitoring of variables believed to be responsive to the fatigue-recovery state (15,23,26,39). Given this state has been proposed as having physiological, psychological, and social bases (21), and being influenced by both training and nontraining stressors, it is not surprising that no single reliable diagnostic marker has been identified to monitor its changing status (23). Psychometric questionnaire scales that seek to capture the full complexity of stress and recovery, such as the recovery-stress questionnaire for athletes (RESTQ-76 Sport) (20), likely provide the most comprehensive assessment of athlete status. Although a shortened recovery-stress questionnaire version is available (52 items), it is apparent that a majority of strength and conditioning and sports science practitioners working in high-performance sport chose to implement modified custom designed forms in their athlete-monitoring regimes (38). Reasons for doing so included concerns over cost, length of questionnaire, and time required to complete and analyze, and a perceived lack of sport specificity of existing questionnaires. Despite a lack of empirical validity, practitioners believed their modified self-report practices, implemented on a frequent and often daily basis over extended periods, provided them with valid information. The results of the present study support this belief; however, a number of cautionary comments and recommendations are warranted. To be more broadly effective and utilized, shortened or modified versions of monitoring questionnaires should meet the expectations of both practitioner and researcher; be published in a manner that provides guidelines for implementation, analysis, and interpretation; and seek to establish the validity and test–retest reliability under varying circumstances.
In summary, this study presents player self-ratings of wellness collected over an entire season in professional AFL footballers. The data is extensive (2,583 questionnaires collected on 183 training or competition days) and represents a good level of compliance. Other more detailed psychometric scales have been used within sport to evaluate athlete responses to training and competition, yet these have typically been administered on a small number of occasions or over a relatively short period of time (9,19,28). This study demonstrates that player self-monitoring through the use of a daily wellness questionnaire can provide valuable insight into the adaptive responses of athletes when training and competing and begins to provide some ecological validity for this approach. These practical tools are useful in an applied environment in that they are brief and easy to administer and can be used on a regular basis over an extended period of time. Future work should look to establish their construct validity and test–retest reliability as is the case with other psychometric scales used in sport research.
Subjective ratings of wellness appear sensitive to changes in load and individual circumstances and provide a useful tool to monitor adaptive responses to the rigorous demands of training, competition, and life as a professional athlete. The influence of individual player characteristics such as maximum speed, age, and playing experience suggest that players can respond differently to training and competition load and should be, wherever possible, managed individually.
Competition breaks within the season have physical and psychological benefits such that team sports with long competitive seasons should look for opportunities to periodically unload their players. This can be achieved through competition scheduled breaks (e.g., a bye), periodized unloading within the training plan, or a reactive response to the interpretation of monitoring and performance data at the both the individual and team level.
Although the findings in this study provide support for using self-report questionnaires, it is the regular (i.e., daily, weekly) review of this serial data and the conversations that ensue as a result with the athlete and within the performance management team that provide the most value. Interpretation of the data should be undertaken cautiously, in conjunction with other data and observations, and from an understanding of individual player baseline levels. To be useful, the approach also needs to be valued by players and coaches, requires a good level of compliance, and be conducted in a supportive environment underpinned by a desire to protect player welfare and team performance.
1. Ascensão A, Rebelo A, Oliveira E, Marques F, Pereira L, Magalhães J. Biochemical impact of a soccer match—Analysis of oxidative stress and muscle damage markers throughout recovery
. Clin Biochem 41: 841–851, 2008.
2. Bahnert A, Norton K, Lock P. Association between post-game recovery
protocols, physical and perceived recovery
, and performance
in elite AFL players. J Sci Med Sport. doi 10.1016/j.jsams.2012.05.008, 2012.
3. Beckmann J, Kellmann M. Procedures and principles of sport psychological assessment. Sport Psychol 17: 338–350, 2003.
4. Brukner P, Khan K, Holzer K. The tired athlete. In: Clinical Sports Medicine. Sydney, Australia: McGraw Hill, 2006. pp. 875–887.
5. Chapman DW, Newton MJ, Zainuddin Z, Sacco P, Nosaka K. Work and peak torque during eccentric exercise do not predict changes in markers of muscle damage. Br J Sports Med 42: 585–591, 2008.
6. Cormack SJ, Newton RU, McGuigan MR, Cormie P. Neuromuscular and endocrine responses of elite players during an Australian Rules Football season. Int J Sports Physiol Perform 3: 439–453, 2008.
7. Coutts AJ, Reaburn P. Monitoring changes in rugby league players' perceived stress and recovery
during intensified training. Percept Mot Skills 106: 904–916, 2008.
8. Coutts A, Reaburn P, Piva T, Murphy A. Changes in selected biochemical, muscular strength, power, and endurance measures during deliberate overreaching and tapering in rugby league players. Int J Sports Med 28: 116–124, 2007.
9. Coutts AJ, Wallace LK, Slattery KM. Monitoring changes in performance
, physiology, biochemistry, and psychology during overreaching and recovery
in triathletes. Int J Sports Med 28: 125–134, 2007.
10. Driver HS, Rogers GG, Mitchell D, Borrow SJ, Allen M, Luus HG, Shapiro CM. Prolonged endurance exercise and sleep disruption. Med Sci Sports Exerc 26: 903–907, 1994.
11. Fietze I, Strauch J, Holzhausen M, Glos M, Theobald C, Lehnkering H, Penzel T. Sleep quality in professional ballet dancers. Chronobiol Int 26: 1249–1262, 2009.
12. Friden J, Seger J, Ekblom B. Sublethal muscle fibre injuries after high-tension anaerobic exercise. Eur J Appl Physiol 57: 360–368, 1988.
13. Gabbett TJ, Domrow N. Relationships between training load, injury, and fitness in sub-elite collision sport athletes. J Sports Sci 25: 1507–1519, 2007.
14. Gastin PB, Fahrner B, Meyer D, Robinson D, Cook J. Influence of physical fitness, age, playing experience and weekly training load on match performance
in elite Australian football. J Strength Cond Res, doi 10.1519/JSC.0b013e318267925f, 2012.
15. Halson S, Jeukendrup A. Does overtraining
exist? An analysis of overreaching and overtraining
research. Sports Med 34: 967–981, 2004.
16. Hooper SL, Mackinnon LT, Howard A, Gordon RD, Bachmann AW. Markers for monitoring overtraining
. Med Sci Sports Exerc 27: 106–112, 1995.
18. Huang YL, Liu RY, Wang QS, Van Someren EJW, Xu H, Zhou JN. Age-associated difference in circadian sleep-wake and rest-activity rhythms. Physiol Behav 76: 597–603, 2002.
19. Kellmann M, Altenburg D, Lormes W, Steinacker JM. Assessing stress and recovery
during preparation for the World Championships in rowing. Sport Psychol 15: 151–167, 2001.
20. Kellmann M, Kallus KW. Recovery
-stress questionnaire for athletes: User manual. Champaign, IL: Human Kinetics, 2001.
21. Kentta G, Hassmen P. Overtraining
—A conceptual model. Sports Med 26: 1–16, 1998.
22. Kentta G, Hassmen P, Raglin JS. Training practices and overtraining
syndrome in Swedish age-group athletes. Int J Sports Med 22: 460–465, 2001.
23. Mackinnon LT, Halson SL, Hooper SL, Jeukendrup AE. Overreaching and overtraining
. In: Physiological Bases of Human Performance
During Work and Exercise. Taylor N.A.S., Groeller H, eds. Edinburgh, United Kingdom: Churchill Livingstone, 2008. pp. 505–520.
24. Main L, Grove JR. A multi-component assessment model for monitoring training distress among athletes. Eur J Sport Sci 9: 195–202, 2009.
25. McLellan CP, Lovell DI, Gass GC. Markers of postmatch fatigue in professional rugby league players. J Strength Cond Res 25: 1030–1039, 2011.
26. Meeusen R, Duclos M, Gleeson M, Rietjens G, Steinacker J, Urhausen A. Prevention, diagnosis and treatment of the overtraining
syndrome. ECSS position statement 'task force'. Eur J Sport Sci 6: 1–14, 2006.
27. Mendez-Villanueva A, Buchheit M, Simpson B, Peltola E, Bourdon P. Does on-field sprinting performance
in young soccer players depend on how fast they can run or how fast they do run? J Strength Cond Res 25: 2634–2638, 2011.
28. Nicholls AR, Backhouse SH, Polman RCJ, McKenna J. Stressors and affective states among professional rugby union players. Scand J Med Sci Sports 19: 121–128, 2009.
29. O'Shaughnessy DM. Possession versus position: Strategic evaluation in AFL. J Sports Sci Med 5: 533–540, 2006.
30. Orchard JW. Intrinsic and extrinsic risk factors for muscle strains in Australian football. Am J Sports Med 29: 300–303, 2001.
31. Paddon-Jones D, Keech A, Lonergan A, Abernethy P. Differential expression of muscle damage in humans following acute fast and slow velocity eccentric exercise. J Sci Med Sport 8: 255–263, 2005.
32. Raglin JS. Psychological factors in sport performance
: The mental health model revisited. Sports Med 31: 875–890, 2001.
33. Raudenbush SW, Bryk AS. Hierarchical Linear Models: Applications and Data Analysis Methods. Thousand Oaks, CA: Sage Publications, 2002.
34. Richmond LK, Dawson B, Stewart G, Cormack S, Hillman DR, Eastwood PR. The effect of interstate travel on the sleep patterns and performance
of elite Australian rules footballers. J Sci Med Sport 10: 252–258, 2007.
35. Ross A, Leveritt M, Riek S. Neural influences on sprint running: training adaptations and acute responses. Sports Med 31: 409–425, 2001.
36. Rushall BS. A tool for measuring stress tolerance in elite athletes. J Appl Sport Psychol 2: 51–66, 1990.
37. Santos RVT, Tufik S, De Mello MT. Exercise, sleep and cytokines: Is there a relation? Sleep Med Rev 11: 231–239, 2007.
38. Taylor K, Chapman DW, Cronin JB, Newton MJ, Gill N. Fatigue monitoring in high performance
sport: A survey of current trends. J Aust Strength Cond 20: 12–23, 2012.
39. Urhausen A, Kindermann W. Diagnosis of overtraining
: what tools do we have? Sports Med 32: 95–102, 2002.
40. Wisbey B, Montgomery PG, Pyne DB, Rattray B. Quantifying movement demands of AFL football using GPS tracking. J Sci Med Sport 13: 531–536, 2010.