The importance of accurately assessing and balancing increasing training loads (TLs) with recovery is critical to understanding and optimizing performance in any sport (3,25). Although a multitude of factors contribute to the body's adaptive response and subsequent performance, collision-based team sports present a unique challenge with respect to assessing the influence of violent impact forces on TL monitoring.
Training programs are typically prescribed in terms of an external TL. Defined as the work completed by an athlete (e.g., distance ran), external TL is measured independent of internal characteristics (i.e., an athlete's physiology). For example, in track and field, coaches often prescribe training based on distance or time (i.e., 10 × 100 m at 18 seconds each with 90-second recovery). Although external TL is important in terms of planning and training outcomes, the stimulus for adaptation is determined by the relative physiological stress imposed on the athlete or internal TL (27). Therefore, to evaluate training status and accumulative fatigue, monitoring internal TL is considered more relevant and is highly valuable in avoiding injury and symptoms of overtraining (21).
Whereas a number of methods are available for coaches to quantify internal TLs, the practicalities of such techniques do not easily lend themselves to collision-based team sports, such as Canadian football. One heart rate (HR)–based method that has proven useful in quantifying team sport internal TL is the concept of Training Impulse (TRIMP). Originally developed by Banister (3) as a strategy for integrating the components of training into a single term, this system's analysis approach is ideal for evaluating the majority of physical components within team sport training and competition. In Canadian football, however, there are a number of challenges to the TRIMP concept that complicate the accurate measurement of training-imposed stress on a player. First, as with many team sports, the intermittent nature of football with random discrete bouts of activity varying in both intensity and duration poses particular difficulties (1). Secondly, the HR equipment necessary to monitor an entire team is relatively expensive and requires a certain level of expertise to collect and interpret. This is particularly restrictive to a football team, which can consist of rosters of over 80 players, compared with soccer squads of approximately 22 players. Finally, and most importantly, the frequent collisions and invasive nature of wearing HR monitors in addition to football equipment mean that players may be reluctant to wear the devices, and adjustments to equipment is rarely possible. In light of these challenges, there is a need for an alternative method for quantifying internal TLs in Canadian football that is both practical and reliable.
The session-rating of perceived exertion (Session-RPE) method was first introduced by Foster et al. (13) as a simple system for strength and conditioning coaches to monitor the internal TL of several different training modalities. Based on the understanding that athletes can inherently monitor their own stress levels, Session-RPE allows a subjective intensity rating of the entire training session (13). Unlike conventional RPE methods, which rely on RPE values being reported throughout the session, Session-RPE encourages athletes to simplify the countless intensities experienced and views a session with a single global rating. To ensure a rating representative of the entire training session, athletes are asked to indicate intensity by referring to a numerical value on the Foster's Modified Borg (FMB) scale 30 minutes after completing the session (5,13). A single arbitrary unit for global internal TL is then calculated by multiplying the training duration (in minutes) by the RPE intensity value provided retrospectively by the athlete. Coaches are then able to evaluate trends in training, injury, and illness in relation to Session-RPE and the global intensity of the training session (13).
Undoubtedly, the simplicity of Session-RPE and ease of interpretation are the major advantages within team sports compared with other reported methods. Successful early comparisons of Session-RPE with more complex methods of internal TL quantification were reported in endurance sports (12,28), non-collision–based team sports (7,16) and resistance-trained athletes (8). Based on this evidence, research focusing on training modalities (17) and neuromuscular fatigue (20) in collision-based sport has used Session-RPE to quantify TL, despite no preexisting collision-specific validation of the method. More recently, Session-RPE has been validated within the collision-based team sport of rugby league (18). However, because of its continuous nature, rugby league has a greater reliance on aerobic energy provision than Canadian football (14,24). Similar to the American version, Canadian football is distinguished by discrete, intense anaerobic bursts of work coupled with short rest periods between plays. Violent blocking and tackling are inherent within these high-intensity plays and can produce considerable physical and physiological strain (24). Consequently, Canadian football imposes a unique stress on individual players that needs to be recognized and factored into a periodized plan through accurate monitoring of internal TL. Despite extensive research, the influence of frequent impact during Canadian (or American) football on the validity of Session-RPE remains unclear. Therefore, the aim of the current project was to validate Session-RPE as a measure of internal TL within Canadian football practice sessions using HR-based methods as the criterion measure. It was hypothesized that Session-RPE would present significant individual correlations between HR-based internal TL methods collected during Canadian football practice sessions.
Experimental Approach to the Problem
To negate any possible learning effects associated with the reliability of Session-RPE (13), a pilot study was implemented during the entire 2010 Canadian Interuniversity (CIS) football season. Session-RPE internal TL data of all players on the University of Saskatchewan competitive roster (N = 78) was collected for all practice sessions and competitive games.
Of the 62 returning players who completed the 2010 pilot study, 20 were recruited for the current investigation and were also refamiliarized with this FMB scale before the study commencement. The limited number of players recruited compared with the pilot study was solely because of availability of HR equipment. Practice session internal TL data using both Session-RPE and HR response methods were collected and used for comparison during the entire 2011 pre-competitive and competitive CIS football season (11 weeks, 713 practice sessions in total). Session-RPE data were also collected for competitive games solely to provide an indication of TL. The CIS football program at the University of Saskatchewan was selected as it presented a more advantageous model from which to evaluate the influence of heavy collisions on Session-RPE. To clarify, at this level not only do rosters comprise a high performance cohort but practice sessions also involve more collision compared with that of professional American and Canadian football teams who are contractually obliged to minimize the amount of contact per training week.
Twenty male participants were recruited from the competitive roster of the 2011 University of Saskatchewan Canadian football team (Table 1). Players were all regular starters from the 2010 CIS season and were focused upon because of the greater potential TLs associated with full practice session and regular competitive game involvement. The top 3 players on positional depth charts created by the University of Saskatchewan coaching staff were approached and were asked to volunteer. Players were representative of all positions excluding Kickers. It is mandatory for all athletes to be screened and gain medical clearance by a clinical physician before participation in the CIS sport, and therefore, all subjects were considered healthy. Inclusion criteria for the competitive roster were based on subjective observations of an experienced coaching staff during numerous sport-specific performance and selection camps. This rigorous selection process allows a confident definition of this population as high performance. All players gave written consent according the American College of Sports Medicine guidelines. Ethical approval was obtained from the University of Saskatchewan Biomedical Research Ethics Board.
Football Season Structure and Training
For both the 2010 and 2011 CIS seasons, the preseason consisted of a 10-day training camp with 2 football-specific practice sessions per day culminating with 1 exhibition game. Immediately after the preseason, a 9-week competitive season began with 8 competitive games and 1 bye week. During both seasons, the postseason consisted of 1 week of practice before the team was eliminated in the first round of playoffs. Practice sessions were planned solely by the coaching staff, which included tactical, technical, and conditioning components (2010 = 48 sessions; 2011 = 51 sessions). In a typical week, all players participated in 4 practice sessions (Monday, Tuesday, Wednesday, and Thursday) and a competitive CIS game (Friday). Full contact was incorporated into practice sessions during the first 3 days of a training week, where helmets and full pads were worn. During the practice before game day (typically Thursday), contact was minimal for all starters with players wearing just helmet and shoulder pads. Active recovery sessions were held on the day after the game (Saturday) and 1 day was dedicated to complete rest (Sunday). During the competitive season, the weekly training plan varied slightly depending on game day scheduling. All practice sessions were completed on modern artificial turf at the University of Saskatchewan football stadium.
Measures of Internal Load
Practice Session-Rating of Perceived Exertion
Using the FMB scale (Table 2), each player's RPE intensity was collected verbally approximately 30 minutes after completion of every practice session (after a 15-minute cool down and an approximately 10- to 15-minute debriefing by the head coach). This timing ensured that the perceived effort be reflective of the entire session rather than the most recent exercise intensity. The FMB scale has proven frequently successful with athletes in previous Session-RPE studies (7,8,12,16,28), and players were asked to report intensity relative to other football practices, which included all aerobic, anaerobic, and contact components. In accordance with Foster et al. (13), an arbitrary unit that represents each individual player internal TL was then calculated by multiplying the training duration (in minutes) by training intensity (RPE value retrospectively indicated). Training or practice duration was defined as the time elapsed from the beginning of team warm-up until the completion of the last training component prescribed by the head coach.
Competitive Session-Rating of Perceived Exertion
Rating of perceived exertion intensity was also collected verbally approximately 30 minutes after competitive games from players who competed in a minimum of 3 quarters. Internal TL was calculated by replacing training duration with the time elapsed from kickoff to the final whistle. Competitive Session-RPE was collected for the purpose of tracking TL (Figures 2A, B) and was not used for comparison against HR data because competitive games were not conducive to wearing HR monitors.
Thirty to 60 minutes before a practice session, each player was fitted with a HR monitor on the chest in accordance with the manufacturer's guidelines (Polar Team2 Pro, VantageNV; Polar Electro, Kempele, Finland). Bespoke padding that protected the player, opponents, and HR monitor from collisions were created using football hip pads and Velcro strips (Figure 1). Each HR monitor consisted of an individually coded transmitter to avoid interference, and the HR was recorded every 3 seconds during each practice session. To reduce HR recording error, the online feature of the Polar Team2 system was used to provide live monitoring of HR responses from all 20 players throughout the practice sessions. Although rare, any irregularities were identified and fixed at the next convenient stoppage to avoid any interference with regular practice. Players were also briefed as to the correct fitting procedure, location of the HR monitor, and how to ensure that the monitors were functioning properly once in place. Upon practice session completion, HR data were saved on a portable PC using the specific Polar Team2 software and subsequently exported and analyzed further using Excel (Microsoft Corporation, Redmond, WA, USA) and custom-written MATLAB routines (Mathworks, Natick, MA, USA).
Two HR-based methods were used as the criterion measure of internal TL, the first of which was Polar TRIMP. Automatically calculated by the Polar Team2 software, this method incorporated Banister's (3) original TRIMP formula in a tailored recovery estimation model (23). The second HR-based method was proposed by Edwards (10). Known as Edwards' TL, this HR-based measure determines internal TL by measuring the product of the accumulated training duration (in minutes) of 5 HR zones by a coefficient relative to each zone (50–60% of HRmax = 1, 60–70% of HRmax = 2, 70–80% of HRmax = 3, 80–90% of HRmax = 4, 90–100% of HRmax = 5) and then summating the results (10). Heart rate maximums were predicted by age (HRmax = 220 − age), and only complete HR data were used for comparison. Incomplete data were typically because of players not being able to complete or participate in full practice because of injury.
Individual Pearson's product moment correlations between Session-RPE TL and HR-based TL were computed using the number of practice sessions as cases for each player according to the methods of Wallace et al. (28) and Impellizzeri et al. (16).
Based on these data, an estimate of the population correlation between Session-RPE TL and HR-based TL was 0.75. Using this estimate of effect size (ES), a power calculation was computed for the correlation using Delta = 2.80 and ES = 0.75, and it was determined that a minimum of 15 participants were required to achieve 80% power at alpha = 0.05. Alpha level was set as 0.05.
Weekly training cycles (periodization) of mean internal TL described using Session-RPE is shown for both the 2010 and 2011 CIS seasons (Figures 2A, B). Competitive Session-RPE was also included, although no HR data were collected during competition. Maximum TL loads were reported after competitive games and during the preseason that included 2 practice sessions per day (days 0 to preseason exhibition game 1, Figure 2B). Weekly practice session TL typically peaked during the second or third session after the game and progressively decreased as the next game approached.
During football practice sessions, significant (p < 0.01) correlations were observed between Session-RPE with Polar TRIMP (r = 0.65–0.91) and with Edwards' TL (r = 0.69–0.91), respectively. Individual correlations are presented in Table 3. Figure 3 is a representative example of how Session-RPE describes similarly both Edwards' TL and Polar TRIMP during football practice sessions.
The present study is the first to apply Session-RPE to Canadian football to monitor internal TL of practice sessions. Significant correlations were demonstrated between 2 criterion HR-based internal TL measures (Edwards' TL and Polar TRIMP) and Session-RPE. As expected and in agreement with the previous literature (7,8,12,16,28), our results support the notion that Session-RPE is a valid tool for monitoring internal TL in numerous sports. Importantly, however, this noninvasive measurement has now been shown to accurately represent internal TL in team sports, which involve frequent impact force through violent collisions.
Significant correlations (ranging from 0.65 to 0.91) reported in the current investigation are similar to other reported Session-RPE and HR-based TL comparisons in soccer (16) (r = 0.50–0.85) and slightly lower than those found in endurance athletes (12) (r = 0.75–0.90). In explanation, Impellizzeri et al. (16) suggested that the reduced correlation strength in soccer compared with endurance events is because of an increased provision of energy via anaerobic pathways leading to an overestimation of internal TL through higher RPE values. Support for this notion is evident in previous research, which reported higher RPE values during intermittent compared with work-matched steady-state protocols (9). Although the metabolic requirements of soccer and football are much different, the superior anaerobic reliance of these sports is a common characteristic, whereas the frequency and intensity of physical contact in football is much greater. Increased neural disruption, muscle tissue damage, and stress because of the violent collisions associated with a football game have been reported (15), which potentially could lead to greater pain perception and increased reported RPE values. Therefore, the higher anaerobic energy provision and the combative nature of football may go some way to explain the slightly reduced correlations in football compared with endurance sports. Furthermore, the unpredictable frequency and nature with which both high-intensity bursts and physical contact occurs during football may account for the wide range of individual correlations seen within the present study (Table 3). Despite the wide range of correlation values, all were statistically significant at the 99% confidence level.
The most endearing element of Session-RPE to Canadian football and other collision-based team sports is the psychobiological nature of the single RPE value. The original Session-RPE concept was based on the inherent ability of an athlete to subjectively monitor both physiological and psychological stress placed on their body during training (12). When both anaerobic and aerobic energy systems are concurrently activated, such as seen in intermittent team sports, research has shown RPE to be a more than reliable measure of exercise intensity (2). Furthermore, RPE is predicted more accurately with the combination of HR and blood lactate concentration as opposed to either variable alone (5). As mentioned previously, it is plausible to suggest therefore that Session-RPE may be sensitive to both the physical damage and mental pain experienced with multiple violent football collisions.
The line separating optimal training and overtraining remains undefined; however, research has shown RPE values to be sensitive to changes in an athlete's physiological and psychological tolerance to training (22). Additional support has been seen in overtraining studies reporting greater RPE values for a given HR during overreaching (19), as well as increased RPE values to a standardized exercise stimulus from athletes in a heightened state of fatigue (26). As suggested by Impellizzeri et al. (16), RPE may be more sensitive than HR to an athlete's fatigue state, and this may partially explain the moderate correlations seen between Session-RPE and HR-based TL observed in some individuals during the length of the current study. Consequently, Session-RPE can be considered a valuable monitoring tool that drives evidence-based guidance of an athlete's training in collision-based sports, to maintain the intricate training-recovery balance and avoid overtraining.
Developing and maintaining fitness over the course of an entire Canadian football season, as well as preventing and recovering from injuries, requires careful planning and implementation of a relatively complex periodized training program (4,6). Logical weekly fluctuation of internal TL calculated via RPE would imply that the application of Session-RPE during a football season is representative of undulating stress placed on individual players and useful in training periodization (Figure 3). Within the present study, further anecdotal support was provided by the experienced coaching staff that recognized an intentional 4-day bell-shaped format in training intensity, which is common in team sports with weekly competition (4). Visual validation of the preparation process convinced the coaches of Session-RPE's applicability, and as the season progressed, TL data and knowledge of results were used to plan adequate recovery for the team before the next match.
Standardizing the timing of rating collection and FMB scale familiarization are 2 areas that can affect the reliability of Session-RPE (13). A definite learning effect was seen within our pilot study, which implemented Session-RPE during the entire 2010 CIS football season (Figure 2A). Despite all players being familiarized with the FMB-scale before this pilot study, the results suggested that a learning effect may have occurred characterized by an intensity overestimation compared with the current data. For example, on reflection of certain sessions, players would often comment on how they would have ranked a previous session lower if they had already experienced the current session. This type of learning effect was only seen during the 2010 season and may have affected the outcome of the current project if a pilot study had not been implemented. Because only returning players were used in the current investigation and were all refamiliarized with the FMB scale, the pilot study was a preexisting solution to any learning effect. In addition, the inclusion of the maximal tests during the mandatory preseason testing ensured that all players experienced intensities that theoretically equate to 10 or “maximal effort” on the FMB scale. This provided the players with an exertion reference from which all other sessions could be gauged.
Based on the body of literature and our previous pilot study data, it was hypothesized that Session-RPE would accurately represent internal TL during training sessions within Canadian football. The data presented here confirm this hypothesis. However, a refinement of this method may be required for Session-RPE to truly reflect the physical stress placed on players during a competitive game. Anecdotal evidence from 2 seasons worth of Session-RPE monitoring (Figures 2A, B) suggests that the stress imposed on each individual by a competitive game may not be truly reflected by the Session-RPE method. Players may over, or in some cases, underestimate intensity depending on position played, tactical involvement, time of possession, number of plays, and tackles made or received during a game. On several occasions after the game, players stated that the FMB scale should be extended beyond 10 (maximal) or another scale used to represent the pain and discomfort experienced at “game intensity.” Indeed, a recent review on the use of RPE within sport outlines that Borg's Category-Ratio (CR-10) scale, which is often erroneously referenced in Session-RPE studies, permits numerical values beyond 10, whereas the FMB scale does not (11). By acknowledging the possibility of beyond maximal efforts and accepting RPE values of 11, 12, or even 13, the CR-10 scale (5) may therefore be more representative of the magnitude of exertion experienced during competitive football games. Because of the constraints of acquiring HR data during competitive games, the current investigation did not include competitive game HR-based TL but acknowledges the importance of further research in this area.
Optimizing performance begins with knowing what an athlete is currently doing. As such, Session-RPE represents an inexpensive, noninvasive, and highly practical internal TL monitoring tool that can be used by all Canadian football players who, because of the combative nature of the sport, may be reluctant to wear HR monitors. The practical implications of such a tool are extensive with an almost limitless number of questions relating to individual responses to collision-based training or competitive stress. In North America, the number of participants in football at both ends of the health to high performance spectrum is considerable. Moreover, when extrapolating the potential findings to all collision-based sports (e.g., Rugby, Lacrosse, or Ice Hockey), the application of this monitoring tool can impact and benefit a large proportion of the international health and sporting community. For example, using Session-RPE to quantify internal TL can help in educating parents and grass-root coaches in the intricacies of training plans and recovery strategies. Additionally, by providing evidence and increasing awareness of individual responses to collision-based physical stimulus, Session-RPE data can help maximize physical development in the youth and elite while minimizing overtraining, illness, and injury across the board. Concussion injuries are particularly poignant currently within football along with concerns surrounding the long-term health effects that repeat incidence may have on an individual. In this case, application of the Session-RPE method can not only aid the careful management of players back to full training but also provide a valuable tool to begin investigating the relationship between fatigue and concussion incidents.
Even at the higher echelons of professional collision sports, any form of training, performance, recovery modalities, or rehabilitation interventions can be investigated with accurate determination of internal TL. Again it is the simplicity of Session-RPE that lends itself well to all high performance environments. Thereby forwarding Session-RPE research this project will also have a far-reaching impact on all areas of sport sciences targeting elite athletic preparation.
The authors would like to express their deepest gratitude to all the players and coaching staff on the University of Saskatchewan football team for participating in this study. We would also like to extend our gratitude to Dean Carol D. Rogers and Dr. Philip Chilibeck from the University of Saskatchewan College of Kinesiology for their invaluable advice and guidance.
1. Achten J, Jeukendrup AE. Heart rate monitoring: applications and limitations. Sports Med 33: 517–538, 2003.
2. Bangsbo J. The physiology of soccer: With special reference to intense intermittent exercise. Acta Physiol Scand Suppl 619: 1–155, 1994.
3. Banister E. Modeling elite athletic performance. In: Physiological Testing of the High-Performance Athlete. MacDougall J.W., ed. Champaign, IL: Human Kinetics, 1991. pp. 403–424.
4. Bompa TO. Periodization: Theory and Methodology of Training. Windsor, Canada: Human Kinetics, 1999.
5. Borg GAV, Hassmen P, Langerstrom M. Perceived exertion
in relation to heart rate and blood lactate during arm and leg exercise. Eur J Appl Physiol 65: 679–685, 1985.
6. Busso T, Denis C, Bonnefoy R, Geyssant A, Lacour JR. Modelling of adaptations to physical training by using a recursive least squares algorithm. J Appl Physiol 82: 1685–1693, 1997.
7. Coutts AJ, Murphy A, Pine M, Reaburn P, Impellizzeri F. Validity of the session-RPE method for determining training load in team sport athletes. J Sci Med Sport 6: 525–535, 2003.
8. Day ML, McGuigan MR, Brice G, Foster C. Monitoring exercise intensity during resistance training using the session-RPE scale. J Strength Cond Res 18: 353–358, 2004.
9. Drust B, Reilly T, Cable NT. Physiological responses to laboratory-based soccer-specific intermittent and continuous exercise. J Sports Sci 18: 885–892, 2000.
10. Edwards S. High performance training and racing. In: The Heart Rate Monitor Book. Edwards S., ed. Sacramento, CA: Feet Fleet Press, 1993. pp. 113–123.
11. Eston R. Use of ratings of perceived exertion
in sports. Int J Sports Physiol Perform 7: 175–182, 2012.
12. Foster C. Monitoring training in athletes with reference to overtraining syndrome. Med Sci Sports Exerc 30: 1164–1168, 1998.
13. Foster C, Florhaug JA, Franklin J, Gottschall L, Hrovatin LA, Parker S, Doleshal P, Dodge C. A new approach to monitoring exercise training. J Strength Cond Res 15: 109–115, 2001.
14. Gabbett T, King T, Jenkins D. Applied physiology of rugby league. Sports Med 38: 119–138, 2008.
15. Hoffman JR, Maresh CM, Newton RU, Rubin MR, French DN, Volek JS, Sutherland J, Robertson M, Gomez AL, Ratamess NA, Kang J, Kraemer WJ. Performance, biochemical, and endocrine changes during a competitive football game. Med Sci Sports Exerc 34: 1845–1853, 2002.
16. Impellizzeri FM, Rampinini E, Coutts AJ, Sassi A, Marcora SM. The use of RPE-based training load in soccer. Med Sci Sports Exerc 36: 1042–1047, 2004.
17. Lockie RG, Murphy AJ, Scott BR, Janse de Jonge XA. Quantifying session ratings of perceived exertion
for field-based speed training methods in team sport athletes. J Strength Cond Res 26: 2721–2728, 2012.
18. Lovell T, Sirotic AC, Impellizzeri FM, Coutts AJ. Affecting perception of effort (session-RPE) during rugby league training. Int J Sports Physiol Perform 2012. Epub ahead of print.
19. Martin DT, Anderson MB. Heart rate-perceived exertion
relationship during training and taper. J Sports Med Phys Fitness 40: 201–208, 2000.
20. McLean BD, Coutts AJ, Kelly V, McGuigan MR, Cormack SJ. Neuromuscular, endocrine, and perceptual fatigue responses during different length between-match microcycles in professional rugby league players. Int J Sports Physiol Perform 5: 367–383, 2010.
21. Meeusen R, Duclos M, Gleeson M, Rietjens G, Steinacker J, Urhausen A. Prevention, diagnosis and treatment of the Overtraining Syndrome. Eur J Sport Sci 6: 1–14, 2006.
22. Morgan WP. Psychological components of effort sense. Med Sci Sports Exerc 26: 1071–1077, 1994.
23. Nissila J, Kinnunen H. Heart rate based training load and recovery time estimation. Polar Electro Oy 1: 1–4, 2008.
24. Rhea MR, Hunter RL, 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.
25. Smith D. A framework for understanding the training process leading to elite performance. Sports Med 33: 1103–1126, 2003.
26. Synder AC, Jeukendrup AE, Heeselink MK, Kuipers H, Foster C. A physiological/psychological indicator of overreaching during intensive training. Int J Sports Med 14: 29–32, 1993.
27. Viru A, Viru M. Nature of training effects. In: Exercise and Sport Science. Kirkendall W.G.. Philadelphia, PA: Lippincott Williams and Wilkins, 2000. pp. 67–95.
28. Wallace LK, Slattery KN, Coutts AJ. The ecological validity and application of the session-RPE method for quantifying training loads in swimming. J Strength Cond Res 23: 33–38, 2009.
Keywords:Copyright © 2013 by the National Strength & Conditioning Association.
contact sport; perceived exertion; internal training load; TRIMP; athlete monitoring