It is known that the ability to appropriately distribute energy expenditure throughout an exercise task (pacing) is extremely important for athletic performance (2,13). In this regard, pacing has been referred to as self-regulation of power/speed during endurance competitions in which athletes are free to vary the work rate (e.g., time trials (TT)) (9). Several studies suggest that athletes develop a stable pacing pattern (PP) in their career, which seems to be consistent from one competition to the next (30,34). However, other studies have shown that physiological, psychological, and environmental factors can affect TT performance and/or pacing (36). These factors include hypoxia (27), hydration status (10), caffeine ingestion (38), motivation (8), and music (5).
The effects of fatigue on PP are, however, poorly understood. This is surprising, considering that many athletes have to compete in several competitions within a short period, resulting in accumulated fatigue (e.g., cycling stage races). It is already known that accumulated fatigue affects several physiological parameters, e.g., decreased blood lactate (BLa), HR, and/or V˙O2 (25,37), mostly explained by reduced sympathetic nervous system activity (19). However, it is unclear whether this also affects PP. Amann and Dempsey (4) previously described the effects of locomotor muscle fatigue on overall 5-km TT performance. Prefatiguing exercise consisted of constant-workload cycling until exhaustion, which was carried out 4 min before a 5-km cycling TT (4). After those prefatiguing trials, 5-km performance time increased by approximately 2%–6% with a significantly lower power output during the trial. The authors speculated that peripheral fatigue might prevent the muscle from responding to the same extent to an identical level of motor command (4). Interestingly, power output rose to almost identical levels in the final 200 m despite different levels of preexisting fatigue. According to the authors, this indicates that the locomotor muscles remained responsive to central motor drive throughout exercise, even in conditions of severe peripheral fatigue (4). This maintained capability to increase power output underlines the fact that subjects seem to (consciously and/or subconsciously) choose motor power output at a certain submaximal level, presumably to avoid further accumulation of peripheral fatigue before the end of the trial (4). In conclusion, these results seem to emphasize the crucial role of locomotor muscle fatigue for performance via its inhibitory influence on central motor drive (4). However, it remains unclear whether this might also have an effect on the PP.
A recently published study described the effect of a preexercise eccentric fatiguing protocol (100 drop jumps) on PP in a 15-min cycling trial (9). The TT was carried out 30 min after the jumps. The authors did not find a difference in PP between both conditions, even though overall performance was worse in the fatigued trial (9). Both the studies described focused on acute effects of locomotor muscle fatigue. However, in several endurance competitions, athletes have to compete in different races over several days and/or weeks (e.g., stage races, heats, and finals), supposedly leading to accumulation of fatigue. For example, road cyclists compete on 90–100 competition days, comprising 1-d races, 1-wk tour races, and 3-wk tour races (1). Within each of these races, cyclists may perform different competition requirements (e.g., flat, long stages, TT, uphill ascents), resulting in exhaustion of various physiological systems (1,26). Depending on the competition and the stage type, HR values range between 51% and 89% of HRmax, with power outputs between 192 and 380 W (26). This leads to high demands of aerobic and anaerobic pathways and might rather induce multifaceted accumulation of fatigue than acute locomotor muscle fatigue.
In this regard, it is already known that overall performance deteriorates after an intensive training phase and improves again when athletes recover (15). However, whether this also affects pacing is still not clear. Because it is assumed that pacing strategy is regulated to prevent premature fatigue, overall pacing might be altered by interventions that result in either different amounts of stored energy before exercise or altered substrate use during exercise (35). As already mentioned, because of different competition requirements, various physiological systems are pushed to exhaustion during stage races (1), which might possibly result in accumulated fatigue of various physiological pathways. Furthermore, the physiological pathways that influence pacing (such as neuromuscular recruitment or energy metabolism) require different durations to ensure their recovery (2). Up to now, it is unclear whether this difference in recovery also has an influence on PP. Therefore, the aim of this study was to analyze the influence of training-induced accumulated fatigue on PP and overall performance in a 40-km cycling TT. In addition, effects of a short recovery phase on pacing and performance were examined. It was hypothesized that TT performance decreases after the intensive training phase and increases after 72 h of recovery. Regarding PP, it was hypothesized that participants will be slower in all 4-km splits when fatigued yet without a change in the overall pattern.
Twenty-three well-trained male cyclists volunteered for this study (age, 28.8 ± 7.6 yr; height, 179.9 ± 5.6 cm; weight, 73.7 ± 7.7 kg; training amount, 10,022 ± 4027 km·yr−1). The study was undertaken in accordance with the Declaration of Helsinki and was approved by the regional ethics committee before commencement (Ärztekammer des Saarlandes, Saarbrücken, Germany). Before testing, all participants read and signed an informed consent and provided details on their training/racing history. All cyclists had to undertake a health examination (including resting ECG) to prove eligibility for intense training. Participants had to fulfil the following criteria: at least 5000-km cycling training a year and competition experience at or higher than national level. Participants had to fill out a nutrition and a sleeping diary over the span of the study.
Each participant undertook three self-paced 40-km cycling TT on a Cyclus 2 ergometer (RBM Elektronik-Automation GmbH, Leipzig, Germany). The first TT was carried out before an intensive training period of 6 d (TT1), the second one, after the training phase (TT2). An additional trial was performed after 72 h of recovery (TT3), during which cyclists had to abstain from any type of exercise.
Before TT1, participants were asked to refrain from exercise (>48 h) and to arrive in a fully recovered state. This was controlled for using written training diaries. On the day before and on the TT1 testing day, participants completed a food diary and were instructed to replicate their nutrition as closely as possible before each trial. Trials were conducted at the same time of the day (±1 h) to minimize diurnal variation. Ambient laboratory temperature was between 18.4°C and 22.4°C. All TT were completed on an electromagnetically braked cycle ergometer, which has been shown to produce valid indices of power output (29). A racing bike frame is placed on an electronic brake, and the bicycle chain over the pinion drives the braking mechanism. The axles, to ingest fork and rear, are constructed elastically. Thus, a “natural” cycling movement can be simulated. Participants adjusted the ergometer to their preferred racing position (which was recorded and replicated for each trial), wore their own cycling shoes, and used their own pedals.
Gas exchange parameters (V˙O2 and V˙CO2) were measured continuously using an automated online metabolic cart (Meta Lyzer 3; Cortex, Leipzig, Germany). The gas analyser and flow turbine were calibrated before each test using a certified standard gas (15.0% O2, 5.0% CO2) and ambient air as well as a 3-L syringe for volume calibration (Hans Rudolph, Kansas City, MO). HR was recorded and analyzed using PolarPro Trainer 5 (Polar Electro, Kempele, Finland). Capillary whole blood samples (20 μL) were taken from the hyperemized earlobe and analyzed for BLa concentration (automated enzymatic-amperometric method; Greiner BioChemica, Flacht, Germany). An electrical fan was positioned approximately 0.5 m in front of the ergometer for cooling during each trial. A questionnaire had to be completed by each participant before each TT and before blood samples were taken. Blood samples were drawn in the morning before each TT from an antecubital vein in a supine position. Samples consisted of 2.7-mL EDTA blood and 9-mL whole blood. Whole blood was centrifuged within 20 min after sampling. All hematological measures were taken between 8:00 and 10:00 a.m. in the mornings of testing to avoid circadian variations.
The 6-d training period consisted of two cycling sessions a day, as follows: the first one at 9:00 a.m. and the second one at 2:00 p.m. The morning sessions consisted of either constant high-intensity endurance training (1 h at 95% of the lactate threshold (LT) according to Stegmann et al. (32)) or high-intensity interval training (3 × 5 × 30 s all-out Wingate Sprints) in an alternating manner. In the afternoon sessions, participants had to cycle for 3 h at 80% of their LT (moderate-intensity endurance training). On training day 4, only the afternoon session was conducted. For training prescriptions, an incremental cycling test to voluntary exhaustion was carried out 5 h after TT1 to determine LT. All training sessions were supervised and conducted outside on the participants’ own bicycles. HR was recorded and analyzed using PolarPro Trainer 5 (Polar Electro, Kempele, Finland) to ensure that participants trained at their required intensity. As described in the introduction, various physiological systems might be pushed to exhaustion during several-day stage races (1); therefore, high-volume and high-intensity sessions were chosen to ensure high stress for different metabolic pathways and hence induce multilevel fatigue close to real cycling tours.
Before each trial, participants performed a standardized warm-up (Lamberts and Lambert submaximal cycling test (23)), after which they were given 5 min to relax and prepare themselves for the subsequent TT. A flat 40-km TT profile was created using the Cyclus2 software and was used for all trials. Participants were instructed to complete the 40-km distance as fast as possible. Because several studies showed that the absence of feedback negatively influences PP (3,24), visual feedback on the distance covered, power, pedaling frequency, and HR was available during all trials. Furthermore, conditions were chosen to resemble real competition as closely as possible to ensure external validity. The same range of electronic gear ratios was used for each trial, and the participant started each trial with the same gear ratio; they were permitted to adjust this throughout the trial to realize their preferred cadence. At 4-km intervals, participants were asked to report their RPE (7), capillary whole blood samples were obtained, and HR values were recorded. V˙O2, V˙CO2 and RER were measured during the whole trial. Unfortunately, because of inconsistencies of measurements, gas exchange parameters are only available from 18 of the 23 participants. Because it has recently been shown that the presence of a male or female observer has a significant influence on RPE (39), participants were tested by a person of the same gender on every test day.
Venous blood measures
From serum analysis, the following parameters were determined: creatine kinase (CK, IFCC enzymatic test) and urea (urea-GLDH enzymatic-UV-test) (all by UniCell DX 600; Beckman Coulter, Krefeld, Germany).
Recovery–Stress Questionnaire for Athletes (REST-Q Sport)
This questionnaire consists of 52 items (REST-Q-52), which can be assigned to 19 subscales and enables a detailed multidimensional analysis of the individual recovery–stress state. The items have to be self-rated on a 7-point Likert scale ranging from 0 (never) to 6 (always), indicating how often the subject has participated in various activities during the last 72 h. The 52 items can be assigned to 10 stress-associated subscales (summed up to a total stress score) and nine recovery-related subscales (total recovery score). The questionnaire shows satisfying reliability and convergent validity (21) and has been used in several studies on fatigue in athletes. Participants were asked to complete the REST-Q at the beginning of all testing sessions before blood measures and TT. For a detailed review of the REST-Q Sport, the reader is referred to the study of Kellmann (20).
Data are presented as means and SD. A one-way repeated-measures ANOVA was performed to examine differences between final times and mean physiological (HR, BLa, V˙O2, and RER) and perceptual responses (RPE) between trials. Furthermore, a one-way repeated-measures ANOVA was used for the comparison of laboratory parameters (CK, urea) and REST-Q Sport scores over time. To compare pacing, velocity in all trials was expressed relative to average race velocity (normalized mean velocity). This approach of expressing pacing as the difference between current and overall mean velocity is well accepted (2). A repeated-measures ANOVA was used to compare pacing (normalized velocity and power output) and pattern of physiological/perceptual responses between trials (factor 1, test day; factor 2, 4-km intervals). When significant main effects were observed, a Scheffé post hoc test was performed. P < 0.05 for the α error was accepted as the level of significance for statistical comparisons. Magnitude-based inferences were also conducted to determine the smallest worthwhile difference in overall performances between trials (18). This approach represents a contemporary method of data analysis that uses confidence intervals to calculate the probability that a difference is practically meaningful. The smallest worthwhile difference was set as 0.3 of the typical variation in an athletes’ performance (1.0% (16)). Where the chance of benefit and harm were both calculated, a qualitative descriptor was assigned to the following quantitative chances of performance effect: 0.5%–5%, very unlikely; 5%–25%, unlikely; 25%–75%, possibly; 75%–95%, likely; 95%–99.5%, very likely; >99.5%, most likely (6). Cohen d effect sizes and thresholds (0.2, 0.6, 1.2, 2.0, and 4.0 for trivial, small, moderate, large, very large, and extremely large (17) were also used to compare the magnitude of the difference in overall performance time.
40-km TT overall performance
Time to completion increased from TT1 to TT2 by 1.8% (±3.0%) and decreased from TT2 to TT3 by 2.0% (±2.7%; P < 0.05; F2,44 = 6.4; η2 = 0.23) (Table 1). TT1 and TT3 showed no significant difference (P > 0.05). Mean and individual times to completion are shown in Figure 1. These changes in time to completion are reflected in mean power output during the TT, which was 4.1% (±7.3%) higher in TT1 and 5.9% (±7.5%) higher in TT3 versus TT2 (P < 0.05; F2,44 = 7.5; η2 = 0.25) (Table 1). Corresponding analysis of magnitude-based differences for time to completion showed that TT2 was very likely to result in worse performance when compared with TT1 whereas TT3 was most likely to result in better performance when compared with TT2 (Table 1).
Training, laboratory, and REST-Q results
Mean training load over the 6 d was as follows: for 95% LT, HR, 143.1 ± 11.1 bpm; RPE, 7.3 ± 1.7; session RPE, 701.9 ± 171.4; for 3 × 5 × 30 s all-out, HR, 179.3 ± 16.8 bpm; RPE, 7.6 ± 1.6; session RPE, 611.1 ± 141.1; for 80% LT, HR, 139.4 ± 10.9; RPE, 7.6 ± 1.5; session RPE, 1165.3 ± 330.4. Participants were in their required HR zones for 63.2% of the time during the 95% LT sessions and for 45.2% during the 80% LT sessions, respectively. In the first three training sessions, HR zones were reached for 70% of the time in both sessions. CK and urea increased significantly from TT1 to TT2 (P < 0.05) and decreased from TT2 to TT3 (P < 0.05) (Table 2). Furthermore, total stress score of the REST-Q increased significantly from TT1 to TT2 (P < 0.001) (Table 2). Correspondingly, the total recovery score was significantly lower in TT2 compared with that in TT1 (P < 0.05) (Table 2). Both scores were back to baseline on the day of TT3 (TT1 vs TT3, P > 0.05) (Table 2).
Normalized PP for the comparison of all three trials are shown in Figure 2A. PP was significantly different in TT2 compared with that in TT1 (P < 0.001) and TT3 (P < 0.01; F18,396 = 3.3E+00; η2 = 0.13). The first 4 km were significantly slower in TT2 (P < 0.001), whereas the increase in velocity between kilometers 36 and 40 was significantly greater compared with both other trials (P < 0.05). No difference in the pattern could be observed between TT1 and TT3 (P > 0.05). The corresponding power output is displayed in Figure 2B, showing a significantly lower power output in each stage during TT2 compared with that in both of the other trials (test effect, P < 0.001, F2,44 = 8.2; η2 = 0.27). Similar to the normalized pacing profile, the pattern of power output was significantly different in TT2 compared with that in TT1 and TT3 (P < 0.001; F18,396 = 3.7; η2 = 0.14) (Fig. 2B), revealing lower power outputs in the first 4 km of the trial and higher increase in power from kilometers 36 to 40 (P < 0.001).
Within-trial effects showed that the first and last 4 km were significantly faster than the middle part of the trial in TT1 (kilometers 16 to 32, P < 0.05). In TT2, the last 4 km were significantly faster than all other 4-km sections (P < 0.001), whereas in TT3, the last section was faster than all 4-km parts between 12 and 36 (P < 0.01).
Physiological and perceptual response during TT
Mean and maximum HR, BLa, V˙O2, and RER values are shown in Table 1. RER, BLa (Fig. 3A and B) and HR (Fig. 4A) were significantly lower in TT2 compared with those in TT1 and TT3 (P < 0.001). RPE showed no difference between trials over the whole race (P > 0.05) (Fig. 4B).
The purpose of the present study was to examine the influence of accumulated fatigue on the PP of trained cyclists in a 40 km TT. The new finding of this study was that fatigue seems to reversibly change PP at the beginning and end of the race. Furthermore, overall physiological responses (HR, BLa, and RER) were lower during the fatigued TT, yet perceived exertion remained the same. Participants adopted a parabolic-shaped pattern in both recovered conditions, whereas in the fatigued condition, the pattern was even from the beginning with a greater endspurt in the last 4 km (slow–fast pacing (2)). This is in contrast to recently published findings by De Morree and Marcora (9) who observed reduction in overall 15-min TT performance, without a change in PP after a muscle-fatiguing exercise before the trial. Hence, accumulated fatigue seems to exert a stronger effect on PP than acute locomotor muscle fatigue. This may be due to the multifaceted fatigue induced in the current study, leading to exhaustion of several metabolic and cardiovascular pathways. Athletes showed the same PP after 2 d of recovery as in the first TT. Hence, as previously suggested, pacing seems to be the result of afferent feedback and signaling alterations, thereby allowing higher power output to be maintained when glycogen stores are loaded (28,36). Furthermore, the reestablished PP displays that the change in the fatigued trial can be directly referred to the effects of accumulated fatigue.
Interestingly, RPE patterns were the same during all three conditions, even though power output and physiological parameters are lower. De Morree and Marcora (9) recently proposed that self-paced performance can be predicted by a psychobiological model, which is based on the motivational intensity theory (40). This model predicts that fatigued participants decide to reduce their pace so that RPE does not reach its maximum before the end of the TT (9,40). The results of the present study are consistent with this assumption because RPE reached its maximum at the end of each TT and its course showed no difference between trials. Furthermore, the endspurt was greater in the fatigued TT. Because finishing the race is paramount, athletes seem to have chosen a slightly conservative pace for most of the trial. However, near the end when the risk of not finishing the race is negligible, participants significantly increase their power output.
In this regard, it has recently been suggested that pacing should be considered as a neural “buffering” process in the distribution of effort to prevent premature physical exhaustion (11,31). This buffering process may avoid the necessity of discontinuing an exercise bout before its scheduled finish by reducing power output at the beginning of the race. Because finishing the 40-km TT was the main goal of the cyclists in the current study, it can be assumed that they consciously downregulated power output until the “finish line” was so close that premature fatigue was very unlikely. Interestingly, participants’ power output was already reduced in the first minute of the fatigued TT. Therefore, the current findings suggest that fatigued athletes reduced their starting power as a combination of anticipation and peripheral feedback to assure that the exercise bout can be completed without premature fatigue (36).
HR, BLa, and RER during the TT were significantly reduced after 6 d of hard training. This is in accordance with several studies reporting reduced capability to activate the sympathetic nervous system after intensive training phases, leading to disturbed glycolytic energy mobilization and cardiovascular responses (12,14,19). The reduced lactate and RER values might indicate a shift of the energy-supplying process in favor of an increased fat and decreased CHO use (19). Reduced glycogenolysis and glucose transport in the muscle and the liver might lead to lower glycolytic activity (19) and, hence, reduced lactate and RER values. However, as described previously, cyclists reduced their power output to prevent premature fatigue; hence, it might also be possible that they underestimated their ability in the second TT and consequently were unable to achieve similar values as those in the first trial.
The lower HR values could also be the result of locomotor muscle fatigue leading to reduced power output achieved with maximal effort (25,37). A previous study by Amann and Dempsey (4) showed that locomotor muscle fatigue has a significant influence on power output during a 5-km cycling TT (4). The authors suggested that muscle fatigue plays a crucial role in exercise performance and that peripheral power output seems to be a carefully regulated variable (4). Therefore, it might be that participants could not cope with exercise of near-maximal intensity because of local muscle fatigue, which might lead to lower HR during the second TT (19). After 2 d of recovery, BLa and RER were similar as those in the first TT, assuming full restoration of the glycogen stores. Thus, HR values were still significantly lower than before the training phase, which could be due to a positive effect of the training phase or insufficient recovery, because various biological systems require different times to recover. However, a possibly insufficient recovery of the cardiovascular system did not influence reestablishing of PP and overall performance.
Even though it has been stated that the assessment of a decrement in sports-specific performance represents the gold standard of measuring short-term fatigue in athletes (22), not all participants showed worse TT performance after 6 d of intensive training in the current study. When taking into account the known day-to-day variability of 40-km cycling TT (average CV of about 1%) (17), only 10 of the 23 cyclists showed meaningful decrease in overall performance above the CV. However, because laboratory parameters and questionnaire results resembled comparable fatigue in all participants, analysis was conducted for the whole sample. In this regard, it might be speculated that fatigue-related performance decrements are delayed in comparison with other measurements, which leads to the assumption that sports-specific performance might not be the gold standard to measure short-term fatigue and recovery in athletes. A reason for the small changes in some of the cyclists could also be a familiarization effect in the second TT. Hence, pacing might not have been optimal in TT1 in all cyclists. This is a limitation of the study because variability of TT performance is lower between a second and third trial compared with that between the first and the second one (33,34). However, because national-level cyclists were recruited, adding a fourth trial would not have been feasible because of logistical constraints.
The results of this study show for the first time that cyclists reversibly change their PP during a 40-km TT because of accumulated fatigue. PP changed at the beginning and the end of the race, which is compatible with the model that pacing includes a combination of anticipation and feedback mechanisms. Participants reduced their power output until premature exhaustion seemed very unlikely. The slower pace at the beginning of the trial may also be a response to compensate for the increase in perception of exertion induced by fatigue. According to the psychobiological model, fatigued participants seem to reduce their power output so that RPE does not reach its maximum before the end of the TT. As described earlier, different cycling competition requirements result in exhaustion of various physiological systems (1), which might lead to a comparable accumulation of fatigue during several-day races and, hence, similar effects on pacing and overall performance.
The present study was initiated and funded by the German Federal Institute of Sport Science. The research was realized within RegMan—Optimization of Training and Competition: Management of Regeneration in Elite Sports (IIA1-081901/12-16).
The authors declare no conflicts of interest.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Abbiss CR, Laursen PB. Models to explain fatigue during prolonged endurance cycling. Sports Med
. 2005; 35 (10): 865–98.
2. Abbiss CR, Laursen PB. Describing and understanding pacing strategies during athletic competition. Sports Med
. 2008; 38 (3): 239–52.
3. Albertus Y, Tucker R, St Clair Gibson A, Lambert EV, Hampson DB, Noakes TD. Effect of distance feedback on pacing strategy and perceived exertion during cycling. Med Sci Sports Exerc
. 2005; 37 (3): 461–8.
4. Amann M, Dempsey JA. Locomotor muscle fatigue modifies central motor drive in healthy humans and imposes a limitation to exercise performance. J Physiol
. 2008; 586 (1): 161–73.
5. Atkinson G, Wilson D, Eubank M. Effects of music on work-rate distribution during a cycling time trial. Int J Sports Med
. 2004; 25 (8): 611–5.
6. Batterham AM, Hopkins WG. Making meaningful inferences about magnitudes. Int J Sports Physiol Perform
. 2006; 1 (1): 50–7.
7. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc
. 1982; 14 (5): 377–81.
8. Corbett J, Barwood MJ, Ouzounoglou A, Thelwell R, Dicks M. Influence of competition on performance and pacing during cycling exercise. Med Sci Sports Exerc
. 2012; 44 (3): 509–15.
9. de Morree HM, Marcora SM. Effects of isolated locomotor muscle fatigue on pacing and time trial performance. Eur J Appl Physiol
. 2013; 113 (9): 2371–80.
10. Dugas JP, Oosthuizen U, Tucker R, Noakes TD. Rates of fluid ingestion alter pacing but not thermoregulatory responses during prolonged exercise in hot and humid conditions with appropriate convective cooling. Eur J Appl Physiol
. 2009; 105 (1): 69–80.
11. Edwards AM, Polman RC. Pacing and awareness: brain regulation of physical activity. Sports Med
. 2013; 43 (1)1: 1057–64.
12. Faude O, Meyer T, Urhausen A, Kindermann W. Recovery training in cyclists: ergometric, hormonal and psychometric findings. Scand J Med Sci Sports
. 2009; 19 (3): 433–41.
13. Foster C, Schrager M, Snyder AC, Thompson NN. Pacing strategy and athletic performance. Sports Med
. 1994; 17 (2): 77–85.
14. Halson SL, Bridge MW, Meeusen R, et al. Time course of performance changes and fatigue markers during intensified training in trained cyclists. J Appl Physiol (1985)
. 2002; 93 (3): 947–56.
15. Halson SL, Jeukendrup AE. Does overtraining exist? An analysis of overreaching and overtraining research. Sports Med
. 2004; 34 (14): 967–81.
16. Hopkins WG. A new view of statistics. 1997–2004 [updated 1997–2004; cited 14 Nov 2013]. Available from: http://sportsci.org/resource/stats/index.html
17. Hopkins WG. Measures of reliability in sports medicine and science. Sports Med
. 2000; 30 (1): 1–15.
18. Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc
. 2009; 41 (1): 3–13.
19. Jeukendrup AE, Hesselink MK, Snyder AC, Kuipers H, Keizer HA. Physiological changes in male competitive cyclists after two weeks of intensified training. Int J Sports Med
. 1992; 13 (7): 534–41.
20. Kellmann M. Preventing overtraining in athletes in high-intensity sports and stress/recovery monitoring. Scand J Med Sci Sports
. 2010; 20 (2 Suppl): 95–102.
21. Kellmann M, Kallus KW. Mood, recovery-stress state, and regeneration. In: Lehmann M, Foster C, Gastmann U, Keizer H, Steinacker JM, editors. Overload, Fatigue, Performance Incompetence, and Regeneration in Sport
. New York (NY): Plenum; 1999. p. 101–17.
22. Lamberts RP. Predicting cycling performance in trained to elite male and female cyclists. Int J Sports Physiol Perform
. 2014; 9 (4): 610–4.
23. Lamberts RP, Swart J, Noakes TD, Lambert MI. A novel submaximal cycle test to monitor fatigue and predict cycling performance. Br J Sports Med
. 2011; 45 (10): 797–804.
24. Mauger AR, Jones AM, Williams CA. Influence of feedback and prior experience on pacing during a 4-km cycle time trial. Med Sci Sports Exerc
. 2009; 41 (2): 451–8.
25. Meeusen R, Duclos M, Foster C, et al. Prevention, diagnosis, and treatment of the overtraining syndrome: joint consensus statement of the European College of Sport Science and the American College of Sports Medicine. Med Sci Sports Exerc
. 2013; 45 (1): 186–205.
26. Mujika I, Padilla S. Physiological and performance characteristics of male professional road cyclists. Sports Med
. 2001; 31 (7): 479–87.
27. Peiffer JJ, Abbiss CR. Influence of environmental temperature on 40 km cycling time-trial performance. Int J Sports Physiol Perform
. 2011; 6 (2): 208–20.
28. Rauch HG, St Clair Gibson A, Lambert EV, Noakes TD. A signalling role for muscle glycogen in the regulation of pace during prolonged exercise. Br J Sports Med
. 2005; 39 (1): 34–8.
29. Reiser M, Meyer T, Kindermann W, Daugs R. Transferability of workload measurements between three different types of ergometer. Eur J Appl Physiol
. 2000; 82 (3): 245–9.
30. Skorski S, Faude O, Caviezel S, Meyer T. Reproducibility of pacing profiles in elite swimmers. Int J Sports Physiol Perform
. 2014; 9 (2): 217–25.
31. St. Clair Gibson A, Noakes TD. Evidence for complex system integration and dynamic neural regulation of skeletal muscle recruitment during exercise in humans. Br J Sports Med
. 2004; 38 (6): 797–806.
32. Stegmann H, Kindermann W, Schnabel A. Lactate kinetics and individual anaerobic threshold. Int J Sports Med
. 1981; 2 (3): 160–5.
33. Stone MR, Thomas K, Wilkinson M, St Clair Gibson A, Thompson KG. Consistency of perceptual and metabolic responses to a laboratory-based simulated 4,000-m cycling time trial. Eur J Appl Physiol
. 2011; 111 (8): 1807–13.
34. Thomas K, Stone MR, Thompson KG, St Clair Gibson A, Ansley L. Reproducibility of pacing strategy during simulated 20-km cycling time trials in well-trained cyclists. Eur J Appl Physiol
. 2012; 112 (1): 223–9.
35. Tucker R. The anticipatory regulation of performance: the physiological basis for pacing strategies and the development of a perception-based model for exercise performance. Br J Sports Med
. 2009; 43 (6): 392–400.
36. Tucker R, Noakes TD. The physiological regulation of pacing strategy during exercise: a critical review. Br J Sports Med
. 2009; 43 (6): 1–9.
37. Urhausen A, Kindermann W. Diagnosis of overtraining: what tools do we have? Sports Med
. 2002; 32 (2): 95–102.
38. Wiles JD, Coleman D, Tegerdine M, Swaine IL. The effects of caffeine ingestion on performance time, speed and power during a laboratory-based 1 km cycling time-trial. J Sports Sci
. 2006; 24 (11): 1165–71.
39. Winchester R, Turner LA, Thomas K, et al. Observer effects on the rating of perceived exertion and affect during exercise in recreationally active males. Percept Mot Skills
. 2012; 115 (1): 213–27.
40. Wright RA. Refining the prediction of effort: Brehm’s distinction between potential motivation and motivation intensity. Soc Pers Psychol Compass
. 2008; 2 (2): 682–701.