Fatigue in soccer players occurs not only toward the end of the match but also temporarily during the game and is manifested as a reduction in repeated sprinting ability (RSA) (23,29). The RSA is an important determinant of soccer performance not only because players perform a number of repeated bouts during a match but also because it can be used as a good predictor of high-intensity performance during a match (36). Training strategies to improve RSA include performance of maximal or near maximal sprints with short rests (30) and interval endurance training, because aerobic fitness is a significant determinant of fatigue resistance in this type of exercise (5).
Recent studies have suggested that alternative methods, such as resistance training with moderate loads (50-70% of maximal strength with 10-20 repetitions), are effective in reducing fatigue during RSA (11,16). However, the use of strength training to improve RSA has not been examined in detail. Most studies have focused on the manipulation of the rest interval between sets (16,37), while still employing low to moderate loads and 10-20 repetitions, which are typical of hypertrophy or muscle endurance resistance training (22). The physiological basis for this improvement has not been clarified, but increases in muscle oxidative capacity and capillary density (7) and improved H+ regulation and increased muscle strength per se may be involved (11).
On the other hand, it has been shown that low volume, high-intensity resistance training with 4 sets × 4-6 repetitions of half-squat with 85-90% of 1 repetition maximum (1RM) improves not only strength, speed, and power but also running economy (RE) in soccer players (17). The fact that a parameter related to endurance performance is improved by a typical strength training program may have implications for performance maintenance during RSA. In contrast with the moderate to high repetition protocols previously used to improve RSA, this type of low repetition strength training mainly induces neural adaptations, such as improved motor unit coordination, with little or no hypertrophy (1,12).
To our knowledge, no study has directly compared the effects of very high load, low volume resistance training with a more traditional muscle hypertrophy endurance regimen on RSA. Thus, the aim of this study was to compare these 2 different resistance training protocols in a realistic environment, that is, during the course of preseason training, in professional soccer players. It was hypothesized that the high load, low volume resistance training program would reduce fatigue during RSA to a greater extent than the traditional muscle hypertrophy endurance regimen.
Experimental Approach to the Problem
The effects of 2 different resistance training protocols consisting of half-squats on RSA and aerobic performance were examined during 6 weeks of preseason training. A 2-week period of preconditioning and familiarization with all the tests and training procedures preceded the main 6-week training period. The following dependent variables were measured within a period of 5 days, both before (pretraining) and after training (posttraining): lean leg volume (LLV), peak and mean power output (MPO) during an RSA test, half-squat strength (1RM), maximal oxygen uptake (V̇O2max), ventilatory threshold (VT), and RE during a treadmill running test, and soccer-specific aerobic field tests. The independent variables were the 2 resistance training protocols. The pre and post-training tests were performed at the same time of the day, whereas diet was recorded for 2 days before each test during the pretraining period and replicated for the 2 days preceding each test at the posttraining period. After the pretraining tests, players were divided into 2 equal groups (n = 10) matched for pretraining cycling power output and V̇O2max.
The 2 groups followed the same training program (Table 1) involving 10-12 morning and afternoon sessions per week (mean session duration: 73 ± 3 minutes) and 1 day of complete rest. Resistance training was performed 3 times per week and included 8-12 upper and lower body exercises. The only difference in training between the 2 groups was in the load and repetitions for the half-squat exercise. One group performed 4 sets of 5 repetitions of half-squat aiming to increase maximal strength (S-group). The load was equal to 90% of 1RM, and sets were separated by 3-minute rest intervals. Emphasis was placed on maximal mobilization during the concentric action (1). The other group performed 4 sets of 12 repetitions of half-squat against a resistance equal to 70% of 1RM, emphasizing both the eccentric and concentric actions with controlled movement speed and an interval of 1.5 minutes between sets. This scheme was expected to induce greater muscle hypertrophy (H-group) and strength increases. To maintain an optimum training stimulus, training load was readjusted at the end of 3 weeks of training for both groups, by performing a 5RM test (S-group) or a 12RM test (H-group).
Twenty male professional soccer players (age: 22.3 ± 1.1 years, body mass: 75.4 ± 2.0 kg, height: 179.5 ± 1.5 cm) volunteered to participate in the study. Written Informed consent was obtained from each subject after they were informed about the procedures, the risks involved, and their right to terminate participation at will. This research was approved by the University of Athens Ethics Committee and all procedures conformed to the ethical conduct set forth by the World Medical Association and the specific policies of the American College of Sports Medicine with regards to human experimentation.
Anthropometric and Half-Squat Strength Measurements
Body fat was estimated from 7 skinfolds (18), and LLV was determined using a previously validated anthropometric method (20). According to this method, the volume of the leg was partitioned into 6 segments, which were similar to truncated cones by measuring 7 circumferences and heights above the floor at specific sites with the subjects standing (20). Skinfold thicknesses were measured with a Harpenden fat caliper at 4 sites, that is, the anterior and posterior thigh in the midline at the one-third subischial height level and the medial and lateral calf at the maximal calf circumference. Because the calipers pick up a double layer of skinfold tissue under a pressure of 10 g mm−2, the reading is converted to a true single measurement using a regression equation applicable to each sex and fat site. This was based on a comparison between x-ray fat and caliper fat using the linear relationship, which we have found exists between the 2 (20). The subcutaneous fat volume was subtracted from the total leg volume to obtain LLV. Test-retest reliability for all the dependent variables measured in this investigation was determined in separate experiments by calculating the intraclass correlation coefficient (ICC) using a 2-way mixed model. The ICC for LLV measurement was 0.96.
The players were familiar with half-squats as part of their regular strength training programs. A 1RM test on a Smith machine was performed to determine maximal half-squat strength (90° knee angle measured with a digital goniometer). After a general warm-up (10 minutes of low-intensity treadmill running), 8 repetitions with an estimated 50% of 1RM were preformed, using each subject's previous training experience and after 1 minute of rest, 3 repetitions with an estimated 70% of 1RM were performed (34). After 3 minutes, subsequent trials were performed for 1 repetition with progressively heavier weights until the 1RM was determined within 3 attempts, using 3- to 5-minute rest periods between trials (27). The ICC for the half-squat was 0.99.
Force-Velocity and Repeated Sprinting AbilityTest
Sprints were performed on a modified friction-loaded cycle ergometer (Monark, model 864, Monark Exercise AB, Varberg, Sweden) that was equipped with a photocell. Flywheel velocity was sampled at 200 Hz (Biopac Systems Inc., Goleta, CA, USA) and instantaneous power output was averaged over each pedal downstroke after being corrected for changes in kinetic energy of the flywheel (2,24).
After full familiarization with maximal sprinting on a cycle ergometer, subjects performed a force-velocity test consisting of 4 6-second maximal sprints from a standing start against resistive loads ranging from 2.5 to 10% of body mass in random order with full recovery (>4 minutes). The individual power vs. pedal rate relationships were derived from the power and pedal rate data during the acceleration phase of the 4 sprints of the force-velocity test (2). The force-pedal rate data were fitted using linear least squares regression, whereas a quadratic polynomial was fitted the power-pedal rate data. These equations were used to calculate the optimal pedal rate, that is, the rate at which peak power is generated (Vopt) and the corresponding optimal resistive load (Fopt).
On a separate occasion, subjects performed the RSA test, which consisted of 10 maximal 6-seconds sprints separated by 24 seconds of passive recovery against a resistive load corresponding to 60% of Fopt. Strong verbal encouragement was given during each sprint. Every sprint session was preceded by a standardized warm-up consisting of 5-minute cycling at 35 W (pedal rate 70 rpm), followed by 2 30-second cycling bouts at 50 and 60 W (pedal rate 100 and 120 rpm, respectively). A 5-minute stretching period separated the warm-up from the first sprint. The highest power output generated over a pedal stroke was defined as peak power output (PPO), while MPO for each sprint was also recorded. Power output was expressed in relative values (watts per liter of LLV). The ICC for PPO and MPO over the 10 sprints ranged from 0.990 to 0.996. The RSA was assessed by calculating the total work in absolute units (kJ) during the 10 sprints, and during the first (sprints 1-5) and the last half of the test (sprints 6-10). Also, the relative decrement of peak and mean power over the 10 sprints (%dec) was calculated as follows:
where total power output = sum of power outputs from all sprints and ideal power output = highest power output × number of sprints performed.
This method has been shown to be the most valid and reliable measure of fatigue during maximal sprint exercise (13). The ICC for %dec for peak and mean power was 0.91 and 0.94.
Aerobic Performance Tests
V̇O2max was measured during an incremental treadmill running test to exhaustion, where running speed was increased by 0.5 km·h−1 every minute, starting from a speed of 10-12 km·h−1 V̇O2max was defined as the highest 30-second average of V̇O2 during the test if at least 2 of the following criteria were satisfied: (a) a plateau of V̇O2 (<2-ml·kg−1·min−1 increase) despite an increase of treadmill speed, (b) inability to maintain the running speed, (c) a respiratory exchange ratio >1.1, (d) heart rate above 95% of the predicted maximum. Measures of gas exchange were obtained by an open-circuit automatic spirometric system (Cosmed Quark b2, Pavona di Albano - Rome, Italy). The VT was determined by the computer software using the V-slope method (3) and verified by experienced researchers. The coefficient of variation for the detection of the thresholds was <3%. Heart rate was continuously recorded during the test (Polar, S610i, Polar Electro Oy, Kempele, Finland). The RE was assessed as the oxygen uptake at the speed corresponding to the individual pretraining (VT) and was expressed as relative oxygen uptake per kg body mass per km (ml·kg−1·min−1). The ICC for V̇O2max, VT, and RE were 0.94, 0.93, and 0.92, respectively. Aerobic fitness was also assessed using the Yo-Yo intermittent endurance test and Hoff's dribbling track test (DTT) as modified by Chamari et al. (9). In the modified DDT test, players were asked to cover the maximum distance by dribbling the ball through a specific track of 290 m per tour in a period of 10 minutes. The ICCs for the Yo-Yo and the DTT tests 0.96 and 0.94, respectively.
Changes in LLV, 1RM, and aerobic fitness parameters and field tests were analyzed using 2-way analysis of variance with repeated measures on 1 factor (pre-posttraining). The PPO and MPO over the 10 sprints were analyzed separately for each group, using 2-way analysis of variance with repeated measures on both factors (pre-posttraining and sprint number). When significant effects were found (p ≤ 0.05), differences were located using a Tukey post hoc test.
Effect size for main effects and interaction was estimated by calculating partial eta squared (η2p) values using the SPSS v. 15 statistical package. Effect size for pairwise comparisons was assessed with Cohen's d using the pooled SD of the 2 means compared. Effect sizes were classified as small (0.2), medium (0.5), and large (>0.8).
Results are presented as mean ± SE for 10 (H-group) and 9 (S-group) individuals, because 1 player did not complete all posttraining measurements because of an injury that was unrelated to the training and testing procedures.
Body mass remained unchanged after training, and % body fat was decreased in a similar fashion in both groups (H-group: from 9.3 ± 0.9 to 8.2 ± 0.7% and S-group: from 7.5 ± 0.8 to 6.8 ± 0.8%; main effect pre-post, p < 0.01, η2p = 0.55; no group vs. pre-post interaction). Maximal half-squat strength increased significantly in both groups (H-group: from 142 ± 3 to 158 ± 5 kg and S-group: from 152 ± 4 to 179 ± 4 kg; main effect pre-post, p < 0.01, η2p = 0.89). However, the increase in 1RM in absolute values (kg) was 58% higher in the S- compared to the H-group (17.3 ± 1.9 vs. 11.0 ± 1.9%, p < 0.05, d = 1.20). The LLV increased significantly only in the H-group (from 7.77 ± 0.38 to 8.10 ± 0.41 L, group vs. pre-post interaction, p < 0.05, η2p = 0.24), whereas it remained unchanged in the S-group (7.61 ± 0.37 to 7.66 ± 0.30 L). Thus, the increase in 1RM per LLV was >twofold higher in the S-group compared to the H-group (16.3 ± 2.3 vs. 6.6 ± 1.7%, p < 0.01, d = 1.42).
The PPO expressed in absolute values (W) was increased in both groups after training only after sprint 5 (sprint vs. pre-post interaction, p < 0.05, η2p = 0.26 for H-group and η2p = 0.22 for S-group), whereas PPO from sprint 1 up to sprint 5 remained unchanged. The PPO expressed relative to LLV and the percent PPO decrement over the 10 sprints remained unchanged after training (Table 2). In contrast, total work over the 10 sprints was improved after training in both groups confirming a significant increase in RSA (H-group: from 44.7 ± 1.1 to 46.7 ± 0.9 kJ and S-group: from 44.1 ± 1.8 to 46.5 ± 1.8 kJ; main effect pre-post, p < 0.01, η2p = 0.52, no group vs. pre-post interaction). When the RSA test was divided into the first (sprints 1-5) and the second half (sprints 6-10), a significant interaction was found only in the S-group (first-second half of test vs. pre-post interaction, p < 0.01, η2p = 0.70, Figure 1), indicating a greater increase in total work in the second half (8.9 ± 2.6%) compared with the first part of the sprint test (3.2 ± 1.7%).
When power output was expressed per liter LLV, there was no change in MPO in the H-group over the 10 sprints (Figure 2). In contrast, a significant sprint vs. pre-posttraining interaction (p < 0.01, η2p = 0.45) was found in the S-group, showing a better maintenance of MPO per liter LLV during the last 6 sprints posttraining (Figure 2).
The results of the laboratory and field tests of aerobic fitness are presented in Table 3. V̇O2max was similar before training and increased significantly in both groups after training, whereas VT expressed as a percentage of V̇O2max was also not significantly altered posttraining (Table 3). Speed at VT tended to improve more in the S-group compared with the H-group (by 14.3 ± 3.9 vs. 7.9 ± 1.4%, p = 0.12, d = 0.74). However, RE significantly improved by 10.9 ± 2.1% only in the S-group (p < 0.01), whereas the corresponding improvement in the H-group (4.2 ± 2.6%) was not statistically significant (Figure 3). Running speed at V̇O2max increased similarly in both groups (6.7 ± 1.3%, p < 0.01, Table 3). Field performance in the DTT test improved similarly in both groups after the training period (∼10%), whereas the percent improvement in the distance covered during the Yo-Yo test tended to be higher in the S-group than in the H-group (H-group: 23.5 ± 4.0% vs. S-group: 33.2 ± 3.0%, p = 0.067, d = 0.62, Table 3).
The main finding of this study was that performance of half-squats with high load and a low number of repetitions (S-group) during preseason soccer training resulted in a greater increase in leg strength, RSA, and RE, than using a moderate load and number of repetitions protocol (H-group). Adaptations in the S-group were not accompanied by significant muscle hypertrophy, as indicated by LLV increases only in the H-group. An improvement in maximal leg strength with minimal or no increase in leg muscle mass has been previously reported for untrained individuals and for soccer players using a training protocol with the same characteristics as in the S-group (17,25). Also, this training protocol has been shown to improve RE in soccer players, without an increase in V̇O2max or lactate threshold (17). However, this is the first study that examined and compared the effects of this type of strength training (S-group) with a moderate intensity and higher repetitions protocol (H-group) on RSA, in a realistic environment—during the preseason training period in professional soccer players.
The few studies that explored the effects of resistance training on RSA have used only moderate (5 sets × 10 repetitions) (37) or high repetition schemes (3-5 sets × 15-20 repetitions) (11,16). They reported increases in total work between 5.4 and 12.5% using different repeated sprints protocols, ranging from several sprints with long rest (15 × 5-second sprints with 50-second rest) (37) to fewer sprints with short rest (5 × 6-second sprints with 20- or 80-second rest) (11,16). The increase in total work in the H-group (4.6%) falls marginally below this range, and this may be because of the higher initial fitness level of the subjects of this study (professional soccer players) compared to recreationally active men and women who took part in the studies mentioned above. Several adaptations may explain the increase in RSA after resistance training with moderate to high repetitions, including local aerobic adaptations such as increased capillary supply and muscle oxidative potential (7,14), reduced intramuscular H+ accumulation (11) and improved muscle buffer capacity (6). Part of the improvement in RSA in the H-group may also be attributed to the increased V̇O2max (5), which was possibly the result of the standard preseason training program. However, the improvement in RSA in the S-group was greater than that in the H-group, despite the similar increase in V̇O2max in H- and S-groups (Table 3). Moreover, the nature of the resistance training protocol in the S-group (high resistance, low number of repetitions, and long rest intervals) would result in a smaller degree of local aerobic muscle adaptations compared with the H-group (7). Therefore, the greater improvement in RSA in the S-group cannot be attributed to greater local aerobic adaptations in the muscles involved and suggests that other mechanisms, possibly of neural origin, may be responsible for the difference in RSA between the 2 groups.
A typical response to the training protocol followed by the S-group is the increase in maximal strength (17). This is a result of neural adaptations to the high load protocol that involves changes in the pattern of motor unit recruitment and increased rate coding (8,12). One of the key characteristics of the S-protocol is the rapid action during the concentric phase in addition to the high resistance (4). In most muscles, the upper limit of motor unit recruitment is about 85% of the maximal force and an increase in force beyond that is accomplished entirely by rate coding (10). This would imply that the load used by the S-group (90% of 1RM) was an appropriate stimulus for neural adaptations, while the lower resistance in the H-group (70% of 1RM) was effective mainly for increasing strength through muscle hypertrophy (7). However, the greater increase in maximal strength in the S-group compared with the H-group cannot directly explain the greater improvement in RSA in the S-group, because this was observed in the second and not in the first half of the test (sprints 6-10). The possible role of neural adaptations for the greater improvement in fatigue resistance in the S-group is highlighted when power output in each of the 10 sprints is expressed in relative values (Figure 2). In this case, power output was higher after training during the last 6 sprints in the S-group, whereas no improvement was seen in the H-group. This was because of the significant increase in LLV in the H-group, indicating that the improvement in power output was simply a result of a proportional increase in leg muscle mass. In contrast, the increased power output per unit of LLV in the S-group would suggest qualitative changes in the muscles involved, possibly implying neural adaptations (1).
Recent studies have shown that part of the fatigue observed during a repeated sprint cycling protocol (10 × 6-second sprints) is because of neuromuscular and metabolic factors (28,35). Mendez-Villanueva et al. (28) measured changes in electromyographic (EMG) activity of the vastus lateralis muscle during 10 × 6-second sprints interspersed by 30-second rest intervals. When compared with the first sprint, the decrease in the normalized EMG amplitude (root mean square) was 9.2 and 14.3% in the fifth and tenth sprints. There was a very high correlation (r2 = 0.972) between total work and EMG amplitude in sprint 10 expressed as a percentage of sprint 1 values, suggesting that fatigue is accompanied by reductions in neural drive and muscle activation, as confirmed in another similar experiment (decrease in voluntary activation from 95 to 91.5%, p < 0.02) (35). Thus, a suboptimal net motor unit activity may impair the ability to repeatedly generate high power outputs, and it may be speculated that strength training with high loads and rapid concentric actions may improve fatigue resistance by increasing motor unit activity toward the end of the repeated-sprint test.
Alternatively, the greater fatigue resistance in the S-group may be a result of more efficient muscle fiber recruitment patterns because of the nature of the strength exercise protocol (19). A number of studies reported a direct relationship between maximal strength training for neural adaptations and improved work efficiency (17,25,32). This has been explained by adaptations within the nervous system, altering the motor unit recruitment pattern, because RE also depends on neuromuscular characteristics (1,33). A connection between neural adaptations and increased RE is also evident in this study, where an increase in RE of about 11% was observed only in the S-group, whereas a nonsignificant 4% increase was seen in the H-group. It is notable that the greater improvement in RE in the S-group was accompanied by a faster running speed at the VT and a tendency for better performance in the distance covered during the Yo-Yo test (H-group: 23.5 ± 4.0% vs. S-group: 33.2 ± 3.0%, p = 0.067, d = 0.62). Therefore, there may be a common mechanism for the improvements in RE and fatigue resistance during repeated sprints.
One interesting finding of this study was that PPO during the initial sprints remained unchanged in both groups after training (Table 2). A possible explanation for the lack of an increase in peak power after the 6-week period may be that players were overreached because of the intense training that involved double training sessions on most weekdays. A short-term decrement of peak muscle performance is typical of overreaching and may be observed immediately after the end of training studies (21,26). This decrease in performance is usually reversed after a period ranging from a few days to a few weeks of reduced training load (15,26). In thisstudy, RSA was measured within a very short period (1-2 days) after the end of the preseason training program (31), and this may explain the lack of improvement in PPO in the initial sprints. As has been previously assumed by Hoff and Helgerud (17), improvement in peak muscle performance after a preseason training period may be more evident after a period of recovery. Another possible explanation for the lack of increase in PPO in the initial sprint may be a change in muscle fiber composition. Campos et al. (7) have reported similar changes of myosin heavy chain content from fast (IIB) to an intermediate slower type (IIAB) in 2 groups: one training with a load corresponding to 3-5RM and the other with a load equal to 9-11RM. These possible shifts in fiber types toward slower phenotypes may explain the lack of change in PPO of sprint 1 in both groups, despite the increase in maximal strength, because power depends on both force and velocity of contraction.
In conclusion, this study has shown that resistance training using high loads during the preseason may be superior to a moderate-load program, not only because it increases strength without a change in muscle mass but also because it results in a greater improvement in RE, game-specific endurance and RSA in the second half of a 10 6-second sprint test with short recovery. These improvements may be attributed to neural adaptations resulting in qualitative changes in the working muscles, which cannot be achieved using traditional moderate-load strength training.
The results of this study suggest that performance of the half-squat exercise with high load (90% of 1RM) putting emphasis on rapid concentric movement is an effective way to promote not only maximal strength but also aerobic and anaerobic parameters related to fatigue in soccer. This was evident by the greater improvement in both RE and repeated-sprint performance, compared with a moderate intensity (70% of 1RM) half-squat protocol. The lack of an increase in LLV after resistance training with a high load may be advantageous in game sports where an increase in muscle mass, attained with the moderate-load protocol, is not always desirable. Therefore, resistance training with high loads may be preferable when the aim is to improve maximal strength and fatigue during sprinting in professional soccer players.
The tendency of the performance in the Yo-Yo test to be higher in the S-group than in the H-group (p = 0.067, d = 0.62) implies that resistance training with high loads may also be used to improve performance in soccer-specific field tests. The lack of an improvement in peak power during sprinting was possibly a sign of overreaching that may be observed after an intense period of preseason training. Had the posttest been performed after a short period (1-2 weeks) of recuperation and not immediately after the end of training, an improvement in peak power may have been evident. Further studies are necessary to examine the effectiveness of high load half-squat training in players of lower competitive level and fitness status.
1. Almasbakk, B and Hoff, J. Coordination, the determinant of velocity specificity? J Appl Physiol
81: 2046-2052, 1996.
2. Arsac, L, Belli, A, and Lacour, J-R. Muscle function during brief maximal exercise: Accurate measurements on a friction-loaded cycle ergometer. Eur J Appl Physiol
74: 100-106, 1996.
3. Beaver, WL, Wasserman, K, and Whipp, BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol
60: 2020-2027, 1986.
4. Behm, DG and Sale, DG. Velocity specificity of resistance training. Sports Med
15: 374-388, 1993.
5. Bishop, D and Edge, J. Determinants of repeated-sprint ability in females matched for single-sprint performance. Eur J Appl Physiol
97: 373-379, 2006.
6. Bishop, D, Edge, J, and Goodman, C. Muscle buffer capacity and aerobic fitness are associated with repeated-sprint ability in women. Eur J Appl Physiol
92: 540-547, 2004.
7. Campos, GE, Luecke, TJ, Wendeln, HK, Toma, K, Hagerman, FC, Murray, TF, Ragg, KE, Ratamess, NA, Kraemer, WJ, and Staron, RS. Muscular adaptations in response to three different resistance-training regimens: Specificity of repetition maximum training zones. Eur J Appl Physiol
88: 50-60, 2002.
8. Carson, RG. Changes in muscle coordination with training. J Appl Physiol
101: 1506-1513, 2006.
9. Chamari, K, Hachana, Y, Kaouech, F, Jeddi, R, Moussa-Chamari, I, and Wisloff, U. Endurance training and testing with the ball in young elite soccer players. Br J Sports Med
39: 24-28, 2005.
10. Duchateau, J, Semmler, JG, and Enoka, RM. Training adaptations in the behavior of human motor units. J Appl Physiol
101: 1766-1775, 2006.
11. Edge, J, Hill-Haas, S, Goodman, C, and Bishop, D. Effects of resistance training on H+
regulation, buffer capacity, and repeated sprints. Med Sci Sports Exerc
38: 2004-2011, 2006.
12. Gabriel, DA, Kamen, G, and Frost, G. Neural adaptations to resistive exercise: Mechanisms and recommendations for training practices. Sports Med
36: 133-149, 2006.
13. Glaister, M, Stone, MH, Stewart, AM, Hughes, M, and Moir, GL. The reliability and validity of fatigue
measures during short-duration maximal-intensity intermittent cycling. J Strength Cond Res
18: 459-462, 2004.
14. Green, H, Goreham, C, Ouyang, J, Ball-Burnett, M, and Ranny, D. Regulation of fiber size, oxidative potential, and capillarization in human muscle by resistance exercise. Am J Physiol
45: R591-R596, 1998.
15. Halson, SL and Jeukendrup, AE. Does overtraining exist? An analysis of overreaching and overtraining research. Sports Med
34: 967-981, 2004.
16. Hill-Haas, S, Bishop, D, Dawson, B, Goodman, C, and Edge, J. Effects of rest interval during high-repetition resistance training on strength, aerobic fitness, and repeated-sprint ability. J Sports Sci
25: 619-628, 2007.
17. Hoff, J and Helgerud, J. Endurance and strength training for soccer players. Sports Med
34: 165-180, 2004.
18. Jackson, AS and Pollock, ML. Generalized equations for predicting body density of men. Br J Nutr
40: 497-504, 1978.
19. Johnston, RE, Quinn, TJ, Kertzer, R, and Vroman, NB. Strength training in female distance runners: Impact on running economy. J Strength Cond Res
11: 224-229, 1997.
20. Jones, PRM and Pearson, J. Anthropometric determination of leg fat and muscle plus bone volumes in young male and female adults. J Physiol
204: 63P-66P, 1969.
21. Jubeau, M, Zory, P, Gondin, J, Martin, A, and Maffiuletti, NA. Effect of electrostimulation training-detraining on neuromuscular fatigue
mechanisms. Neurosci Lett
424: 41-46, 2007.
22. Kraemer, WJ and Ratamess, NA. Fundamentals of resistance training: Progression and exercise prescription. Med Sci Sports Exerc
36: 674-688, 2004.
23. Krustrup, P, Mohr, M, Steensberg, A, Bencke, J, Kjaer, M, and Bangsbo, J. Muscle and blood metabolites during a soccer game: Implications for sprint performance. Med Sci Sports Exerc
38: 1165-1174, 2006.
24. Lakomy, HKA. Measurement of work and power output using friction-loaded cycle ergometers. Ergonomics
29: 509-517, 1986.
25. Loveless, DJ, Weber, CL, Haseler, LJ, and Schneider, DA. Maximal leg-strength training improves cycling economy in previously untrained men. Med Sci Sports Exerc
37: 1231-1236, 2005.
26. Maffiuletti, NA, Cometti, G, Amiridis, IG, Martin, A, Pousson, M, and Chatard, JC. The effects of electromyostimulation training and basketball practice on muscle strength and jumping ability. Int J Sports Med
21: 437-443, 2000.
27. Matuszak, ME, Fry, AC, Weiss, LW, Ireland, TR, and McKnight, MM. Effect of rest interval length on repeated 1 repetition maximum back squats. J Strength Cond Res
17: 634-637, 2003.
28. Mendez-Villanueva, A, Hamer, P, and Bishop, D. Fatigue
in repeated-sprint exercise is related to muscle power factors and reduced neuromuscular activity. Eur J Appl Physiol
103: 411-419, 2008.
29. Mohr, M, Krustrup, P, and Bangsbo, J. Match performance of high-standard soccer players with special reference to development of fatigue
. J Sports Sci
21: 519-528, 2003.
30. Mohr, M, Krustrup, P, Nielsen, JJ, Nybo, L, Rasmussen, MK, Juel, C, and Bangsbo, J. Effect of two different intense training regimens on skeletal muscle ion transport proteins and fatigue
development. Am J Physiol Regul Integr Comp Physiol
292: R1594-R602, 2007.
31. Mujika, I, Padilla, S, Pyne, D, and Busso, T. Physiological changes associated with the pre-event taper in athletes. Sports Med
34: 891-927, 2004.
32. Osteras, H, Helgerud, J, and Hoff, J. Maximal strength-training effects on force-velocity and force-power relationships explain increases in aerobic performance in humans. Eur J Appl Physiol
88: 255-263, 2002.
33. Paavolainen, L, Häkkinen, K, Hämäläinen, I, Nummela, A, and Rusko, H.Explosive-strength training improves 5-km running time by improving running economy and muscle power. J Appl Physiol
86: 1527-1533, 1999.
34. Prestes, J, Frollini, AB, de Lima, C, Donatto, FF, Foschini, D, de Cássia Marqueti, R, Figueira, A Jr, and Fleck, SJ. Comparison between linear and daily undulating periodized resistance training to increase strength. J Strength Cond Res
23: 2437-2442, 2009.
35. Racinais, S, Bishop, D, Denis, R, Lattier, G, Mendez-Villaneuva, A, and Perrey, S. Muscle deoxygenation and neural drive to the muscle during repeated sprint cycling. Med Sci Sports Exerc
39: 268-274, 2007.
36. Rampinini, E, Bishop, D, Marcora, SM, Ferrari Bravo, D, Sassi, R, and Impellizzeri, FM. Validity of simple field tests as indicators of match-related physical performance in top-level professional soccer players. Int J Sports Med
28: 228-235, 2007.
37. Robinson, JM, Stone, MH, Johnson, RL, Penland, CM, Warren, BJ, and Lewis, RD. Effects of different weight training exercise/rest intervals on strength, power, and high intensity exercise endurance. J Strength Cond Res
9: 216-221, 1995.