Various physical abilities (3,8,37) and technical and tactical skills (4,32,34) have been pointed out as important for match running performance in soccer. The sport is hallmarked by its explosive and intermittent activities of high speed running and sprinting interspersed by periods of low intensity exercise (7,26,31). Previous studies on male and female soccer players indicate that maximal running speed (the highest speed achieved when running) (13,20,26,35), repeated sprinting (repeated bouts of sprinting) (19,20,25,30), and high-intensity running (the sum of the distance covered during a soccer match in moderate speed, high speed, and sprint running) (8,18,19,21,23–25) abilities are of paramount importance for reaching elite levels (16,27). Thus, scientists and practitioners are continuously seeking to understand how to maintain and improve these physical components within soccer players.
A number of fitness tests have been used to obtain objective information about players' fitness levels, to prepare the necessary short- and long-term training programs, and to provide motivation and feedback to the players (22,30,38). Laboratory tests (e.g., treadmill test of maximal oxygen consumption) are traditionally considered as the gold standard (10), but in later years more soccer-specific field performance tests have been designed with the aim to better simulate soccer match play to improve test specificity (10,21,38). The Yo-Yo intermittent recovery tests level 1 (IR1) and 2 (IR2) (23,24) have become common physical capacity field tests for soccer players (3,21). Both tests claim to measure players' ability to perform and recover from intense activity (3,21). The Yo-Yo IR1 test measures the players' ability to repeat intermittent activity with a high aerobic component throughout the test reaching maximal heart rates (HRs) at the end of the test, but the IR1 performance also depends on a considerable energy contribution from the anaerobic system (3,22). The Yo-Yo IR2 test has been observed to involve the anaerobic energy delivery system even more, with marked reductions in creatine phosphate (CP) and glycogen during the test (3). Furthermore, both levels of the IR test have shown to result in muscle and blood lactate accumulation, but with the average muscle lactate accumulation in the Yo-Yo IR1 test being only about one-fifth compared with the IR2 test (3,21).
The capacity of the anaerobic energy system is also commonly evaluated with various repeated sprint ability (RSA) field tests (2,36). Possible limiting factors for the ability to repeatedly sprint have previously been suggested to be limited energy supply from anaerobic energy sources and accumulation of metabolites (e.g., H+ and inorganic phosphate) and interstitial potassium, but also limitations in the aerobic glycolysis are thought to influence the RSA (6,14). However, the relative contribution from the various energy systems is speculated to be altered with sprint duration, number of sprints, and the recovery time between sprints (higher aerobic contribution with increased duration and number of sprints) (6,14,36). In addition, single sprint tests of short duration are considered relevant for soccer (13), and single sprint tests highly trigger the anaerobic energy delivering systems as it places high demands on phosphocreatine degradation and anaerobic glycolysis (14,36).
Although the aim of these respective tests is to measure specific physical capacities, the overlap in information attained from the tests can be considerable, and practitioners and scientists continue to debate which tests to use (38). Because these tests are relatively time consuming and may be physically and mentally strenuous for the players, it is of great practical interest to evaluate the relation between the different tests. This could reduce the number of tests necessary for attaining sufficient information about the physical capacity of the players and are therefore able to devote more hours to training.
It is scientifically well accepted that lowered submaximal HRs during exercise indicate improved aerobic fitness (9) and it has been proven that HR during submaximal versions of the Yo-Yo intermittent endurance test level 2 (Yo-Yo IE2) and Yo-Yo IR1 could provide information about test performances and training adaptations as well as soccer match locomotor activity (21). Hence, to investigate and establish the relationship between submaximal HR responses during the test and maximal test performances could be of great importance for the practitioner, as this could facilitate the use of submaximal HRs from the field tests to estimate players' physical capacity and to monitor training-induced changes. Furthermore, this could reduce the physical and mental burdens of testing on soccer players. However, to the best of our knowledge, only sparse research has focused on the relationship between submaximal HR responses and test performances as well as between these (e.g., Yo-Yo IR1, Yo-Yo IR2, RSA, and single sprint speed) field test performances in soccer players.
Accordingly, the aim of this study was to investigate the relationship between Yo-Yo IR1 and IR2 test performances, RSA (7 × 35 m), maximal sprinting speed (10, 20, and 35 m), and submaximal HRs after 2 and 4 minutes of the Yo-Yo IR tests. Second, we aimed to investigate the reproducibility of these submaximal HR measures. On the basis of which energy systems that theoretically affect performance in the present field tests, we expect to find large correlations between the 2 Yo-Yo IR test performances, between the Yo-Yo IR2 and the RSA test performances, and between the RSA and single sprint test performances. Moreover, relationships would also be expected between submaximal HRs and Yo-Yo IR test performances as well as between performances of the Yo-Yo IR1 and RSA tests, and between sprint speed test performance and Yo-Yo IR test performances.
Experimental Approach to the Problem
To assess performances in Yo-Yo IR test levels 1 and 2, sprint speed on 10, 20, and 35 m, RSA on 7 × 35 m, and submaximal HRs after 2 and 4 minutes of the Yo-Yo IR tests, we used a descriptive correlative-explorative study design and tested players from the 3 best leagues in Norway during 3 sessions within 3 consecutive days. The first day consisted of testing sprint speed and RSA, whereas the second and third days consisted of Yo-Yo IR1 and IR2 testing, respectively. We then applied Pearson's correlation tests to assess the relationships between the test performance results. In addition, as a regression model has the ability to assess the impact of an independent variable controlled for the other independent variables in the model, we used multiple regressions also. Furthermore, to test the reproducibility of the HR measures from the Yo-Yo IR tests, we performed additional tests and then calculated the reliability of the HR measures according to the guidelines given by Hopkins (17).
Fifty-seven male soccer players (characteristics in Table 1) participated in this investigation. To be included in the study the players had to play regularly at one of the 3 highest levels (n = 13 professionals from top division, n = 10 from second division, n = 34 from third division) of Norwegian soccer and volunteer to participate in this project. In addition, the players had to be categorized by the team's medical staff as completely injury free. All tests were performed within 3 consecutive days at the start (within 3 weeks) of the players' competitive season. All the players gave their written informed consent to participate in the study, and together with the clubs involved, all players have approved the use of the depersonalized data. The study has been conducted in accordance to the Helsinki declaration. The study has also been accepted by the Privacy Ombudsman for Research (project number 23335), the Norwegian Social Science Data Service.
Before the experiment, the subjects received information about the importance of upholding normal dietary routines and fluid intakes through all 3 days of testing. In addition, players received written and verbal information about the test procedures. All players had been familiarized to the tests before the experiment as all teams used the same tests at the beginning of their preseason training (to reduce measurement errors caused by learning effects) (31).
Body mass (kg) was attained using an electronic scale (A&D Company Limited, Tokyo, Japan) and height (m) by using a stadiometer (KaWe Medizintechnik, Asperg, Germany). The results of these measurements were later rounded to the nearest 0.1 kg and 0.005 m for weight and height, respectively, and both were measured before all other tests. Test day 1 further consisted of the sprint and repeated sprint test. The sprint speed and 7 × 35 m repeated sprint tests were both measured using a timing system (Brower Timing Systems, Draper, UT, USA) with 3 sets of infrared beams/transmitters, each consisting of an infrared sender (IRD-T175) and an infrared emitter with antennas. Each unit was mounted on a tripod. Data were sent from the timing gates directly to the handheld coach monitor. The test-retest reproducibility of the current system has been evaluated elsewhere, and the system is considered to be a useful tool for measuring running speed (36). Furthermore, data were manually plotted in Microsoft Excel (Microsoft Corp., Redmond, WA, USA) and saved on a PC (Pentium 4 with Microsoft Windows XP). Before these tests, all players conducted a standardized warm up consisting of various running exercises with increasing intensity toward the start of the test, and ultimately, players performed three 40-m sprints close to maximum speed separated by 2 minutes of rest. Then, the sprint test consisted of 3 trials of 35 m maximum sprints with a minimum of 3-minute rest between them. During the sprint test, time was measured for the 0–10, 0–20, and 0–35 m sprint. Subjects started from a standing position with their front foot on a starting line 30 cm behind the first photo cell. When given a signal, they started the sprint using the shortest time possible to reach the photo cells at the finishing line. Timing started automatically when the subjects broke the beam of the photo cell placed at the starting line. The interval times were recorded when the subjects passed the photo cells at 10 m for the 0–10 m sprint, at 20 m for the 0–20 m, and at 35 m for the 0–35 m. The between-trials reproducibility of the sprint speed measures was later calculated.
A 10-minute rest period was then given before the repeated sprint test, which, in turn, consisted of 7 maximum 35-m sprints interspersed with 25 seconds of active recovery. The mean sprint time was calculated as it has been described as a reproducible and useful measure for evaluating RSA (15,20). A previous study investigating the short-term reproducibility (testing twice within 1 week) of a 6 × 40 m RSA test demonstrated a coefficient of variation (CV) of only 0.8% for the mean time measure (20). As the present study did not retest the players, we are not able to provide reproducibility measures for the RSA test within the present cohort.
On the second day of testing, players performed the Yo-Yo IR 1 test, while performing the Yo-Yo IR 2 on the third day of testing. All players were familiar with the test procedure of both Yo-Yo IR tests. Before the Yo-Yo IR testing, all players conducted the same standardized warm up as they performed in relation to the repeated sprint testing, although ultimately followed by three 40-m shuttle run sprints at near maximum speed. Both levels of the Yo-Yo IR test consist of 20-m shuttle runs at increasing velocities (controlled by sound signals from a CD player) with 10 seconds of active recovery in-between. The Yo-Yo IR1 starts at 10 km·h−1 and then consists of 4 running bouts with increasing speed from 10 to 13 km·h−1, 7 bouts from 13.5 to 14 km·h−1, and thereafter stepwise 0.5 km·h−1 speed increments every 8 running bouts. The Yo-Yo IR2 test starts at 13 km·h−1 (21,24) and continues with 4 running bouts of increasing speed from 13 to 16 km·h−1, 7 bouts of increasing speed from 16.5 to 17 km·h−1, and after that continues with stepwise increases of 0.5 km·h−1 every 8 running bouts. The stepwise speed increments were followed until players either twice had failed to reach the line at the time of the sound signal from the test CD (objectively evaluated by experienced fitness coaches) or felt too exhausted to continue at the required speed (subjectively evaluated), as previously described (24). The total distance covered by the player until he stopped was then registered. During both Yo-Yo IR tests, HR was recorded in 5-second intervals (Polar Team2; Polar, Kempele, Finland). Data were later read with the accompanying software (Polar Team2 software) installed on a personal computer (Dell XPS M1330; Dell Products, Plano, TX, USA). A Denon CD player (DC 1015; Denon Brand Company, Kanagawa, Japan) with a Sony amplifier (F590ES) and Sony loudspeakers (SS-E420; Sony Corporation, Tokyo, Japan) was used to play the Yo-Yo IR CD tracks. The Yo-Yo IR tests were conducted on the same indoor artificial turf as the sprint and repeated sprint test to ensure stable conditions throughout all tests, using previously described procedures (23,24). Verbal encouragement was given from the same test leader, as well as the respective team coaches, throughout all tests. Previous studies demonstrated a high reproducibility in the Yo-Yo IR1 and IR test performances, as no significant differences were found across 2 tests within 1 week (CV of 4.9 and 9.6%, respectively) (23,24). However, the present study did not retest the players. Hence, we cannot provide reproducibility measures (e.g., intraclass correlation [ICC]) for the Yo-Yo IR1 or IR2 performances within the present cohort.
To investigate test-retest reproducibility for the submaximal HR measures, we conducted additional Yo-Yo IR1 and Yo-Yo IR2 tests on 2 later occasions, as it was not possible to do it within the 3 days of performance testing because of time limitations. The cohort used for this purpose was very similar to the other study participants with regards to sex and test performances. From these additional tests, we calculated ICC and CV for HR after 2 (n = 33), 4 (n = 33), and 6 minutes (n = 37) of the Yo-Yo IR1 test and 2 minutes of the Yo-Yo IR2 test (n = 10), respectively.
SPSS 17.0 (SPSS, Inc., Chicago, IL, USA) for Windows were used for data analyses. All measured variables, assessed by a Shapiro-Wilk's test, were found to be normally distributed. To assess the relationships between Yo-Yo IR test levels 1 and 2 performances (m), sprint speed (fastest time) on 10, 20, and 35 m, RSA (mean time) on 7 × 35 m (and 7 × 10 m and 7 × 20 m split-times), and submaximal HRs after 2 and 4 minutes of the Yo-Yo IR tests, descriptive correlations between performance measures were used. Correlations were determined by using Pearson's product-moment correlation coefficient (r). To investigate the independent associations between sprint speed, RSA, and submaximal HRs on Yo-Yo IR performances, a series of regressions, controlling for age, body mass, and stature were performed. In addition, a 2-way mixed ICC reliability test and the CV between sprint speed trials, and for submaximal HRs, were calculated in accordance with the guidelines provided by Hopkins (17). Furthermore, we used Stata 11 (StataCorp., College Station, TX, USA) to investigate the association between sprint speed, RSA, and submaximal HRs and Yo-Yo IR performances by doing a series of ordinary least square (OLS) regressions (we applied OLS because the independent variables were on interval level) with each of the above independent variables and a set of control variables (age, body mass, and stature). Higher values on the beta coefficients indicate a larger impact on the dependent variable. In general, 1 level increase in the given independent variable will lead to an increase/decrease in the dependent variable corresponding to “beta.” Magnitude of correlation coefficients were considered as trivial (r < 0.1), small (0.1 < r < 0.3), moderate (0.3 < r < 0.5), large (0.5 < r < 0.7), very large (0.7 < r < 0.9), and nearly perfect (r > 0.9) and perfect (r = 1.0) (17). In addition, as a robustness check, all regressions were carried out with log-transformed variables to reduce the effects of potential outliers. However, the results did not differ significantly, and, hence, the absolute values were used for further analysis. The 0.05 level of significance was adopted for all statistical tests.
Descriptive and performance characteristics are presented in Table 1 and correlations between performance measures are presented in Table 2. We found a very large correlation between the Yo-Yo IR test levels 1 and 2 performances. Repeated sprint ability at 10 and 20 m (split-times) and 35 m all correlated moderately and largely to Yo-Yo IR1 performances. In addition, RSA at 20 m (split-time) and 35 m moderately to largely correlated to Yo-Yo IR2 performance. A small and moderate correlation between sprint speed at 20 and 35 m and Yo-Yo IR1 performance were found. Sprint speed over 35 m moderately correlated to Yo-Yo IR2 performance. In addition, sprint speed correlated largely to very largely to RSA. The correlations between RSA measures (mean time at 10 m split, 20 m split, and 35 m) were found to be very large (Table 2). Furthermore, the relationships between sprint performances at 10, 20, and 35 m were very large to nearly perfect. Intraclass correlation measures of sprint performance were 0.94 (CV = 0.7%), 0.97 (CV = 1.4%), and 0.96 (CV = 1.9%) for 10-, 20-, and 35-m sprint speed, respectively.
As seen in Table 3, within the Yo-Yo IR1 test, moderate and large correlations were found between submaximal HRs after 2 and 4 minutes and the test performance. The Yo-Yo IR2 test performance largely correlated to submaximal HRs after 2 minutes. Intraclass correlation measures of submaximal HRs after 2, 4, and 6 minutes of the Yo-Yo IR1 test were 0.92 (CV = 4.1%, n = 33), 0.93 (CV = 3.8%, n = 33), and 0.90 (CV = 4.0%, n = 37), respectively. For submaximal HR after 2 minutes of the Yo-Yo IR2, we found an ICC of 0.72 (CV = 2.9%, n = 10).
The OLS-regressions (Table 4) revealed significant associations between sprint speed at 20 and 35 m and the Yo-Yo IR1 test performance, whereas significant associations only between 35 m and the IR2 test performance. Furthermore, RSA at 35 m was found to be positively related to both levels of the Yo-Yo IR test. In addition, submaximal HRs both after 2 and 4 minutes proved to be associated with Yo-Yo IR1 and IR2 performances. No other than the displayed measures were found to significantly relate the Yo-Yo IR test level 1 or 2 performances.
The relationship between the Yo-Yo IR test performances was very large and we found large correlations between RSA at 35 m and Yo-Yo IR test performances. However, RSA at 10 and 20 m (split-times) only moderately correlated to performances in the Yo-Yo IR tests. Moreover, the relationship between performances in the RSA test and sprint speed test was found to be large to very large. Ultimately, larger correlations than expected were found between submaximal HRs after 4 minutes of the Yo-Yo IR 1 test and 2 minutes of the Yo-Yo IR 2 test and test performances. These findings may suggest inclusion of one of the Yo-Yo tests and a RSA test in a general soccer-specific field test protocol for high-level soccer players, along with submaximal Yo-Yo tests with HR measurements for frequent, time-efficient exercise testing in periods with no maximal tests.
The present finding of a very large correlation between Yo-Yo IR1 and Yo-Yo IR2 performances is in line with previous findings among Danish and Norwegian elite and subelite players (21), whereas slightly higher than shown among other professional and amateur soccer players (33). This may well be because of the fact that both tests elicit maximal HRs and have a high anaerobic energy contribution (3,21), although it should be mentioned that running speeds and the rate of anaerobic energy turnover are much higher in the IR2 test (3,21). Because of the nature of high-level soccer matches, with high running speeds and high anaerobic turnover frequent high-intensity activities, it might be that the Yo-Yo IR2 test is more suitable for the high-level male player.
The present study also revealed that RSA on 10 m, 20 m (split-times), and 35 m correlated moderately to largely to the Yo-Yo IR1 test performance, whereas only 20 m split-time and 35 m RSA correlated to the Yo-Yo IR2 test performance. Assessed with the regression analysis, we found an independent association between RSA on 35 m and both Yo-Yo IR1 and IR2. On the contrary, no significant association between IR2 performance and 5 × 30 m RSA were previously shown (24). Other studies using similar (11) and other RSA protocols (1,5,29) reported somewhat lower correlation coefficients between RSA and Yo-Yo IR test performances than the present study. As the IR1 and IR2 tests have been shown to highly trigger the aerobic energy delivery system (3), and Da Silva et al. (11) found moderate (r = 0.38–0.49) correlations between 7 × 35 m RSA performance and aerobic variables (e.g., V[Combining Dot Above]O2max), we speculate that what we found reflected (as a large correlation) in the relationship between the present IR performances and RSA at 35 m. Furthermore, Pyne et al. (29) speculated that shorter running distances in RSA protocols cause the anaerobic energy delivery to be the main contributor (i.e., CP degradation and muscle buffer capacity), whereas the aerobic system is more pronounced with protocols of longer sprinting distances. The latter could be why the present data indicate stronger relationships between 35 m RSA and Yo-Yo IR test performances compared with 10 and 20 m RSA split-times, respectively, as they are linked closer to CP degradation and muscle buffer capacity (i.e., the anaerobic energy system). This assumption is strengthened by the very large correlations between sprinting speed and RSA performances as single short sprints are claimed to mainly rely on energy from the anaerobic system (14,36).
Furthermore, we observed small to moderate correlations between sprinting speed and Yo-Yo IR test performances. As shown in Table 2, significant independent associations between 20- and 35-m sprinting speed and the Yo-Yo IR1 test performance, and 35-m sprinting speed and Yo-Yo IR2 test performance, were revealed by the performed OLS-regressions. In line with our findings of weak relationships, previous studies found no significant correlations (24,28) between Yo-Yo IR1 or Yo-Yo IR2 performances and sprinting speed. We interpret the present and previous results to indicate that sprint performance and Yo-Yo IR test performance mainly rely on different physiological capacities as sprinting speed mainly relies on anaerobic energy delivery (14,36), while the Yo-Yo IR tests also depend on significant contributions from the aerobic system (3,21,23,24). Thus, it seems that both sprint performance and IR test performance should be assessed when investigating the soccer players' physical capacity.
Yo-Yo IR1 submaximal HRs after 2 and 4 minutes correlated moderately and largely with Yo-Yo IR1 performance, respectively, whereas the Yo-Yo IR2 submaximal HR after 2 minutes correlated largely to IR2 performance. In addition, the regression analysis showed a significant association between the submaximal HRs after 2 and 4 minutes and performance on both Yo-Yo IR test levels. Previous studies also using elite team sport athletes have revealed nonsignificant moderate correlations between submaximal HRs after 3 minutes and Yo-Yo IR1 performance (23) and after 2 and 3 minutes of the Yo-Yo IR2 test (24). However, Krustrup et al. found very large correlations between IR1 performance and HR after 6 and 9 minutes, which may be because of the fact that the subjects in that study were not well-trained soccer players with a high anaerobic intermittent exercise performance.
Ultimately, we recognize that a possible limitation to the present study could be the relatively short recovery periods between the various test days (24 hours) as the availability of glycogen could be essential to all of the test performances (3,24,36). In addition, recovery time between the speed test and the RSA test could affect the present results. However, it has been indicated that 3 minutes between intratest repetitions and 10 minutes between the various tests, and with no activity of above 10 seconds duration, should be sufficient for the rebuilding of energy stores to normal values (12). Furthermore, although previously being described as both reliable (20) and valid (30), the authors recognize the possible limitation of the use of RSA mean time when it comes to discriminating between fast, but highly fatigable, and slow sprinters. Ultimately, the investigation of relationships between the present test performances, and the interpretation of our findings, would have been further substantiated by including physiological data, such as V[Combining Dot Above]O2 and blood lactate. This was, however, not possible to include in the current investigation.
The Yo-Yo IR 1 and 2 test performances, as well as sprint and RSA performances, correlated very largely, and it may therefore be considered using only one of the Yo-Yo tests and a RSA test, in a soccer-specific field test battery. However, based on the nature of the high-level soccer match, with high running speeds and high anaerobic turnover from high-intensity activities, we suggest the Yo-Yo IR2 test for the high-level male player. Moreover, as submaximal HRs during Yo-Yo IR tests seem highly reliable, and related to test performance, we recommend the use of the submaximal HR measures for frequent time-efficient and nonexhaustive testing of intermittent exercise capacity of high-level soccer in periods where maximal tests are avoided because of players' mental and physical burdens.
We like to thank the players for their efforts and Shaher Shalfawi for his contribution with the statistics during the project. We also like to thank Luke Conolly for proof reading the article.
1. Aziz AR, Chia M, Teh KC. The relationship between maximal oxygen uptake and repeated sprint performance indices in field hockey and soccer players. J Sports Med Phys Fitness 40: 195–200, 2000.
2. Aziz AR, Mukherjee S, Chia MY, Teh KC. Validity of the running repeated sprint ability test among playing positions and level of competitiveness in trained soccer players. Int J Sports Med 29: 833–838, 2008.
3. Bangsbo J, Iaia FM, Krustrup P. The Yo-Yo intermittent recovery test
: A useful tool for evaluation of physical performance in intermittent sports. Sports Med 38: 37–51, 2008.
4. Bangsbo J, Norregaard L, Thorso F. Activity profile of competition soccer. Can J Sport Sci 16: 110–116, 1991.
5. Bishop D, Edge J, Davis C, Goodman C. Induced metabolic alkalosis affects muscle metabolism and repeated-sprint ability. Med Sci Sports Exerc 36: 807–813, 2004.
6. Bishop D, Girard O, Mendez-Villanueva A. Repeated-sprint ability—Part II: Recommendations for training. Sports Med 41: 741–756, 2011.
7. Bloomfield J, Polman R, O'Donoghue P. Physical demands of different positions in FA Premier League soccer. J Sports Sci Med 6: 63–70, 2007.
8. Bradley PS, Sheldon W, Wooster B, Olsen P, Boanas P, Krustrup P. High-intensity running in English FA Premier League soccer matches. J Sports Sci 27: 159–168, 2009.
9. Buchheit M, Simpson MB, Al Haddad H, Bourdon PC, Mendez-Villanueva A. Monitoring changes in physical performance with heart rate measures in young soccer players. Eur J Appl Physiol 112: 711–723, 2012.
10. Castagna C, Manzi V, Impellizzeri F, Weston M, Barbero Alvarez JC. Relationship between endurance field tests and match performance in young soccer players. J Strength Cond Res 24: 3227–3233, 2010.
11. Da Silva JF, Guglielmo LG, Bishop D. Relationship between different measures of aerobic fitness and repeated-sprint ability in elite soccer players. J Strength Cond Res 24: 2115–2121, 2010.
12. Dick FW. Sports Training Principles. London, United Kingdom: A & C Black, 2007.
13. Faude O, Koch T, Meyer T. Straight sprinting is the most frequent action in goal situations in professional football. J Sports Sci 30: 625–631, 2012.
14. Girard O, Mendez-Villanueva A, Bishop D. Repeated-sprint ability—Part I: Factors contributing to fatigue. Sports Med 41: 673–694, 2011.
15. Glaister M, Howatson G, Pattison JR, McInnes G. The reliability and validity of fatigue measures during multiple-sprint work: An issue revisited. J Strength Cond Res 22: 1597–1601, 2008.
16. Helgerud J, Engen LC, Wisløff U, Hoff J. Aerobic endurance training improves soccer performance. Med Sci Sports Exerc 33: 1925–1931, 2001.
17. Hopkins WG. Measures of reliability in sports medicine and science. Sports Med 30: 1–15, 2000.
18. Iaia FM, Bangsbo J. Speed endurance training is a powerful stimulus for physiological adaptations and performance improvements of athletes. Scand J Med Sci Sports 20(Suppl. 2): 11–23, 2010.
19. Iaia FM, Rampinini E, Bangsbo J. High-intensity training in football. Int J Sports Physiol Perform 4: 291–306, 2009.
20. Impellizzeri FM, Rampinini E, Castagna C, Bishop D, Ferrari Bravo D, Tibaudi A, Wisloff U. Validity of a repeated-sprint test for football. Int J Sports Med 29: 899–905, 2008.
21. Ingebrigtsen J, Bendiksen M, Randers MB, Castagna C, Krustrup P, Holtermann A. Yo-Yo IR2 testing of elite and sub-elite soccer players: Performance, heart rate response and correlations to other interval tests. J Sports Sci 30: 1337–1345, 2012.
22. Ingebrigtsen J, Dillern T, Shalfawi SA. Aerobic capacities and anthropometric characteristics of elite female soccer players. J Strength Cond Res 25: 3352–3357, 2011.
23. Krustrup P, Mohr M, Amstrup T, Rysgaard T, Johansen J, Steensberg A, Pedersen PK, Bangsbo J. The yo-yo intermittent recovery test
: Physiological response, reliability, and validity. Med Sci Sports Exerc 35: 697–705, 2003.
24. Krustrup P, Mohr M, Nybo L, Jensen JM, Nielsen JJ, Bangsbo J. The yo-yo IR2 test: Physiological response, reliability, and application to elite soccer. Med Sci Sports Exerc 38: 1666–1673, 2006.
25. Krustrup P, Zebis M, Jensen JM, Mohr M. Game-induced fatigue patterns in elite female soccer. J Strength Cond Res 24: 437–441, 2010.
26. Little T, Williams AG. Specificity of acceleration, maximum speed, and agility in professional soccer players. J Strength Cond Res 19: 76–78, 2005.
27. Mohr M, Krustrup P, Bangsbo J. Match performance of high-standard soccer players with special reference to development of fatigue. J Sports Sci 21: 519–528, 2003.
28. Mujika I, Santisteban J, Impellizzeri FM, Castagna C. Fitness determinants of success in men's and women's football. J Sports Sci 27: 107–114, 2009.
29. Pyne DB, Saunders PU, Montgomery PG, Hewitt AJ, Sheehan K. Relationships between repeated sprint testing, speed, and endurance. J Strength Cond Res 22: 1633–1637, 2008.
30. Rampinini E, Bishop D, Marcora SM, Ferrari Bravo D, Sassi R, 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.
31. Rampinini E, Coutts AJ, Castagna C, Sassi R, Impellizzeri FM. Variation in top level soccer match performance. Int J Sports Med 28: 1018–1024, 2007.
32. Rampinini E, Impellizzeri FM, Castagna C, Coutts AJ, Wisløff U. Technical performance during soccer matches of the Italian Serie A league: Effect of fatigue and competitive level. J Sci Med Sport, 12: 227–233, 2009.
33. Rampinini E, Sassi A, Azzalin A, Castagna C, Menaspa P, Carlomagno D, Impellizzeri FM. Physiological determinants of Yo-Yo intermittent recovery tests in male soccer players. Eur J Appl Physiol 108: 401–409, 2010.
34. Reilly T, Bangsbo J, Franks A. Anthropometric and physiological predispositions for elite soccer. J Sports Sci 18: 669–683, 2000.
35. Reilly T, Williams AM, Nevill A, Franks A. A multidisciplinary approach to talent identification in soccer. J Sports Sci 18: 695–702, 2000.
36. Spencer M, Bishop D, Dawson B, Goodman C. Physiological and metabolic responses of repeated-sprint activities: Specific to field-based team sports. Sports Med 35: 1025–1044, 2005.
37. Stølen T, Chamari K, Castagna C, Wisløff U. Physiology of soccer: An update. Sports Med 35: 501–536, 2005.
38. Svensson M, Drust B. Testing soccer players. J Sports Sci 23: 601–618, 2005.