Secondary Logo

Journal Logo

Original Research

Validity and Reliability of a Submaximal Intermittent Running Test in Elite Australian Football Players

Veugelers, Kristopher R.1,2; Naughton, Geraldine A.1; Duncan, Craig S.3; Burgess, Darren J.1,2; Graham, Stuart R.2

Author Information
Journal of Strength and Conditioning Research: December 2016 - Volume 30 - Issue 12 - p 3347-3353
doi: 10.1519/JSC.0000000000001441
  • Free



Evaluating sport-specific abilities is an important component of high performance programs (2). Physiological capacities and responses to training are ideally assessed via valid, reliable, and relevant testing that requires a maximal effort representative of competition (26). For example, the yo-yo intermittent recovery 2 (YYIR2) test is capable of determining an athlete's capacity to perform intense intermittent exercise and is reported to have a strong positive relationship with match high speed running distance in both elite soccer (2,22) and Australian rules football (ARF) (24). The YYIR2 test also has the potential to differentiate between fitness levels, among playing positions, playing standards, successful or unsuccessful teams, and season phases (14,17). Importantly, research has demonstrated that ARF players who have a higher intermittent running capacity as assessed by the YYIR2 test can produce higher match exercise intensity and accrue more ball disposals during matches (23,24). Superior YYIR2 test results are also reported to positively influence match performance as assessed by coaches' votes (23). Match performance in team sports is likely influenced by a multitude of factors, such as technical skill, tactical ability, team strength, opposition strength, and score line (21,27,28). However, current research highlights the potential of the YYIR2 test in evaluating physiological capacities that can affect ARF match performance.

The validity, reliability, and physiological responses to YYIR2 test performance are well established (2,17). However, exposing athletes to maximal testing on a regular basis is rarely appropriate in elite team sport given the resultant increase in residual fatigue would likely compromise recovery for subsequent training and competition (26). Consequently, routine fitness testing in team sport environments must be balanced with the management of player's well-being and fatigue to ensure minimal disruption to the overall program (26).

Submaximal testing may provide a viable alternative for monitoring physiological capabilities and responses to training. In contrast to maximal testing, submaximal tests can be implemented frequently as a monitoring tool without adversely affecting the normal training process or causing excessive fatigue (26). The submaximal heart rate responses of soccer players to the YYIR1 and YYIR2 tests have been found to be negatively correlated with maximal test performance (2,15). This suggests that submaximal versions of the YYIR tests may be capable of predicting maximal test performance. To date, previous testing has investigated only associations between heart responses and performance within the same maximal test. The capacity of a stand-alone submaximal intermittent running test to evaluate the intermittent running capacity of team sport athletes remains unknown.

Monitoring heart rate (HR) during submaximal exercise may also provide objective information about the body's physiological responses to variations in training load (3). Heart rate variables, such as exercise heart rate (HRex) and heart rate recovery (HRR), have been used to unobtrusively and non-invasively monitor training status in endurance sports with varying levels of success (3). For example, a more rapid HRR and decreased HRex have been shown to be indicative of improved fitness (3,19) and states of overreaching (1,4). Conversely, slower HRR and increased HRex are potential indicators of deconditioning (19,20). When considered in the context of a specific training phase, such variables may provide valuable insight into an individual's readiness to perform, which could prove to be particularly useful in high performance settings.

The benefits of using HR responses to a stand-alone submaximal intermittent running test for monitoring fitness and fatigue in elite team sport athletes remain uncertain. Despite potential advantages of submaximal testing, modified tests require assessment for validity and reliability before they can be deemed effective. Therefore, the objective of this study was to establish the validity and reliability of a submaximal intermittent running (SIR) test in a group of elite ARF players.


Experimental Approach to the Problem

Validity of the SIR test was investigated using correlations between HR responses from the SIR and measures from YYIR2 testing (e.g., maximal distance and HR responses). The SIR and YYIR2 tests were completed 48 hours apart. This timeframe was chosen to represent a balance between minimizing residual fatigue and ensuring that fitness levels of athletes remained stable between tests. The protocol was completed twice during 2 different weeks in the preseason period, which both followed reduced load weeks. Table 1 presents a typical weekly preseason training schedule, including validity testing for the SIR test. The day-to-day reliability of the SIR test was then evaluated over 3 trials on successive days following a subsequent de-loading week in the preseason phase. Heart rate was monitored continuously during each test using Firstbeat HR monitors, and data were downloaded using the proprietary software (Firstbeat Technologies, Jyväskylä, Finland). The validity of these HR devices has been established (9).

Table 1.:
Typical preseason training schedule including testing protocol for validity.*


Participants were 45 senior and rookie-listed professional ARF players from one Australian Football League (AFL) club (mean ± SD; age 23 ± 4 years, height 188 ± 8 cm, body mass 85 ± 8 kg, time spent on an AFL list 6 ± 4 years). After approval from the Human Ethics Research Committee at the Australian Catholic University, written informed consent was obtained from all participants to use their data for research purposes. All participants completed testing as part of the normal training regime and were familiar with testing procedures before the study. Participants were free from any injury that may have limited their ability to complete testing.


The relationship between HR and exercise can be influenced by a variety of factors, including mode of exercise, training status, exercise duration, environmental conditions, time of day, hydration status, and caffeine intake (19). Therefore, the following standardized conditions were used to minimize confounding factors with the potential to mask a change in training status: (a) all SIR and YYIR2 tests were performed indoors at the same time of day (am) on the same artificial turf surface; (b) athletes participated in a standardized 10-minute warm-up before testing that consisted of various running-based exercises of increasing intensity; (c) athletes were advised to maintain their normal daily routine on the morning of testing (e.g., similar breakfast and fluid consumption, caffeine intake, same clothing, and shoes); and (d) the same sports science staff administered each test to ensure procedures remained consistent.

Yo-Yo Intermittent Recovery 2 Test

The YYIR2 test was performed using previously established procedures (17). The test consisted of repeated 2 × 20-m shuttle runs at a progressively increased speed. The speed was controlled by audio beeps from a prerecorded source. Athletes had a 10-second active recovery period between each 2 × 20-m shuttle run that consisted of a 5-m shuttle completed at walking pace. The first level of the YYIR2 test commenced at 13 km·h−1 and was followed by stepwise speed increments until either volitional exhaustion or failure to reach the finishing line occurred twice in the allocated time. The test result was reported as the total distance covered. Heart rate was recorded continuously throughout the protocol for each athlete using a Firstbeat HR monitor placed around the chest.

Submaximal Intermittent Running Test

The SIR test followed a similar protocol to the YYIR2 test. However, modifications were based on predetermined testing priorities to (a) impose only a submaximal intensity to minimize additional fatigue, (b) elicit a submaximal intensity that minimizes day-to-day variability in HR, and (c) allow for easy incorporation into a warm-up.

Subsequently, the SIR test consisted of 2 × 18-m repeated shuttle runs and was terminated after 4 minutes. Using 18 m rather than 20 m was a strategy designed to subtly reduce the intensity of the demands. Day-to-day variation in HR has been described as lowest during exercise of higher intensities, for example, 85–90% of maximal HR (HRmax) (18,19). Preliminary testing of a group of athletes from the same population as tested in the current study (n = 17) indicated that the relative intensity of HR at the end of the final working stage (4-minute mark) of the SIR test was 83 ± 4% of HRmax. Therefore, given the desired balance for low variability and for the test to elicit a truly submaximal HR response between 75 and 85% of HRmax (7), it was postulated that the use of a fixed 4-minute protocol of repeated 18-m shuttle runs would provide an acceptable intensity for testing in the chosen group of athletes. The protocol was further strengthened by the moderate inverse correlation previously reported between submaximal HR and YYIR2 performance before but not after 4 minutes of the YYIR2 test (14).

Following the 4 minutes of submaximal running, a standardized recovery period of 3 minutes was implemented in which players were required to remain in a stationary standing position. Heart rate was recorded continuously during both the running and the recovery periods using a Firstbeat HR monitor placed around the chest.

Measuring Heart Rate Responses to Exercise

HRex was calculated as the average HR (expressed as a percentage of HRmax) during the final 30 seconds of the 4-minute running period of the SIR test (7). Each athlete's HRmax was estimated as the peak HR reached during previous YYIR2 testing, reported to be a valid estimate of maximal HR response (2). To explore the suitability of shorter testing protocols, HRex was also calculated at the 2- and 3-minute mark of the SIR test. In addition, HRex was monitored throughout YYIR2 testing to provide comparisons with the SIR test.

Heart rate recovery was determined at 3 time points; 1 minute (HRR60s), 2 minutes (HRR120s), and 3 minutes (HRR180s) following the 4-minute running period of the SIR test. Heart rate recovery was calculated as the absolute difference between 4-minute HRex and HR after each minute of the recovery period (7). To account for any potential change in 4-minute HRex that may have impacted the measurement of HRR, all HRR data were expressed as a percentage of 4-minute HRex (7). This relative index is also likely to minimize any interpersonal differences (8).

Statistical Analyses

Analyses were conducted using SPSS for Windows and a reliability-specific spreadsheet (13). Descriptive statistics are reported as mean ± SD. A Shapiro–Wilk test was used to verify the normal distribution of the data. Statistical significance was set at p ≤ 0.05.


Linear regression analysis was used to examine the degree of association between HR responses to SIR test and YYIR2 test and YYIR2 test performance. Magnitudes of Pearson's correlation coefficients were assessed based on the following recommendations: 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), nearly perfect (r > 0.9), and perfect (r = 1) (12). Paired t-tests were used to establish any differences in HRex at corresponding time points under maximal and submaximal conditions. Statistics are reported with 95% confidence limits.


Test–retest reliability of the SIR test was determined using a reliability-specific spreadsheet (13), which calculated the change in mean, intraclass correlation coefficient (ICC), and typical error of measurement (TE) expressed as a coefficient of variation (CV). Statistics are reported with 95% confidence limits. The smallest worthwhile change (SWC) was also used to compliment TE and assess test usefulness. The rearrangement of Cohen's d effect size calculation enables the SWC to be determined by multiplying the smallest worthwhile effect (0.2) by the between-subject SD (11). A test's capacity to detect change is considered “good” when TE ≤ SWC, “satisfactory” when TE = SWC, and “marginal” when TE ≥ SWC (10).



Descriptive statistics for SIR and YYIR2 test protocols during validity testing are presented in Table 2. A total of 38 players completed submaximal and maximal testing under standardized conditions (21 ± 2.1°C, 41 ± 3% relative humidity). The mean distance traveled during the YYIR2 test was 1,141 ± 318 m. In comparison, the SIR test had a capped distance of 468 m. The HRex was consistently lower at each corresponding time point in the SIR test compared with the YYIR2 test (p < 0.01).

Table 2.:
Descriptive statistics (group mean ± SD) for validity testing (n = 38).*

Linear regression analyses for validity testing are presented in Table 3. Consistently, large inverse correlations were reported between 2-, 3-, and 4-minute HRex during the SIR test and YYIR2 test performance, as denoted by total distance covered (r = −0.58 to −0.61, p < 0.01). Large inverse correlations also existed within the YYIR2 test between total distance and HRex at 2, 3, 4, 5, and 6 minutes (r = −0.50 to −0.60, p < 0.01). In contrast, the relationship between HRex at 7 and 8 minutes of the YYIR2 test and total distance covered during the YYIR2 were moderate in strength (r = 0.42–0.48, p < 0.01). Submaximal intermittent running test HRR120s and HRR180s were also moderately correlated to YYIR2 test total distances (r = 0.32–0.35, p ≤ 0.05). A small correlation was observed between HRR60s during the SIR test and YYIR2 test distance (r = 0.24, p = 0.07).

Table 3.:
Pearson's correlations between heart rate response and yo-yo intermittent recovery 2 test distance (n = 38).*

The relationship between 4-minute HRex during SIR and YYIR2 test conditions and YYIR2 test performance is illustrated in Figure 1. Thirty percent of the variation in YYIR2 test performance was explained by HRex at the 4-minute mark of the test. In comparison, 34% of the variation in YYIR2 test performance was explained by HRex at the 4-minute mark of the SIR test.

Figure 1.:
Relationship between 4-minute exercise heart rate (HRex) during the submaximal intermittent running (SIR) and yo-yo intermittent recovery 2 (YYIR2) tests and YYIR2 test performance.


Descriptive statistics for reliability testing are shown in Table 4. A total of 25 players completed all 3 SIR tests on successive days under standardized conditions (23 ± 1.4°C, 39 ± 2% relative humidity).

Table 4.:
Descriptive statistics (group mean ± SD) for reliability of submaximal intermittent running test (n = 25).*

Inferential statistics for the reliability trials are displayed in Table 5. Strong correlations for ICCs were observed for all HRex and HRR measures (r = 0.90–0.97) (12). Coefficient of variation ranged between 1.3 and 9.2% for all variables. Four-minute HRex recorded both the strongest ICC (r = 0.97) and the lowest CV (1.3%). Of all test parameters, 4-minute HRex was the only variable to achieve a TE < SWC.

Table 5.:
Inferential statistics for reliability of submaximal intermittent running test (n = 25).*


The YYIR1 and YYIR2 tests have been comprehensively evaluated for field-based sports science testing (2). Despite the depth of research, this seems to be the first study to investigate the validity and reliability of an SIR test in elite ARF players. Heart rate responses from the SIR test provided a valid indicator of YYIR2 performance in elite ARF players. Furthermore, HR responses during the SIR test were found to have acceptable day-to-day reliability, with HRex at 4 minutes being the most reliable and sensitive measure associated with YYIR2 performance. These findings support the use of an SIR test to indicate intermittent running capacity in elite ARF players as an alternative, nonfatiguing method to maximal YYIR2 testing.

The first major finding of the study was that linear regression analyses showed HRex at 2, 3, and 4 minutes of the SIR test to have large inverse relationships with YYIR2 test distance in this study's participant cohort (r = −0.58 to −0.60, p < 0.01). Therefore, HR responses to an SIR test lasting no longer than 4 minutes can provide a valid indicator of maximal intermittent running performance.

The magnitude of relationships found between HRex during the SIR test and YYIR2 test distance was consistent with results from previous investigations of intermittent running tests in soccer. For example, moderate inverse relationships within the same test were observed between HRex after 3 minutes and YYIR1 distance (16) and also between 2 and 3 minutes and YYIR2 distance covered in elite soccer players (15,17).

Similar relationships between submaximal HR responses and YYIR1 and YYIR2 total distances have also been reported in sub-elite soccer players. For example, moderate-to-large correlations were observed between minutes 2 and 4 of HRex and YYIR1 total distances, and minute 2 of HRex and YYIR2 total distances (15). In contrast, HRex at 2 minutes of the YYIR2 was largely correlated with YYIR2 distances covered in sub-elite but not in elite players (14). Interestingly, HR after 2 and 4 minutes of the YYIR2 tests was reported to be lower (9 and 6%, respectively) for elite compared with sub-elite players (14). It is possible that such differences may be the result of both a higher fitness level and the greater relative intensity of training and matches in elite compared with sub-elite players. Subsequently, the most effective submaximal tests may be those that are designed to ensure that the specific population being tested generates a relative HR response within the most sensitive HR range.

The second major finding of the study was that the reliability of HR measures during the SIR test compared favorably with the lowest CVs previously recorded. For example, CV for both HRex (1.3–2.0%) and HRR (4.4–9.2%) reported in this study were lower than previous investigation (5). However, it is acknowledged that the CV for HR measures can vary substantially depending on testing protocol, training status, age, environmental conditions, and analytical variations (5,8,19). Nevertheless, in this group of athletes, the SIR test can elicit reliable day-to-day HR responses. Homogeneity of fitness, training, and physical characteristics of players in the current study may have contributed to the low CV values observed.

Of all the SIR test HR variables collected, 4-minute HRex recorded the lowest CV (1.3%) and was the only measure determined to have a TE < SWC. These findings support previous research showing HR measured during exercise of higher rather than lower intensities is more reliable and sensitive to change (18,19). In addition, the stronger correlations reported in the current study between total distance and HRex compared with HRR agrees with a previous investigation in elite ARF (6). Collectively, these results suggest that HR responses from an SIR test provide valid and reliable associations with YYIR2 performance in elite ARF players, with 4-minute HRex providing the most sensitive and reliable measure.

Some limitations are associated with monitoring HR. The homogeneity of HR responses observed in athletes tested for the current study may not extend to other sporting populations. Subsequently, using the SIR test with different athletes may require consideration of the appropriateness of exposure to the same absolute workload. Diversity in physical attributes within any group may result in inconsistent HR responses between individual players. This may affect the interpretation of test results given day-to-day reliability of HR is improved at higher exercise intensities (18) and speed of HRR can be influenced by initial HR (25).

The findings from this study reflect only one of many potentially modified submaximal protocols to indicate maximal YYIR2 performance. Although the modified 18-m distance of the SIR test was an acceptable protocol in this group of athletes, this modification may not be suitable for a different cohort. Shorter testing durations than 4 minutes may also be beneficial, although these may not be suitable if the test is designed to double as a warm-up before activity. Conversely, protocols longer than 4 minutes may exacerbate fatigue in some players and subsequently interfere with the submaximal nature of the test. Nevertheless, the ability to modify test distance or time to coincide with the most reliable and sensitive HR intensities may be advantageous in a variety of team sport settings. Furthermore, there may be scope to create a number of modifications based on positional differences within team sports that have a less homogenous playing group than the one used in the current study. However, population-specific reliability would need to be established for any modifications to the current testing protocol.

Practical Applications

Monitoring of HR within the SIR test can provide valid and reliable associations with YYIR2 test performance in elite ARF players. In particular, 4-minute HRex in the SIR test was the most effective indicator of intermittent running capacity. The submaximal nature of the test provides broad appeal across multiple team sports: it can be administered as part of a warm-up, does not cause excessive fatigue, and can be applied routinely in a large group of athletes. Further study should focus on the applicability of a modified SIR test in other team sports and assessing the potential of using individual HR responses to the SIR test for monitoring changes in training status.


The authors would like to thank the players and staff of the Port Adelaide Football Club for their support.


1. Aubry A, Hausswirth C, Louis J, Coutts AJ, Buchheit M, Le Meur Y. The development of functional overreaching is associated with a faster heart rate recovery in endurance athletes. PLoS One 10: e0139754. doi: 10.1371/journal.pone.0139754.
2. 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.
3. Borresen J, Lambert MI. Autonomic control of heart rate during and after exercise. Sports Med 38: 633–646, 2008.
4. Bosquet L, Merkari S, Arvisais D, Aubert AE. Is heart rate a convenient tool to monitor over- reaching? A systematic review of the literature. Br J Sports Med 42: 709–714, 2008.
5. Buchheit M. Monitoring training status with HR measures: Do all roads lead to Rome?. Front Physiol 5: 1–19, 2014.
6. Buchheit M, Racinais S, Bilsborough JC, Bourdon PC, Voss SC, Hocking J, Cordy J, Mendez-Villanueva A, Coutts AJ. Monitoring fitness, fatigue and running performance during a pre-season training camp in elite football players. J Sci Med Sport 16: 550–555, 2013.
7. Buchheit M, Simpson M, Al Haddad H, Bourdon P, Mendez-Villanueva A. Monitoring changes in physical performance with heart rate measures in young soccer players. Eur J Appl Physiol 112: 711–723, 2012.
8. Daanen H, Lamberts R, Kallen VL, Jin A, Van Meeteren NLu. A systematic review on heart-rate recovery to monitor changes in training status in athletes. Int J Sports Physiol Perform 7: 251–260, 2012.
9. Firstbeat Technologies Ltd. Accuracy of Firstbeat Beat-to-Beat Heart Rate Monitor. White Paper. Jyväskylä, Finland: Firstbeat Technologies Ltd.; 2013.
10. Hachana Y, Chaabène H, Ben Rajeb G, Khlifa R, Aouadi R, Chamari K, Gabbett TJ. Validity and reliability of new agility test among elite and subelite under 14-soccer players. PLoS One 9: 1–6, 2014.
11. Hopkins W. How to interpret changes in an athletic performance test. Sportscience 8: 1–7, 2004.
12. Hopkins W. Measures of reliability in sports medicine and science. Sports Med 30: 1–15, 2000.
13. Hopkins W. Reliability (Excel Spreadsheet). Available at: Accessed May 1/2015.
14. 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.
15. Ingebrigtsen J, Brochmann M, Castagna C, Bradley P, Ade J, Krustrup P, Holtermann A. Relationships between field performance tests in high-level soccer players. J Strength Cond Res 28: 942–949, 2014.
16. 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.
17. 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.
18. Lamberts R, Lambert MI. Day-to-day variation in heart rate at different levels of submaximal exertion: Implications for monitoring training. J Strength Cond Res 23: 1005–1010, 2009.
19. Lamberts R, Lemmink KAPM, Durandt JJ, Lambert MI. Variation in heart rate during submaximal exercise: Implications for monitoring training. J Strength Cond Res 18: 641–645, 2004.
20. Lamberts R, Swart J, Capostagno B, Noakes TD, Lambert MI. Heart rate recovery as a guide to monitor fatigue and predict changes in performance parameters. Scand J Med Sci Sports 20: 449–457, 2010.
21. Liu H, Gómez MA, Gonçalves B, Sampaio J. Technical performance and match-to-match variation in elite football teams. J Sports Sci 34: 509–518, 2016.
22. 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, 2003.
23. Mooney MG, Cormack S, O'Brien B, Morgan WM, McGuigan M. Impact of neuromuscular fatigue on match exercise intensity and performance in elite Australian football. J Strength Cond Res 27: 166–173, 2013.
24. Mooney MG, O'Brien B, Cormack S, Coutts A, Berry J, Young W. The relationship between physical capacity and match performance in elite Australian football: A mediation approach. J Sci Med Sport 14: 447–452, 2011.
25. Pierpont GL, Stolpman DR, Gornick CC. Heart rate recovery post-exercise as an index of parasympathetic activity. J Auton Nerv Syst 80: 169–174, 2000.
26. Pyne DB, Spencer M, Mujika I. Improving the value of fitness testing for football. Int J Sports Physiol Perform 9: 511–514, 2014.
27. Sullivan C, Bilsborough JC, Cianciosi M, Hocking J, Cordy J, Coutts AJ. Match score affects activity profile and skill performance in professional Australian football players. J Sci Med Sport 17: 326–331, 2014.
28. Sullivan C, Bilsborough JC, Cianciosi M, Hocking J, Cordy JT, Coutts AJ. Factors affecting match performance in professional Australian football. Int J Sports Physiol Perform 9: 561–566, 2014.

athlete monitoring; yo-yo intermittent recovery test; fitness test; submaximal; heart rate; team sport

© 2016 National Strength and Conditioning Association