Applicability of Field Aerobic Fitness Tests in Soccer: Which One to Choose?

desire to make tness testing cheaper and easier to conduct in a team-sport setting has led to the development of numerous eld aerobic tness tests. This has contributed to a growing confusion among strength and conditioning coaches about which one to use. The main aim of this narrative review was to examine the reliability, validity, sensitivity and usefulness of the commonly used eld aerobic tness tests and to provide practical guidelines for their use in soccer. The University of Montreal track test (UMTT) and Vam Eval test seem the best options for estimation of maximal oxygen uptake (VO 2max) while the highest signal-to-noise ratio of the 30-15 intermittent tness test (30-15IFT) suggests its superior sensitivity to track changes in tness. The UMTT and 30-15IFT are the best solutions for prescription of long and short high-intensity interval training sessions, respectively. All eld tests mostly present with marginal usefulness, but the smallest worthwhile change for UMTT or Vam Eval test, Yo-YoIRT2 and 30-15IFT are smaller than their stage increment making the improvement of only one stage in the test performance already worthwhile. Strength and conditioning coaches are advised to choose the test based on their speci c purpose of testing. Keywords: 30-15 intermittent tness test;

present with marginal usefulness, but the smallest worthwhile change for UMTT or Vam Eval test, Yo-YoIRT2 and 30-15IFT are smaller than their stage increment making the improvement of only one stage in the test performance already worthwhile. Strength and conditioning coaches are advised to choose the test based on their speci c purpose of testing. Keywords: 30-15 intermittent tness test; Yo-Yo intermittent recovery test; 20-m shuttle run test; University of Montreal track test; Vam Eval test; maximal aerobic speed; exercise prescription; validity; reliability; sensitivity; usefulness 1. Introduction Fitness testing can be considered as a basic professional activity for sport scientists and strength and conditioning coaches [1,2]. It can be conducted for a variety of reasons including assessment of physiological capacities [1,3], talent identi cation and selection [4], training and performance monitoring [5], evaluation of training program effectiveness [1,6] and training prescription [7]. Due to these multipurpose requirements, one tness test can hardly be used as an ideal tool able to provide useful information for all aspects of tness testing. This has led to the development of numerous eld aerobic tness tests generally measuring the same generic tness quality (i.e., maximal aerobic power), but with better or limited applicability for the speci c purpose of testing. Sport-speci c tness tests, for example, appear to show greater ecological validity, but, on the other hand, have limited convergent validity. This makes them appropriate for speci c- tness assessment, but rather poor in assessment of basic tness capacities (e.g., the maximal oxygen uptake (VO2max)) [8]. Similarly, some tests present with limited practical validity and can, therefore, hardly be used to accurately prescribe exercise which is probably the most important part of strength and conditioning coaches' job [3]. Although laboratory incremental exercise test is considered a “gold standard” for testing VO2max[9], eld aerobic tness tests have emerged as time and resource-saving alternatives. During the last four decades, several eld tests, including the University of Montreal track test (UMTT) [10], along with its modi cation the Vam Eval test [11], J. Funct. Morphol. Kinesiol.2021,6, 69.

a “gold standard” for testing VO2max[9], eld aerobic tness tests have emerged as time and resource-saving alternatives. During the last four decades, several eld tests, including the University of Montreal track test (UMTT) [10], along with its modi cation the Vam Eval test [11], J. Funct. Morphol. Kinesiol.2021,6, 69.

J. Funct. Morphol. Kinesiol.2021,6, 69 2 of 23 multistage 20-m shuttle run test (20mSRT) [12], Yo-yo intermittent recovery test level 1 and 2 (Yo-YoIRT1 and 2) [8] and 30-15 intermittent tness test (30-15IFT) [13] have gained popularity and are widely used in practice for the assessment of aerobic tness. These tests are different in nature as they include multistage continuous straight-line[10,11] , shuttle [12] and intermittent shuttle [8,13] runs to exhaustion. Due to differences in exe- cution, these eld tests provide different end-test speeds which are speci c to the nature of the effort made during the test. Speci cally, introducing changes of direction every 20 m into straight-line running yields higher oxygen uptake (VO2), heart rate, blood lactate concentration and Rating of Perceived Exertion responses [14,15] which leads to exhaustion at signi cantly lower end-test speeds during shuttle tests in comparison to incremental straight-line tests [16]. Similarly, omitting inter-effort recoveries while performing the 30-15IFT results in reaching exhaustion at signi cantly lower end-test speed [17]. For similar levels of aerobic tness exhaustion will be reached at the lowest running speed in 20mSRT while UMTT or Vam Eval test, Yo-YoIRT and 30-15IFT will have their end-test speeds higher by approximately 2 km/h interval each. This would, for example, make the end-test speed of 30-15IFT approximately 6 km/h greater than the one reached in 20mSRT. Since velocity associated with VO2max(vVO2max) is the preferred method for prescribing exercise intensity for high-intensity interval training (HIIT) [18] this measure should be obtained through a eld test which closely mimics the locomotor activity of a certain HIIT format in order to make it usable for prescription. However, not all mentioned eld tests are speci c to the HIIT formats, so their end-test speeds cannot easily be used for training prescription purposes. Field tests are very popular as they are less time-consuming and cheaper than tests performed in the laboratory. However, the large number of available eld tests has con- tributed to a growing confusion among coaches about which one to use. All mentioned eld tests were nominally created for aerobic tness assessment. This has led

be used for training prescription purposes. Field tests are very popular as they are less time-consuming and cheaper than tests performed in the laboratory. However, the large number of available eld tests has con- tributed to a growing confusion among coaches about which one to use. All mentioned eld tests were nominally created for aerobic tness assessment. This has led coaches to believe that the tests are basically the same and that the choice can be made solely on preference. However, each test had been developed with a speci c purpose and, as such, should be used if and when it has the best metric characteristics for a certain aspect of test- ing. Therefore, the main aim of this paper is to review the available scienti c literature for the purpose of reporting and discussing the reliability, validity, sensitivity and usefulness of the most commonly used eld aerobic tness tests. This will enable the formation of practical guidelines for their proper use in soccer (football). This paper used a narrative review format. In order to retrieve relevant scienti c papers, we searched the Web of Science and PubMed databases using standard search criteria. After accounting for the already retrieved publications, the keywords mentioned in the abstract were used to search for additional scienti c papers. Reference lists of retrieved articles and recently published reviews were examined to nd additional papers not identi ed by the keywords-based search. Only full-text articles published in English were included in the review. The searching process included articles retrieved until 1 March 2021. 2. Assessment of Maximal Oxygen Uptake The general importance of aerobic tness in soccer is well documented [19,20]. How- ever, even though strength and conditioning coaches seek information about their player's VO2max, recent studies show that matching running performance might not be affected by aerobic tness capacities [21–23] as much as previously reported [24–27]. Namely, it appears that playing position and game tactics are more important factors in determining how much a player will run during a match than physical tness [22]. This is further supported by the ndings that improvements in

recent studies show that matching running performance might not be affected by aerobic tness capacities [21–23] as much as previously reported [24–27]. Namely, it appears that playing position and game tactics are more important factors in determining how much a player will run during a match than physical tness [22]. This is further supported by the ndings that improvements in aerobic tness do not necessarily re ect in improvements in high-intensity match running performance in young soccer players [28]. However, aerobically tter players experience reduced individual running demands during the game which is bene cial in terms of reducing the overall fatigue and injury risk [29], as well as the impairment of technical performance [30]. So, even though aerobic tness might

J. Funct. Morphol. Kinesiol.2021,6, 69 3 of 23 not be the primary limiting factor for match running performance in soccer players [21–23], since players cover 9 to 12 km in total distance, perform 150 to 350 high-intensity running activities, and execute 50 to 100 accelerations above 2.5 m/s 2 with 300 changes of directions during the match [31], adjusted aerobic conditioning should de nitely be implemented in the overall training program. Well-developed aerobic tness will enable players to perform their technical and tactical requirements with less physiological load [22] and to quickly recover between high-intensity efforts [32] which have been shown to be typical activities for team sports and especially soccer [27,33]. Probably the most important reason for tness testing is the assessment of physical capacities and abilities [1,3]. Despite recent scienti c evidence related to its association with soccer match performance, when it comes to aerobic or cardiorespiratory tness, many strength and conditioning coaches focus on VO2max. However, it appears that rather than VO2maxper se, match running performance is more related to vVO2max[23] and peak incremental test speed [21]. Both variables represent an integrated measure of VO2maxand running economy, and can therefore be considered an athlete's peak aerobic locomotor ability [7]. This is further supported by the fact that repeated sprint ability (RSA) [34] shows much larger association with peak incremental test speed than VO2maxin team sport players [35]. Therefore, assessing peak aerobic locomotor ability seems to be much more important than VO2maxas it provides a more ecologically valid measure of aerobic tness and can also be used for prescribing exercise. Apart from being less important than end-test speed for assessment of aerobic tness in team sports, another reason disputing the assessment of VO2maxthrough test equations is their low prediction accuracy due to the fact that the calculation presumes a standard running economy. While VO2maxcan be reached in all eld tests when executed to exhaus- tion [12,36,37], the calculation of VO2maxusing performance data obtained through the test shows different levels of accuracy among the tests. Being the most similar to laboratory incremental exercise tests, the UMTT shows the largest

prediction accuracy due to the fact that the calculation presumes a standard running economy. While VO2maxcan be reached in all eld tests when executed to exhaus- tion [12,36,37], the calculation of VO2maxusing performance data obtained through the test shows different levels of accuracy among the tests. Being the most similar to laboratory incremental exercise tests, the UMTT shows the largest correlation with VO2max(r = 0.96) and the lowest standard error of estimate (SEE) of 2.8 mL/kg/min [10] and therefore seems the best option for estimation of VO2max. As originally developed for the assessment of VO2maxin limited spaces such as gyms, the 20mSRT also has a high level of criterion-related validity with SEE of 3.5 mL/kg/min in adults [38] and 4.7 and 5.9 mL/kg/min in healthy adults and children, respectively [39]. This indicates a higher validity for adults than for children [40]. On the other hand, both Yo-YoIRTs and 30-15IFT present lower correlation coef cients and limits of agreement with VO2maxand, therefore, may not be ideal for estimation of VO2max[8,13,41,42]. However, it seems that criterion-related validity of the tests might be tness level dependent as higher correlations and lower SEE of Yo-YoIRT1 were reported for recreational [43] and untrained individuals [44]. 3. Assessment of Speci c Intermittent Endurance Soccer and most other team sports are intermittent activities with high aerobic de- mands [20] placed on players due to frequent changes in types of movement [31] and repetition of high-intensity running and sprinting [45]. Even though high-intensity run- ning can be maintained throughout the match [22], the decrease in occurrences of repeated sprint sequences and number of sprints within a sequence [46] suggests the accumula- tion of fatigue over the course of a match which may negatively impact players' overall physical and technical match performance [47]. As high-intensity running appears to be an important index of match-related physical performance [27], assessing player's ability to repeat such activities and to recover from them quickly seems important. This has led to the development of the Yo-YoIRTs devised with the main purpose of assessing soccer- speci c intermittent endurance [8]. Indeed, signi cant correlations between

technical match performance [47]. As high-intensity running appears to be an important index of match-related physical performance [27], assessing player's ability to repeat such activities and to recover from them quickly seems important. This has led to the development of the Yo-YoIRTs devised with the main purpose of assessing soccer- speci c intermittent endurance [8]. Indeed, signi cant correlations between Yo-YoIRT1 and high-intensity running during the match have been found in young [23,48–50] and senior level [24,26] soccer players. Large and very large signi cant correlations between Yo-YoIRT1 and high-intensity running (r ranging from 0.56 to 0.76) [8,23,24,26,48–50], very high-intensity running (r = 0.59) [50], sprinting (r = 0.63 and 0.76) [23,49], total distance

J. Funct. Morphol. Kinesiol.2021,6, 69 4 of 23 covered (r ranging from 0.53 to 0.65) [24,26,48,50] and high-intensity activity (r ranging from 0.56 to 0.77) [23,26,48,49] performed during the match seem to support the ecological validity of the test [44,51]. The same applies for the Yo-YoIRT2 as a very large correlation (r = 0.72) was obtained between Yo-YoIRT2 and peak high-intensity running in a 5-min period during the match [8]. However, the signi cance of these correlations has lately been brought to question [52–54] as these analyses were performed on pooled data from all the players in a team. This resulted in neglecting the sometimes-substantial differences in physical tness [21,55] as well as the often-substantial differences in match running performances [21,22,56] between players from different playing positions. Namely, even though signi cant correlations were found on pooled data, when analyzed according to playing position, the associations between aerobic tness and match running performances were actually trivial and non-signi cant, with the only exception of strikers [21]. This suggests that tactical roles dictated by playing positions as well as other contextual factors such as score line, team formation and opponent quality rather than physical tness are primarily important in determining player's match running performance [21,53]. Addi- tionally, it is also interesting to notice that other eld tests, such as UMTT and 20mSRT, which are not initially designed to assess speci c intermittent aerobic endurance, also show large to very large correlations with high-intensity running [27,49], very high-intensity running [21,27] and high-intensity activity [49]. In fact, in young soccer players, signi cant correlations with high-intensity running (r = 0.70 vs. 0.65) and high-intensity activity (r = 0.75 vs. 0.73) were greater for 20mSRT than Yo-YoIRT1, raising doubt to the superiority of Yo-YoIRT1 in terms of ecological validity [49]. Generally, these ndings point out that aerobic tness is not a major limiting factor of match running performance [21,52] and that assessing speci c intermittent endurance obviously does not provide an additional bene t in assessing player's physical tness [3]. However, when choosing the test for the purpose of aerobic tness assessment of soccer

in terms of ecological validity [49]. Generally, these ndings point out that aerobic tness is not a major limiting factor of match running performance [21,52] and that assessing speci c intermittent endurance obviously does not provide an additional bene t in assessing player's physical tness [3]. However, when choosing the test for the purpose of aerobic tness assessment of soccer players, the choice of the test used should be based on the player's age, their aerobic tness level and testing time-point. Namely, it has been shown that high-levels of VO2are reached early into the Yo-YoIRT1 and that almost half of the test duration is executed with VO2above 95% VO2max[37]. This is quite different from the VO2response elicited during continuous tests in which VO2kinetics appears to be fairly linear. This means that YoYoIRTs are more metabolically demanding [8,38]. It seems that the difference between ndings obtained in the continuous and intermittent tests increases as players improve their VO2max, their anaerobic capacity and the ability to recover quickly following a high-intensity run so that their anaerobic capacity could be expended slowly during the execution of the test. Indeed, the anaerobic contribution to the intermittent test is higher than during a continuous test as blood lactate concentrations and end-test velocities are signi cantly higher after the intermittent in comparison to the continuous test [17]. This possible differentiation between the 20mSRT and Yo-YoIRT1 as players get tter is further supported by the almost perfect correlation (r = 0.89) in very young soccer players [49], while slightly lower correlations were observed for adults and elite athletes [43]. Therefore, the literature suggests that 20mSRT should be used with younger and less aerobically t players as the protocol involves lower starting speeds and smaller increments at the beginning of the test, while Yo-YoIRT1 is more useful for aerobically tter players and during the in-season period when certain level of conditioning has already been reached [49]. The shorter testing time for Yo-YoIRT1 in young soccer players also makes it a better option for in-season period when time devoted to testing is limited. On the other hand, a very

beginning of the test, while Yo-YoIRT1 is more useful for aerobically tter players and during the in-season period when certain level of conditioning has already been reached [49]. The shorter testing time for Yo-YoIRT1 in young soccer players also makes it a better option for in-season period when time devoted to testing is limited. On the other hand, a very large signi cant correlation obtained between 30-15IFT and mean sprint time of the RSA test (r = 0.88) [57] and Yo-YoIRT1 (r = 0.75) [58] suggest that even 30-15IFT can be used for evaluation of speci c intermittent endurance even though the test was not created for that particular reason [13,42].

This review examines the reliability and usefulness of field aerobic fitness tests in soccer.