The maximal and sub-maximal versions of the Yo-Yo intermittent endurance test level 2 are simply reproducible, sensitive and valid

Abstract

We appreciate Drs. Mendez-Villanueva and Buchheit’s interest in our work (Bradley et al. 2011b) and agree that the relationship between physical capacity and match performance of soccer players is complex. However, there are several flaws in their critique of our study (MendezVillanueva and Buchheit 2011) and some of them need to be corrected as a matter of record. The first point raised by the Letter makes the assertion that our article incorrectly concludes that the Yo-Yo intermittent endurance test level 2 (Yo-Yo IE2) is a valid assessment tool for elite soccer populations. Although no exact measure of physical performance in elite soccer matches exists, the total distance and that covered at high-intensity provide useful indicators (Mohr et al. 2003). We observed a significant correlation between the Yo-Yo IE2 test performance and both total distance (r = 0.74, P \ 0.01) and high-intensity running during a game (r = 0.58, P \ 0.05). At this point, it is worth reminding the authors of the Letter that the strength of the correlations are large to very large, based on the classifications of Cohen (1988) and Hopkins et al. (2009). Thus, it is unclear why Drs. Mendez-Villanueva and Buchheit suggest that this conclusion is erroneous, especially given that their research group has recently based inferences on similar correlation strengths (Mendez-Villanueva et al. 2011). The lower correlation observed for high-intensity running is hardly surprising given that this variable can differ substantially from game to game (Gregson et al. 2010). However, we make reference to this point in the Discussion. Recently, we observed a stronger correlation between Yo-Yo IE2 test performance and high-intensity running (r = 0.67, P \ 0.01) using a larger number of elite players over multiple games (Bradley et al. unpublished data). Another interesting finding from this study was a correlation between the changes in Yo-Yo IE2 test performance and the change in high-intensity running during match-play across the season, whereas this was not evident for maximal oxygen uptake ( _ VO2 max). In further support of the validity of the Yo-Yo IE2 test, other studies have found correlations between Yo-Yo IE2 test performance and high-intensity running during a match for elite futsal referees (r = 0.77, P \ 0.05; Rebelo et al. 2011) and the fatigue index exhibited by elite female soccer players during repeated sprinting (r = 0.76, P \ 0.05; Krustrup et al. 2010). We also refute the assumption that we make inferences in our article about ‘predicting’ match performance from Yo-Yo IE2 test performance. This interpretation is not only a distortion of our findings but also an error of fact as we did not make this suggestion anywhere in the article. However, we do correctly identify that, due to the scattered trend exhibited by the correlation between Yo-Yo IE2 test performance and _ VO2 max, it would be erroneous to predict performance in one variable using the other. We particularly made reference to the test’s lack of predictive ability for _ VO2max rather than high-intensity running, given that this is a common question Communicated by Susan A. Ward.