We would like to thank our colleagues who have written to comment 1,2 on our recent review about the validity of the training load (TL) concept, 3 particularly given their extensive previous scientific and practical contributions on this topic. We are pleased to have the opportunity to extend discussion of the topic further in this reply. We suspect that how TL is conceptualized may lead to important differences in perspective as we find ourselves in broad agreement with several of the points raised in both letters. Therefore, we will respond by briefly outlining our interpretation of the TL concept and the rationale for linking this to an acute performance decrement (APD) and then addressing some specific points that were raised.Most discussions of the concept of TL begin with the seminal work of Banister et al, 4 who proposed that training-induced fatigue could be linked to subsequently observed gains in performance (fitness). Banister's work was based on fundamental principles of training theory, 5,6 and therefore, we conceptualized the TL in the same manner but as a single training dose. The principles of training theory propose that a training session constitutes an overload that causes a training stress and fatigue, which, in turn, provides a signal for adaptation or overcompensation that occurs during the ensuing recovery period (Figure 1). The training session is considered a training dose, and the resultant training adaptations increase fitness and enhance performance. Indeed, the representation of this training process is often described by its effects on performance, that is, an APD occurs in response to the training dose, and then in recovery, performance is restored progressively to an augmented level (Figure 1). Banister et al's 4 thesis was that chronic changes in performance could be modeled as the integrated effects of training fatigue and fitness. As benefits to performance Passfield