Author(s): VAN DER VEGT, N., DEN HARTIGH, R., NEUMANN, N., BRINK, M., Institution: UNIVERSITY OF GRONINGEN, Country: NETHERLANDS, Abstract-ID: 1631

INTRODUCTION:
Adversarial growth (AG) is increasingly conceptualised as a dynamic process within complex dynamical systems approaches to athlete development (1). From this perspective, transitions in an athlete’s psychophysiological state may be preceded by periods of increased instability, referred to as critical fluctuations. While critical fluctuations have been explored as early indicators (EI) in relation to sports injuries, their potential relevance for AG, that is, positive performance improvement following adversity, remains unexplored. Therefore, this study examined whether critical fluctuations could serve as an EI for AG in elite youth football players.
METHODS:
Participants were 29 elite youth football players. Daily psychological and physiological self-report data, together with GPS and heart rate data, were collected on every training and match day across two competitive seasons. This resulted in an average of 306 observations per player (range = 84-430). AG was operationalised as positive transitions in variables related to the athlete’s psychophysiological state (i.e., confidence, perceived recovery and perceived performance), identified using the Pruned Exact Linear Time (PELT) algorithm. The Dynamic Complexity (DC) index, a metric of critical fluctuations, was calculated for each variable using a moving-window approach. Significant DC peaks were identified and aggregated into cumulative complexity peaks (CCPs) using one-sided z-tests. AG was considered correctly predicted if at least one CCP occurred within a predefined time window preceding a positive transition. A F1-score was calculated to account for the class imbalance arising from the lower frequency of AG events relative to non-AG events.
RESULTS:
At least one CCP occurred prior to 35% of AG events within the 10 observations (roughly two weeks) preceding the positive transition. Critical fluctuations showed a specificity of 92%, indicating a high rate of correctly classified non-AG observations. The F1-score was low (0.13), reflecting limited overall discriminative performance of the model due to a high false positive rate.
CONCLUSION:
The findings provide support for studying AG from a complex dynamical systems perspective. This study offers proof of concept that critical fluctuations may function as EI for AG in elite youth football players. Rather than exclusively indicating maladaptive outcomes such as injury, periods of increased instability may reflect windows of opportunity during which athletes reorganise toward higher-functioning states (2). The relatively high false positive rate indicates that critical fluctuations may reflect broader system transitions beyond injury and adversarial growth. Future research should further examine the role and interpretation of critical fluctuations in athletes’ psychophysiological states and explore how practitioners can act on EI to facilitate adaptive development.
(1) den Hartigh et al., 2022; (2) Neumann et al., 2024