Resilience is a key construct to understand when athletes continue to perform optimally, or when they break down. Although there is consensus that resilience can be conceptualized as a dynamic process, it remains an open question whether studying such a process on a group level adequately represents the individuals within a given sample. As a first step to answer this question, we designed a diary study to test whether the statistics for repeated assessments of protective factors and resilience can be generalized from group-level trajectories to the individuals. By tracking resilience and the protective factors over 21 days in athletes, we found divergent patterns of group-level and individual-level statistics for the repeated assessments. This so-called “ergodicity problem” implies that the individual, rather than the group, should be placed at the level of analysis to avoid wrong conclusions and ineffective interventions on their resilience.