The restrictions imposed by the COVID-19 pandemic left many older adults isolated and confined. Under active aging theory, self-management is crucial for well-being among older adults coping with aging. The current between–within subject quasi-experimental study examines how (a) initial self-management and (b) changes in self-management due to independent physical training affect psychological outcomes in a sample of care home residents following the outbreak of the pandemic. A total of 64 older adults (53 females, 11 males), with mean age of 82.23, reported on their self-management abilities and then embarked on six months of training in chair exercises (one session per week). The training exercises were halted after 22 sessions due to the pandemic, but some residents continued to practice independently. Eight weeks after the outbreak of the pandemic, residents who had continued to practice at least once per week (n = 35) and those who had not continued to practice (n = 29) were questioned again about their self-management and about five psychological outcomes: anxiety, traumatic stress, satisfaction, general mood, and post-traumatic growth (PTG). Self-management improved among older adults who independently practiced the exercises, and it declined among those who did not. Pre-pandemic self-management significantly predicted post-outbreak traumatic stress symptoms, anxiety, general mood, and satisfaction with life, but not PTG. However, the difference in self-management between the pre-pandemic and post-outbreak measures was associated with PTG, and made a unique contribution to prediction of the other effects. Self-management abilities among older adults can be seen as a protective factor against adverse psychological outcomes in times of trauma. Further, the improvement in self-management among older adults who independently practiced physical exercises made a unique contribution beyond initial self-management abilities.