The current study investigated the relationship between the frequency of watching sports and depressive symptoms among older adults. This study used cross-sectional data from the Japan Gerontological Evaluation Study, a nationwide mail survey of 21,317 older adults. Depressive symptoms were defined as a Geriatric Depression Scale score of ≥ 5. Participants were queried regarding the average frequency at which they watched sports on-site and via TV/Internet over the past year. Among the 21,317 participants, 4559 (21.4%) had depressive symptoms, while 4808 (22.6%) and 16,576 (77.8%) watched sports on-site and via TV/Internet at least once a year, respectively. Older adults who watched sports on-site a few times/year (prevalence ratio, 0.70; 95% confidence interval, 0.65–0.74) or 1–3 times/month (0.66, 0.53–0.82) were less likely to have depressive symptoms compared to non-spectators after adjusting for frequency of playing sports, exercise activities, and other potential confounders. Meanwhile, a dose–response relationship was confirmed for watching via TV/Internet (prevalence ratio of 0.86, 0.79, and 0.71 for a few times/year, 1–3 times/month, and ≥ 1 time/week, respectively). This study suggested that watching sports on-site or via TV/Internet, regardless of whether they regularly engage in sports, may reduce the risk of depressive symptoms among older adults.