This study aimed to examine the differences in associations between 24-hour movement behaviors and body composition on weekdays versus weekends in preschool children, and to estimate the potential effects of time reallocation among behaviors on body composition. A cross-sectional study was conducted among healthy children aged 3–6 years recruited from urban and suburban kindergartens in Beijing. Movement behaviors were measured objectively using accelerometers on both weekdays and weekends, ensuring comparable assessment across contexts, sleep duration was obtained from parent-completed questionnaires. Body composition was assessed using bioelectrical impedance analysis. Compositional data analysis was applied to build multivariate regression models examining weekday and weekend movement behaviors independently. Isotemporal substitution analysis was used to predict changes in body composition under different time-reallocation among behaviors. Moderate-to-vigorous physical activity (MVPA) on weekdays was significantly associated with lower percent body fat (PBF) and higher skeletal muscle mass index (SMMI), while sedentary behavior (SB) was correlated with lower soft lean mass index (SLMI), fat-free mass index (FFMI), and SMMI. Compositional isotemporal substitution analysis indicated that on weekdays, reallocating 5–30 minutes from SB to MVPA was associated with reduced PBF and higher muscle-related indices, whereas reallocating time from MVPA to SB resulted in adverse effects on these metrics, demonstrating an asymmetric pattern. No significant associations were observed between movement behaviors and body composition on weekends. The findings indicate that the association between MVPA and favorable body composition in preschoolers is significant on weekdays but not weekends, highlighting the context-dependent nature of movement behaviors in relation to body composition.