PURPOSE: To compare weekly training and game demands according to playing position in basketball players. METHODS: A longitudinal, observational study was adopted. Semiprofessional, male basketball players categorized as backcourt (guards; n = 4) and frontcourt players (forwards/centers; n = 4) had their weekly workloads monitored across an entire season. External workload was determined using microsensors and included PlayerLoad™ (PL) and inertial movement analysis variables. Internal workload was determined using heart rate to calculate absolute and relative summated-heart-rate-zones workload and rating of perceived exertion (RPE) to calculate session-RPE workload. Comparisons between weekly training and game demands were made using linear mixed models and effect sizes in each positional group. RESULTS: In backcourt players, higher relative PL (P = .04, very large) and relative summated-heart-rate-zones workload (P = .007, very large) were evident during training, while greater session-RPE workload (P = .001, very large) was apparent during games. In frontcourt players, greater PL (P < .001, very large), relative PL (P = .019, very large), peak PL intensities (P < .001, moderate), high-intensity inertial movement analysis events (P = .002, very large), total inertial movement analysis events (P < .001, very large), summated-heart-rate-zones workload (P < .001, very large), RPE (P < .001, very large), and session-RPE workload (P < .001, very large) were evident during games. CONCLUSIONS: Backcourt players experienced similar demands between training and games across several variables, with higher average workload intensities during training. Frontcourt players experienced greater demands across all variables during games than training. These findings emphasize the need for position-specific preparation strategies leading into games in basketball teams.