Little research documents the contribution of upper limb and total body movement to energy expenditure (EE) during active video gaming. To address this, EE, heart rate (HR), and, upper limb and total body movement were assessed in 11- to 17-year-old adolescents whilst playing three active (Nintendo Wii) and one sedentary (XBOX 360) video games. Non-dominant upper limb activity, EE and HR were significantly greater during Wii Sports boxing [mean 267.2 (SD 115.8) J kg−1 min−1; 136.7 (24.5) beats min−1] than tennis or bowling (P ≤ 0.044). For all active games hip activity best predicted EE (R 2 ≥ 0.53), with two-measure models of HR and single-site activity data, and multi-site activity data, similarly explaining the variance in EE (R 2 ≥ 0.64). The physiological cost of upper-body orientated active video games increased when movement of both upper limbs was encouraged. Improvements in EE explanatory power provide support for multi-site activity monitoring during unique, non-ambulatory activities.