Classic net energy (CNE) is CNE = ME – heat increment (HI), being HI = fed (HP) – fasting (FHP) heat production. Productive energy (PE) measures net energy for gain (NEg) and maintenance (NEm) and is superior to CNE for predicting performance and economics. In vivo and in silico experiments (E) in this study assessed the reliability of CNE assumptions: A1) diet alone determines NE, A2) FHP is a constant of metabolic body weight (MBW), A3) broiler MBW power is 0.70, A4) diet does not influence FHP, A5) old FHP values work, A6) crude protein (CP) only increases HI, A7) HI is the only effect of diet on energy metabolism and A8) CNE validation studies are informative. In E1, 2400 broilers were fed 3 CP levels to induce performance and body composition (Dual-Energy X-Ray Absorptiometry; DEXA) changes. FHP was measured in calorimetry chambers (CHA). Mixed models determined the influence of body composition on FHP and MBW power. In E2, birds were fed 12 diets varying in TDAA and digestible nutrients. Performance, DEXA, and CHA were run (HP, FHP, CNE, PE, body composition). Data was modeled to determine CNE and PE variations associated with other variables, and the influence of changes in TDAA and CP on performance, body composition, and HI, CNE, NEg, NEm, and PE. In E3, a literature meta-regression model reporting increasing broiler HP through the years determined the deviation in FHP in CNE studies using old FHP values. A mechanistic model built in E4 integrated findings explaining divergent influences of CP on CNE and PE. Linear mixed models modeled the data and Tukey test (P<0.05) or mean 95% confidence interval (95CI) compared mean values. A model from E1 (R2=0.98) showed FHP a function of body composition (P<0.001), age (P=0.039), and body weight (BW; P<0.001) or MBW (P<0.001), proving A1 and A2 unsupported. E1 showed fitting metrics remain the same with MBW or BW if body composition is included, implying body composition explained the MBW concept and that A3 was unsupported. E1 also showed the determined MBW was not different than 0.75, based on its 95CI (P>0.05), also implying A3 was unsuported. E2 showed differences (P=0.001) across treatments, implying A4 was unsupported. The meta-regression model from E3 indicated HP increases by 2.68 kcal/kg BW0.75/d each year, implying old FHP overlooked a 53% shift in FHP in 23 years and that A5 was unsupported. The mechanistic model in E4 revealed CP increases HI but also improves protein turnover, feed conversion ratio, and NEg and NEm, proving A6 and A7 unsuported.