Widely different Arrhenius parameters have been published for thermal decomposition of HMX (octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine), RDX (1,3,5-trinitrohexahydro-s-triazine), and NTO (3-nitro-1,2,4-triazol-5-one). Evaluation of these data reveals that an approximately linear relationship exists between In A and E(a) for each compound irrespective of the phase. This kinetic compensation effect accounts for and unifies most of the differences in the reported rates. The range of data is qualitatively attributable to differences in the sample characteristics and experimental conditions. All E(a)-ln A sets lying on or close to the compensation regression line consistently account for the rate of thermal decomposition under the conditions reported. E(a)-ln A sets not lying on or near the compensation regression line represent a different process or are incorrect. The kinetic compensation effect provides a rational method to select Arrhenius parameters to describe the global thermal decomposition rate of an explosive.