Burn rates of thermites are typically calculated in terms of an average particle size that characterizes the bulk mixture. As the particle diameter approaches the nano-scale the burn rate calculation becomes increasingly sensitive to changes in the particle diameter. In this study, burn rate estimates for nano-scale particle composite thermites are statistically evaluated in terms of an integral that employs a probability density function (pdf) for particle size distribution and a diameter dependent burn rate equation. It is shown that the burn rates depend sensitively on the mean particle diameter and the particle size distribution. Both single mode and bimodal particle size distributions were studied. The analysis shows that as the particle size is reduced to the nano-scale, the size distribution, rather than the average particle size alone, becomes increasingly important. Large variability in burn rate is associated with large standard deviations in particle size. Combining nano-scale with bulk-scale particles in a bimodal distribution does not significantly increase the burn rate as compared to a composite consisting of strictly nanoparticles. The results presented here suggest that better reproducibility of the burn rate may be achieved experimentally by selecting a material with a narrow particle size distribution.