The rice stink bug is a major pest of rice in Texas, causing quality related damage. The previous threshold used for assisting in rice stink bug spray decisions lacked flexibility in economic and production decision variables and neglected the dynamics of the pest population. Using stochastic dynamic programming, flexible economic thresholds for the rice stink bug were generated. The new thresholds offer several advantages over the old, static thresholds, including increased net returns, incorporation of pest dynamics, user flexibility, ease of implementation, and a systematic process for updating.