A major use of crop models is to evaluate management strategies. An important question is how accurate models are for such evaluations. The purpose of this study was to determine how to use a combined crop and decision model to evaluate irrigation strategies for corn (maize, Zea mays L.) and to estimate the uncertainty in the criteria used for evaluation. The uncertainty estimation has three steps. First, the sources of uncertainty are identified. We considered uncertainty in the model parameters and model residual error. Second, the uncertainty in each source is quantified. We used a Bayesian approach to obtain a posterior distribution of the model parameters and variances of residual error. Finally, the uncertainties are propagated through to the quantities of interest. In our case, this included calculations for observed quantities—these posterior predictive checks allowed us to verify that our uncertainty estimates were reliable—and predictions of the criteria used to evaluate the irrigation strategies. We considered several criteria including multiyear average yield and interannual yield variability. The uncertainty in average yield was quite small (standard deviation of about 0.2Mg/ha). This is due to the fact that much of the error in yield prediction cancels out when looking at average yield. Three major conclusions are that this model can be a powerful tool for evaluating irrigation strategies, that prediction of average yield can have much less uncertainty than prediction of yearly yield, and that it is essential to verify the reliability of uncertainty estimates using data.