AbstractFor decades, the DoD has employed numerous reporting and monitoring tools for characterizing the acquisition cost estimates of its major programs. These tools have resulted in dozens of studies thoroughly documenting the magnitude and extent of DoD acquisition cost growth. However, little attention has been paid to the behavior of the larger cost component of a system's life cycle cost: Operating and Support (O&S) costs. Consequently, the DoD has little knowledge of how O&S cost estimates behave over time, and virtually no insights regarding their accuracy. In a previous paper, the authors describe an analytical methodology for remedying this deficiency via a study to characterize the historical accuracy of O&S cost estimates. The results of this study are presented here, and indicate there tend to be large errors in DoD O&S cost estimates, and that the accuracy of the estimates improves little over time. However, the results also suggest the possibility of improving the accuracy of these estimates via a stochastically‐based predictive cost model.