We develop a two-factor mean reverting stochastic model for forecasting storable commodity prices and valuing commodity derivatives. We define a variable called “normalized excess supply” based on the observable production rate, consumption rate, and inventory levels of the commodity. Our analysis indicates a strong inverse correlation between normalized excess supply and crude oil spot and futures prices. Our first stochastic factor is normalized excess supply, following a mean reverting process. The second stochastic factor is the deviation of prices from a mean level determined by normalized excess supply. Moreover, we develop valuation models for futures and options on futures contracts based on the two factor model. We apply this model to crude oil prices from 1995 to 2017 via a Kalman filter. We perform out-of-sample tests for forecasting spot prices. Additionally, we develop a scenario analysis framework for incorporating variant views of future supply, demand, and inventories in risk management and valuation of commodity related financial products. We show the utility of the model for investment management professionals and risk managers aiming to incorporate macroeconomic conditions in valuation and risk management endeavors.