Water markets are a promising method for adapting to water scarcity in the western United States, and the Colorado-Big Thompson Project (CBT) market is often held up as a prime example of their potential. While much has been written about the CBT market, the current academic literature tends to eschew structural modeling of supply and demand in favor of fitting hedonic price equations, which ignore many of the market’s unique characteristics. This paper proposes a model of supply and demand for CBT water which accounts for these unique features, including transaction supply, municipality stockpiling, and differences in behavior across different types of water users. The estimation of this model is made possible by novel administrative records data on both transfers and ownership of CBT water, the processing and features of which are described in detail. While the voluminous and messy nature of the data has prevented complete estimation of the model at this point, some preliminary results are presented along with a plan for future work.