Differences in timber market models that arise from alternative assumptions and fundamental differences in theory and structure can lead to disparate published results, as well as confusion among policy analysts and industry. This paper presents a thorough comparison of four widely used models, and then analyzes a set of published market predictions relative to historical data. The goal is to understand how fundamental differences between the models may affect market predictions. Several conclusions are drawn. First, market outlooks are largely a function of exogenous demand growth and timberland management assumptions, although differences in theory and structure are important. Second, spatial equilibrium models adapt to changing conditions by shifting price growth from region to region, while dynamic optimization models shift harvest quantities. Finally, model output must be assessed carefully, and often with a thorough understanding of input assumptions and alternative scenarios.