We examine the application of quantitative spatial models to the growing body of fine spatial data used to study economic outcomes for regions, cities, and neighborhoods. In “granular” settings where people choose from a large set of potential residence-workplace pairs, idiosyncratic choices affect equilibrium outcomes. Using both Monte Carlo simulations and event studies of neighborhood employment booms, we demonstrate that calibration procedures that equate observed shares and modeled probabilities perform very poorly in such settings. We introduce a general-equilibrium model of a granular spatial economy. Applying this model to Amazon’s proposed HQ2 in New York City reveals that the project’s predicted consequences for most neighborhoods are small relative to the idiosyncratic component of individual decisions in this setting. We propose a convenient approximation for researchers to quantify the “granular uncertainty” accompanying their counterfactual predictions.