This article uses data from the 1996 Summer Olympic Games and the 2002 Olympic Winter Games to test the predictions of regional input-output models. Real changes associated with these events are insignificant. Nominal measures of demand overstate demand increases and factor price increases absorb the impact of real increases in demand. Nominal changes appear to be limited to hotel prices. Input-output models of a regional economy are often used to predict the impact of short-duration sporting events. Because I-O models assume constant factor prices and technical coefficients between sectors are calibrated from long-run steadystate relations in the regional economy, the predictions greatly overstate the true impact. Because the predictions of these models are increasingly used to justify public subsidies, understanding these deficiencies is crucial.