Sophisticated investors rely on scenario analysis to select portfolios. We propose a new approach to scenario analysis that enables investors to consider sequential outcomes. We define scenarios not as average values but as paths for the economic variables. And we measure the likelihood of these paths on the basis of the statistical similarity of the paths to historical sequences. We also use a novel forecasting technique called “partial sample regression” to map economic outcomes onto asset class returns. This process allows investors to evaluate portfolios on the basis of the likelihood that the scenario will produce a certain pattern of returns over a specified investment horizon.