A stable predictive relationship between inflation and the output gap, often referred to as a Phillips curve, provides the basis for empirical formulations of countercyclical monetary policy in many models. In this paper, we provide an empirical evaluation of the usefulness of alternative univariate and multivariate estimates of the output gap for predicting inflation. In-sample analysis based on ex post output gap measures suggests that many of the alternative estimates we examine appear to be quite useful for predicting inflation. However, examination of out-of-sample forecasts using real-time estimates of the same measures suggests that this predictive ability is mostly illusory. We find that the usefulness of output gaps as predictors of inflation has been severely overstated and that real-time forecasts using the output gap are often less accurate than forecasts that abstract from the output gap concept altogether.