The output gap-which measures the deviation of actual output from its potential-is frequently used as an indicator of slack in an economy. This paper estimates the Finnish output gap using various empirical methods. It evaluates these methods against economic history and each other by a simulated out-of-sample forecasting exercise for Finnish CPI inflation. Only two gap measures, stemming from a frequency domain approach and the Blanchard-Quah decomposition, perform better than the naïve prediction of no change in inflation-but do not improve upon a simple autoregressive forecast. The pronounced volatility of output in Finland makes it particularly difficult to estimate potential output, producing considerable uncertainty about the size (and sign) of the gap.