This paper combines a data rich environment with a machine learning algorithm to provide new estimates of time-varying systematic expectational errors (belief distortions) embedded in survey responses. We find that distortions are large on average even for professional forecasters, with all respondent-types over-weighting their own beliefs relative to other information. Forecasts of inflation and GDP growth oscillate between optimism and pessimism by large margins, with over-optimism associated with an increase in aggregate economic activity. Biases in expectations evolve dynamically in response to cyclical shocks. Biases about economic growth display greater initial under-reaction while those about inflation display greater delayed over-reaction.