The European Central Bank published a working paper that develops a method for incorporating full probability histograms from surveys of professional forecasters into model based predictive distributions. The approach uses entropic tilting to reweight complete economic scenarios so they exactly match survey probabilities while otherwise remaining as close as possible to the model’s original distribution and preserving its economic relationships. Applied to U.S. Survey of Professional Forecasters data from 1996 through 2025 and forecasts from a Bayesian vector autoregression with time varying volatility, the method improved point and density forecast accuracy. Gains were strongest around the Great Recession and the COVID-19 pandemic and extended beyond targeted variables such as GDP growth, unemployment and core inflation to employment, investment and longer term interest rates. Direct histogram tilting performed about as well as tilting to fitted distribution moments, while requiring fewer assumptions, using less computation and avoiding occasional difficulties in matching fitted moments.