This paper documents a decline in aggregate hedge fund performance over the past decade. We test whether a set of prediction models can select subsets of individual funds that buck the trend and subsequently outperform. Two of the predictors reliably pick funds that lower the volatility and raise the Sharpe ratio of a multi-asset class portfolio relative to a stock/bond portfolio over the full 1997–2016 sample. Hedge fund allocations reduce volatility across two sub-periods but fail to improve the Sharpe ratio from 2008 onwards. Potential explanations for the erosion of hedge fund performance are explored.