Abstract: In this paper we study the persistence and predictability of several statistical parameters of individual hedge fund returns. We find little evidence of persistence in mean returns but do find strong persistence in hedge funds? standard deviations and their correlation with the stock market. Persistence in skewness and kurtosis is low but this could be due to the small size of the sample used. Despite the observed persistence, our study also shows that in absolute terms hedge funds? risk profiles are not easily predicted from historical returns alone. The true value of a hedge fund's track record therefore appears not to lie in its use as a predictor of future performance and risk, but primarily in the insight that it provides in a fund's risk profile relative to that of other funds in the same strategy group. The availability of a track record is important, but for a different reason than many investors think.