This paper examines the methods currently employed to assess investment performance in the light of recent developments in the theory of capital asset pricing. There have been a considerable number of studies in the last decade on whether or not mutual funds' are able to achieve superior investment performance ("beat the market" in some sense). The verdict is virtually unanimous : mutual funds do not consistently outperform the market. This evidence is having a traumatic impact on the US securities industry, resulting in proposals for far-reaching changes in the management and organisation of mutual funds. The validity of these research findings is, of course, dependent on the appropriateness of the methodology underlying the empirical tests. Recent work suggests that the methodology is suspect and can yield highly misleading results. Our purpose is to assess this claim and to see whether anything can be rescued. We argue that the damage is not as serious as first appearances suggest.