The remainder of the paper is as follows. The next section defines the standard approach and the notion of luck. Then, we define the F DR and explain our new methodology which allows us to compute the F DR among the best and worst funds separately. Section 3 presents the performance measures, the estimation technique to compute the p-values as well as the mutual fund data. Section 4 contains the empirical analysis of the impact of luck on performance across the four investment categories. Section 5 concludes. An appendix gathers proofs and results of a Monte-Carlo study on the accuracy of our new measures of luck