This article investigates the behavior of return ranks of different period lengths for four sets of consistent out- and underperformers in two well-controlled peer groups of large blend and intermediate-term bond mutual funds. It shows that short-term return ranks are increasingly biased toward the median, even when the underlying fund performance is stable. Short-term ranks thus systematically overestimate the performance of consistent underperformers and vice versa. This complicates their use for performance evaluations. This article describes how the bias of short-term ranks arises, how it affects the practice of performance measurement, and how these shortcomings can be surmounted. In the course of doing this, the article defines a new measure of risk-adjusted performance, which assesses the strength and consistency of outperformance, and it describes a helpful improvement to Philip’s statistical control chart approach to performance measurement (https://ssrn.com/abstract=371121).

The Right and Wrong of Ranks: Why Short-Term Ranks are Poor Performance Proxies and What to do About It
Armin Grueneich, Ph.D.