A new method for the computation of target scores in interrupted, limited-over cricket matches

Since the limited-over games in cricket are result-oriented, an interruption most frequently due to rains (for the team batting second), requires that a revised target is set. After initial experimentation with some ad hoc (and unsatisfactory) methods, the International Cricket Council (ICC) - the governing body of international cricket - has, since 1998, employed the Duckworth-Lewis (D/L) method for revising targets in interrupted matches. Though based on sound mathematical principles and generally satisfactory, on several occasions, the targets set using the D/L method have seemed quite inappropriate. The method presented in this article is based on mathematical models describing the natural development of the innings. The method employs the concept of normal (PAR) and target scores. Regression equations obtained from a detailed statistical analysis of a data set of closely fought matches, are used to construct easy-to-use tables for employing the method in the field, though a user-friendly, interactive computer program has also been developed. The method is capable of satisfactorily handling any number of interruptions during any stage of the game, as will be demonstrated with a few illustrative examples. More importantly, in the few situations where the D/L method seems to lead to inappropriate targets, those obtained with the present method are seen to be quite satisfactory. A large number of players, umpires, cricket administrators, critics and cricket enthusiasts who have evaluated the method seem to find it consistently superior to the currently followed D/L method.