A comment on the Duckworth–Lewis–Stern method

Stern (2016) documents the evolution of the Duckworth– Lewis (DL) method from its inception (Duckworth & Lewis, 1998) through the Professional edition (Duckworth & Lewis, 2004) up to the contemporary edition, the Duck worth–Lewis–Stern (DLS) method. The DL method’s innovation was the concept of a batting team’s remaining resources, a function of the overs they still have to face and their remaining wickets. If the match was interrupted, quantifying the resources each team received during the match allowed a “fair” result to be reached by observing which team scored proportionally more runs in their available resources. Stern details the increase in scoring rates over the last 25 years in international cricket and the resulting modification of the DL method. Duckworth and Lewis (2004) observed that when making a large total, teams appear to score a larger proportion of their runs earlier in their innings than the original DL method predicted. By introducing a match factor, reflecting the size of the score being chased, the Professional DL allows the chasing team’s resources to be straightened towards the average run rate (ARR) line, indicating a constant rate of run scoring throughout the innings, if a large total was required. The DLS method introduces greater flexibility, allowing the degree of straightening to depend on the number of overs remaining. The DLS method is currently applied in international cricket. Jewson (2015) conducted a similar analysis exploring, whether the same run scoring patterns are observed in English domestic cricket. The results of this analysis are briefly presented in Section 2 of this article and largely agree with Stern’s. Additionally, in Section 3 we explore what it means for a target adjustment rule to be fair, demonstrating how the goalposts for fairness were moved when the original DL was updated to the Professional method.