Duckworth-Lewis-Stern Method Explained: How Rain Rules Work in Cricket
A clear explanation of the Duckworth-Lewis-Stern (DLS) method — why cricket needed a rain rule, how it calculates revised targets, its history from Frank Duckworth and Tony Lewis, and the controversies it still generates.
The Duckworth-Lewis-Stern (DLS) method is cricket's mathematical system for calculating revised targets when rain or other interruptions reduce the number of overs available to either team in a limited-overs match. Developed by statisticians Frank Duckworth and Tony Lewis and introduced in 1997, it replaced simpler and often unfair rain rules with a system based on the concept of 'resources' — the combination of wickets remaining and overs available that determines a team's scoring potential at any point in an innings.
The Problem DLS Solved: The 1992 World Cup Controversy
The inadequacy of previous rain rules was most famously demonstrated in the 1992 Cricket World Cup semi-final between England and South Africa. South Africa needed 22 runs from 13 balls when rain interrupted play. When play resumed — with one fewer over available — the 'most-run-scoring-overs' rule reduced South Africa's target such that they needed 21 runs from 1 ball: a practically impossible target from what had been a achievable one. The obvious injustice of the result — South Africa effectively eliminated by a rain rule rather than cricket — accelerated the search for a fairer system. DLS, introduced in 1997, was the result.
Resources: The Core Concept
DLS is built on the concept that a team's scoring potential depends on two resources: overs remaining and wickets in hand. A team with 10 wickets and 50 overs has 100% of their resources; a team with 10 wickets and 20 overs has roughly 68% of resources; a team with 5 wickets and 20 overs has substantially fewer resources than one with 10 wickets and 20 overs. When interruptions reduce the overs available, the DLS tables calculate what percentage of resources each team has had and adjusts the second innings target accordingly — if the second team has fewer resources available than the first team did, the target is reduced proportionally.
Why DLS Is Still Controversial
Despite being far fairer than previous systems, DLS still generates controversy. Its key criticism is that it is based on historical scoring data and does not account for current match context: if a team has lost 3 wickets cheaply in difficult early-overs conditions and a rain interruption then reduces the match, the DLS target may set a revised target that assumes those cheap wickets were lost in normal scoring conditions. Teams that are behind their par score when rain stops play often argue that the interruption came at a worse moment for them than the mathematical model captures. The model is also less intuitive to understand than simple run-rate calculations, which makes results that seem counterintuitive to fans especially controversial.
Frequently asked questions
Is DLS truly the fairest possible rain rule for cricket, or are there better alternatives that haven't been adopted, and what would a theoretically perfect rain rule look like?
DLS is widely considered the most sophisticated and fairest rain rule that has been practically implemented at scale in international cricket, but it has acknowledged limitations and there are ongoing academic discussions about whether better systems exist. DLS's fundamental limitation: it calculates revised targets based on historical average run-scoring data aggregated across many innings, but no two innings are the same. The run-scoring resources model assumes a 'typical' relationship between wickets, overs, and runs — but if conditions on the day are significantly different from the average (a very difficult pitch, exceptional bowling conditions, or a very flat surface and small ground), the model may over- or under-estimate what is achievable. These deviations are inherent in any model that relies on historical data. Alternative approaches that have been proposed include: dynamic par score systems that adjust for actual scoring rates in the match rather than historical averages; models that weight more recent scoring data more heavily; and systems that explicitly account for pitch conditions. The challenge with more complex systems is that they become harder to explain and less transparent — if a system calculates a revised target through a complex real-time algorithm, teams and spectators cannot verify the fairness of the calculation in real time. The theoretically perfect rain rule would be one that could accurately predict what each team would have scored with the full allocation of resources, given the specific conditions of that match. This is an unsolvable prediction problem — you cannot know what would have happened in the overs that didn't occur. DLS approximates it better than its predecessors with available data. Research continues, but the practical requirements of transparency, implementability, and consistency across very different match conditions make it difficult to design a materially better system that is also practically usable.