DLS Method Explained: How Duckworth-Lewis-Stern Calculates Rain-Revised Targets
How the Duckworth-Lewis-Stern (DLS) method works in rain-affected cricket matches — the two resources concept, why DLS produces targets that sometimes seem counterintuitive, and real examples from famous matches.
The Duckworth-Lewis-Stern (DLS) method is the mathematical formula used by the ICC to calculate revised targets in rain-affected limited-overs cricket matches. Originally developed by statisticians Frank Duckworth and Tony Lewis and first used in international cricket in 1997, the method was updated by Professor Steven Stern in 2014 and now bears his name as well. Its central insight is that a team's batting resources are determined by two factors — overs remaining and wickets in hand — rather than overs alone, which was the flaw in earlier rain-rule methods.
The Two Resources: Overs and Wickets
The DLS method's foundation is the concept that a batting team has two resources available to it: overs remaining and wickets in hand. Both matter. A team with 20 overs remaining and 10 wickets (a fresh start) has all their resources available. A team with 20 overs remaining and only 2 wickets has significantly fewer resources because they must bat conservatively to protect their remaining wickets. The DLS method quantifies what percentage of a team's scoring resources remain at any point in the innings, using a table of resource percentages calculated from large historical datasets of actual match scores. When rain interrupts play and overs are lost, the method calculates what proportion of resources each team had available and adjusts the target accordingly.
Why DLS Targets Can Seem Counterintuitive
DLS sometimes produces targets that seem unfair or counterintuitive to fans — particularly when a team in a strong position appears to have their target raised after a rain break, or when a team that was cruising suddenly needs more runs per over than before. These situations arise because DLS adjusts for the resources available, not just the runs scored. If the fielding team uses the rain break strategically — the interruption helps them by resetting conditions when the bowling team would otherwise have lost more overs in the death — the DLS target might appear to help the fielding team. In practice, DLS is neutral: it doesn't advantage either side, but it replaces what would have happened with what should have happened based on historical scoring patterns. The counterintuitive cases arise when fans compare the DLS target to the raw 'runs needed' rather than to what the team's resources suggest they should score.
Real Examples from Famous Matches
The 1992 Cricket World Cup semi-final between England and South Africa is the most notorious rain-rule failure — but it used the old 'Most Productive Overs' method, not DLS, and is partly responsible for DLS being developed. South Africa needed 22 off 13 balls when rain interrupted; when play resumed with a reduced allocation, they needed 22 off 1 ball — an impossible target that eliminated them. This absurd result could not occur under DLS. A more recent DLS example was the 2019 World Cup semi-final between India and New Zealand at Old Trafford. New Zealand scored 239/8 in 50 overs; rain intervened, and India's revised target was set at 240 in 50 overs under DLS — essentially the same target. India were ultimately dismissed for 221, but the DLS adjustment was minimal, reflecting that India had their full resources available.
Frequently asked questions
Is DLS the fairest possible rain-rule system, and what are the strongest arguments for reforming or replacing it with something else?
DLS is the best rain-rule system that has been practically implemented at international level, but 'best available' is not the same as 'optimal', and there are genuine arguments for refinement and debate about whether the underlying approach is the most equitable available. Why DLS is better than predecessors: the previous rules — Most Productive Overs, Average Run Rate — had well-documented structural flaws. MPO could produce targets that required scoring at physically impossible rates (the South Africa 1992 example). Average Run Rate simply scaled the target proportionally to overs, ignoring that teams score differently in different phases. DLS's resource model corrects both these problems by accounting for wickets lost as well as overs remaining, and by using historically calibrated tables rather than linear scaling. Legitimate criticisms of DLS: the resource percentages in DLS tables are calculated from historical scoring data — they reflect how teams have scored on average, historically. As T20 cricket and ODI cricket have evolved with faster scoring rates (particularly in powerplays and death overs), some argue the tables need more frequent updating than the Stern revision provided. Specifically, DLS may undervalue the importance of early wickets in T20 cricket where death-over hitting means teams with wickets in hand can recover more easily from slow starts than historical data suggests. The 'good wickets lost' argument: DLS treats all wickets equally in terms of resource value — losing your tenth wicket has the same resource cost as losing your first wicket from the tenth wicket's position. Critics argue this is wrong: teams with specialist batters at positions 1-6 lose more resource per wicket at the top of the order than teams with tail-enders. A team that is 50/5 has fewer resources than DLS suggests if their last five wickets are tail-enders. Practical alternatives: some researchers have proposed machine-learning-based target systems that update in real time based on pitch conditions, team batting strength, and current form — but these introduce subjectivity that DLS avoids. The ICC has prioritised the rule's simplicity and neutrality. Overall, DLS is defensible as the best system currently deployable at international level, while acknowledging it is a sophisticated approximation of fairness rather than a perfect solution.