Duckworth-Lewis-Stern (DLS) Method Explained Simply

Where DLS came from

Before the 1990s, rain-affected one-day matches used crude rules such as average run rate or "most productive overs". Both produced absurd results. The most famous was the 1992 World Cup semi-final, where South Africa needed 22 off 13 balls before a short rain delay and were left needing an impossible target off a single ball when play resumed.

Two English statisticians, Frank Duckworth and Tony Lewis, designed a fairer method that was first used in international cricket in 1997. After they retired, Australian statistician Steven Stern became the custodian of the method, and in 2014 it was renamed Duckworth-Lewis-Stern. It is now the standard for ICC limited-overs matches and most domestic competitions.

Stern still maintains and updates the model as scoring patterns change. That matters, because T20 scoring today looks nothing like one-day cricket in the 1990s, and the method has to keep pace with how teams actually bat.

The core idea: resources, not just overs

The key insight is that a batting side has two resources: overs remaining and wickets in hand. Losing overs hurts, but how much it hurts depends on wickets. Ten overs left with eight wickets in hand is worth a great deal. Ten overs left with one wicket in hand is worth much less, because the last pair cannot attack freely.

DLS expresses a team's remaining potential as a percentage of the full resources it had at the start of the innings. At the first ball of a 50-over innings, a team has all of its resources. Every ball bowled and every wicket lost uses some up. When rain removes overs, it removes the resources those overs would have provided.

The exact percentage for each combination of overs and wickets comes from a published table built from historical scoring data. Those values are not something to estimate from memory, and the Professional Edition used at the top level is run through software rather than looked up by hand.

How a revised target is worked out

The basic principle is a comparison. If the side batting second has fewer resources available than the side batting first had, its target is scaled down in proportion. If, for example, Team 2 ends up with three-quarters of the resources Team 1 used, its par score is roughly three-quarters of Team 1's total.

When Team 2 has more resources than Team 1, which can happen when Team 1's innings was cut short unexpectedly, the target goes up. In that case the method adds runs based on an average score for the level of cricket, rather than simply scaling up, so that a very low first-innings score is not multiplied into something silly.

The par score is usually a decimal. The target is the next whole number above it. If Team 2 finishes exactly on the whole-number part of par, the match is a tie.

1. Record the overs available and wickets lost at every interruption in both innings.
2. Enter those details into DLS software at each stoppage.
3. Read off the par score at that point and the revised target for any reduced innings.
4. Announce the new target and overs to both captains and the scorers before play restarts.

Par scores during an innings

During a chase, television shows a "DLS par score" that changes ball by ball. It answers one question: if play stopped for good right now, what score would Team 2 need to be on to win? Par goes up quickly when wickets fall, because each lost wicket uses up resources.

Captains watch par closely when rain is forecast. A chasing side ahead of par can afford to play safely before a storm. A side behind par needs to take risks, but every wicket lost pushes par higher still, which is a tension the method is designed to create.

Common misunderstandings

DLS does not favour the chasing side or the side batting first by design. It tries to set a target that is equally hard to reach as the original one, given what was lost. Results can still feel unfair, especially when an interruption comes at a moment that suited one team.

A minimum number of overs is needed before DLS can produce a result. Under ICC playing conditions, that is 20 overs per side in a one-day international and five overs per side in a T20 international. Below that, it is a no result.

Should local cricket use DLS?

For most club, school and gully matches, the honest answer is no. DLS assumes two teams playing to the scoring patterns of professional cricket, and it needs careful records of every interruption. A 10-over tennis-ball game played on a half-flooded ground is not what it was built for.

Simpler methods work well locally: average run rate, a target based on runs scored in the same number of overs, or a pre-agreed par table for your league. What matters most is agreeing the method before the toss. Keep an accurate over-by-over record of runs and wickets; Cricket Score Counter does this automatically, and it gives you the numbers any method needs if rain arrives.