The Duckworth Lewis problem

The basic chart of the Duckworth-Lewis method for a 50 over game... not a T20

The first T20 between India and Australia raised plenty of eyebrows, not because Australia won, but because of the peculiar scores. India’s score of 169/7 was 11 runs more than Australia’s 158/4, yet they lost by 4 runs, sparking fresh debate about the Duckworth Lewis method and how it’s applied to T20 cricket.

The method itself was designed to apply to 50 over games, taking into account par score based on two resources, wickets and overs left. For so long it has largely succeeded within the 50 over game when setting targets, but time and again it has caused problems for the shortest format. Back in the 2016 IPL, Rising Pune Supergiants were struggling at 103/6 when rain intervened, and Kolkata Knight Riders were set a mere 66 in 9 overs, an absolute doddle with the current level of ball striking.

The coach of RPS at the time Stephen Fleming was critical of the DLS method, saying ‘it’s not made for a T20 game, it’s just ridiculous really and until it’s addressed you just hope the sky doesn’t open up.’ While his team were likely to lose anyway, he does have a point. For T20 cricket, 6 overs can constitute a game, and the shorter the chasing team has to bat the more it plays into their advantage. The end result of course being that it creates unfair results.

Back in 2010 at the World T20, England posted a strong total of 191-5 against the West Indies, only to lose by 8 wickets as the Windies made 60/2 within 6 overs. This exemplifies the problem of wickets in hand for such a short format. Even when teams are set 10 an over, such a short innings and having all 10 wickets in hand enables them to play with aggression to every ball. The batting team know they’re unlikely to get bowled out and scoring at a faster rate becomes easier.

For games shortened to less than 10 overs, the method thus becomes unfair. T20 is based around run rate and not wickets, and teams that are 6 down often still score at 8-10 an over. By failing to reduce the wickets in hand for a revised score and shorter innings, it gives the top order of the chasing team free license to take a gung-ho approach. Unless the bowling team take a wicket every few balls the game is all but over. Perhaps if a team were told they had to chase 60 off 6 overs with only 6 wickets it would make a fairer contest.

Another option to consider would be to raise the minimum number of overs for a game. Deciding any fixture in 6 overs is a grave injustice to the teams and making it 10 overs would at least make it a slightly more even contest preventing a 6 over debacle taking place, where the chasing team canters home.

While matches such as India vs Australia raise the questions, it is good that it is merely a bilateral series and not a tournament. The overriding fear that comes with the use of DLS in T20 is that it will create a huge injustice at an international tournament. Rising Pune Supergiants have already seen it during an IPL semi-final, and to see an international team lose thanks to a 6 over chase in a World T20 knock out game would be a huge shame. South Africa famously were set 22 off 1 ball during the 1992 World Cup, and no one wants to see a similar scenario play out in future.

The DLS method should adapt to the T20 format, or it should no longer be used. Teams will continue to suffer, and tournaments will continue to be affected if it continues to set unfair benchmarks. In reality, T20 is so dynamic it is almost impossible to accurately set scores and chases, and none of us want to see a team eliminated on the basis of a 6 over innings nor teams score more runs and lose. Avoiding these scenarios is not going to be easy, but progress must be made before the next World T20 in 2020. The hopes of a nation shouldn’t rest on the outdated calculations of two old English statisticians.