Duckworth-Lewis-Stern (DLS) Method - Simplified

Background

Cricket is a lengthy sport. It's shortest international format, T20Is, last for up to three hours per game. One Day International matches (the more conventional format) last for up to eight hours. Test matches (the original format) can last for up to five days! 

It is also a 'fragile' sport i.e it takes much less adverse conditions to stop a cricket match than it would a football match, for example. One example; Cricket cannot be played in the rain, not even a drizzle, if it's constant. This means when rainfall becomes constant, umpires have to call players off the field and have the grounds covered.

The challenge then becomes that it's a lengthy sport that is susceptible to weather stoppages. This means that there is a high probability of having matches abandoned due to inadequate playing time, especially during the rainy season. 

To counteract this problem, there was need to develop a system of compromise that could be used to restructure the game whilst accommodating the time lost to weather stoppages; In came Duckworth-Lewis-Stern (DLS) method.

Old Methods Used

DLS method wasn't an instant solution. It was designed after a few other methods had been implemented before. Each of the older methods used had it's own flaws, and so DLS method was built off the foundations of the older methods with the hope of removing score prediction flaws.

1. Average Run Rate Method

The average run rate method was fairly simple. It used the team that batted first's run rate to determine the second batting team's target score.

 

Run Rate: Average amount of runs scored per over. i.e Run Rate (RR) = Runs scored/ Overs bowled

 

The average run rate method would therefore use the formula:

 

Team B Target Score = (Team A Average Run Rate x Overs left for Team B) + 1

 

So, for team B it was just a matter of matching team A's run rate plus a single run in the overs they had available.

But then this was one of the method's flaws.

            (i) This gave an unfair advantage to the team batting second because it's easier to maintain a required run rate over fewer overs as there is less need to preserve your wickets as compared to a 50 over innings.

 

For example;

If Team A reached a score of 300 runs after 50 overs, their Run Rate would have been 6 runs per over, which is an excellent score in ODI cricket. If Team B then lost 30 overs due to bad weather and only had 20 overs available, their target over 20 overs would be:

Target = (6 x 20) + 1 = 121 Runs

 

A score of 121 runs in 20 overs is a very low target in T20 cricket.

Therefore the average run rate method would diminish the effort made by the team that batted first. 

            (ii) The second and most obvious flaw was that the average run rate method did not take into account the number of wickets lost. There was no accurate reflection of the balance of the game at the stop of play.

 

Scenario:

Team A's first innings score after 50 overs amounts to 250 runs. Run rate = 5. Team B's target = 251.

Team B's run chase is going terribly, and after 20 overs they score 120 runs for the loss of 9 wickets and have two tail-enders at the crease. Run rate = 6.

Suddenly torrential rain falls and the game is forced to end. Even though it's fairly clear that Team A were fully in control of this match, Team B would be awarded the victory for their superior run rate.

             

             (iii) The method could also be easily manipulated. 

As in the scenario above, when there are signs of incoming bad weather, teams batting second just had to push their run rate up without much care of the loss of wickets.

2. Most Productive Overs

The most productive overs method was designed as a counter measure to the flaws of the Average Run Rate Method. However, this method overcompensated the blemishes of the Average Run Rate Method and ended up giving an unfair advantage to the team batting first; Where the Average Run Rate Method gave an unfair advantage to the team batting second.

Most Productive Overs Method, as the name suggests, would use runs scored off the most productive overs of Team A to determine the target for Team B.

 

The Most Productive Overs method was calculated using the following formula:

 

Team B target (X overs) = Total of Team A's score in their most productive X overs + 1

 

Again, this had several flaws:

            (i) Firstly, and obviously, it took no account whatsoever of the most effective overs of the bowling side. Team B could have bowled a significant number of maiden overs, yet the formula did not take those into consideration.

            (ii) Again, the method did not take into account the number of wickets lost.

How DLS Method Works

DLS method is basically a mathematical formula that uses available data to predict scores, hence calculate hypothetical scores. It was introduced in 1997 and adopted by the ICC in 1999.

It's first use was in 1997 in a game between Zimbabwe and England, where Zimbabwe went on to win by 7 runs with the use of DLS.

 

What makes the DLS method more dependable is that unlike the previous method, DLS method takes into account all "Resources" each team has available to them. That means the method takes into account the wickets in hand and overs left.

 

Furthermore, DLS method considers historical data. Some cricket venues have higher average game scores than others. DLS method takes into account scores achieved over numerous matches, and at different venues, to calculate par scores. 

Such numbers can help better predict scores in the event of stoppages. The more matches played and recorded, the better a predictive tool DLS becomes. 

If there is a delay in play after Team A has already completed it's full batting quota,

 

Team B par score = Team A score x Team B's resources

Remember, a team's available resources means the wickets in hand and overs the team has left to bat.

 

If the match is interrupted and Team A fails to complete its full batting quota,

 

Team B par score = Team A score x (Team B resources/ Team A resources)

 

If the par score is a non-integer, as in most cases, it is rounded up to the next integer. That becomes the chasing team's revised target. The figure is rounded down to determine the par score (draw target).

  

DLS not available? - This simply means either the resource values are not available or a computer with an up-to-date DLS software is not available.

For the best score prediction results, it is always best to an up-to-date software.