New format leads to rethink June 17, 2009

Twenty20 leads to Duckworth-Lewis review

Cricinfo staff
29

A report in the Guardian claims that the Duckworth-Lewis system used in rain-affected matches will be reviewed in the coming months to take into account Twenty20 matches.

The Duckworth-Lewis method was introduced in 1997 after several failed attempts to come up with a way to make rain-affected one-day games more competitive. Since then, although there have been minor changes to the way run chases are calculated, no major overhaul has been undertaken. But the different needs of Twenty20 cricket means the time has come for a rethink.

"People have suggested that we need to look very carefully and see whether in fact the numbers in our formula are totally appropriate for the Twenty20 game," Frank Duckworth, one of the co-inventors, told the newspaper. "We thought it was appropriate to wait until the end of this competition when we've got a lot more Twenty20 data on our database.

"If there are any changes these should be ready for the commencement of the southern hemisphere season on 1 October." He added that if there were alterations, they were unlikely to make a significant difference to the calculations.

An ICC spokesman said it was "happy" with the Duckworth-Lewis system although it was always receptive to alternatives put forward by member boards. Duckworth and Tony Lewis are contracted by the ICC to do updates every few years - the last one was in 2006 - to reflect the changing nature of the game.

Although it has it critics, not least because of its complexity which often leaves crowds, and even commentators and players, bemused, few dispute that it a much better system than any that came before it.

There are alternatives, and the Indian Cricket League used a local system known as VJD. This was dismissed by Duckworth, who accused the inventor of trying to "give people what they feel is a fair answer" and of "fiddling his figures to do it".

Comments have now been closed for this article

  • Kirkstar on June 18, 2009, 20:55 GMT

    I dont know what's wrong with these losers. The WI had 9 over to get half of England's 160 runs. There was no advantage under the circumstances. There were only 3 power play overs. If we were to consider any thing near fair. An evenly balanced system would have asked 160 in 10 overs. Thus from the start WI were the ones under pressure. It is beyond absurd that to even suggest that the WI should also do it with no more than 5 wickets (another half). I concur with the sentiments that when ECB and CA are on the winning side they defend the Duckworth method. India is reknown for its software industry. Maybe the best work around is to design a software that applies a 'sympathy calculation' when ECB or CA are the input variables, and default to a drawn match when the play together

  • cornplus on June 18, 2009, 14:23 GMT

    If T20 was designed to get cricket popular where it isnt (North America & Europe) then the ECB & CA should stop changing/asking for review of the rules/regulations when decisions do not go in their favour. Let's face the fact the chasing 160 over the course of 20 overs requires a totally different approach than chasing 80 in 9 overs. I believe that the D/L target was reasonably fair althought it favoured the chasing team but England need to accept that we have no control over the elements and it could have affected any team. Many of the wickets that England took were due to poor shot selection by the West Indies Batsmen trying to get to 80runs required in about 4-5 overs; my point is that the number of wickets west indies lost should not be an issue. Let me be bold enough to make my June 21st predictions: WI vs Pakistan

  • Nadzzz on June 18, 2009, 11:10 GMT

    When you win D/L is fair when you lose you complain. It's simple as that. Losers always looks for excuses especially underperformers like England & NZ. No big title ever, no big player in last 20 years.(look for figures & numbers, honestly).

  • BryanLara on June 18, 2009, 10:59 GMT

    If the match had gone England way,we would not be reviewing D/L method.Get with England and Austraila fans.Law and regulations are laws and regulation.They are meant to be ammended but not to the ECB and CA liking.

  • aftab38 on June 18, 2009, 10:10 GMT

    I am not even sure that the revised target was correctly calculated. When I type 161 over 20 overs into the D/L calculator on their website, the target required to win after 9 overs is given as 84. Could the ICC/match referee perhaps have messed up? not for the first time ....

  • carthorse on June 18, 2009, 9:13 GMT

    Come on. Duckworth Lewis is fine save the wicket imbalance- Obvious and indisputable. The number of wickets, irrespective of how many overs, should be adjusted in a format relevant to runs required compared to total of team batting first. For example : Windies needed less than half England's runs so should have had four or five wickets only. How can half the runs be chased with the same number or wickets? Currently every team batting first is under a huge disadvantage as results show.

  • nafzak on June 18, 2009, 5:55 GMT

    Usingthe DL method, Eng had to score an easy 135 to win in 20 in a 50 over ODI earlier this year in the WI. No one complained about Duckworth Lewis then. Why now? I am tired of the disrespct that is shown to us in the non-white countries. I am sorry, but that is true. Look at the NZ capt. recently going to the umpire and asking to check on Umar Gul -all because NZ could not play spin. It's the same BS.. change the rules if you can't beat them. Call them for chucking. Pietersen switch hits and it's okay, but the bowler cannot switch from over to around. Isn't that unfair? The rules on bouncers were changed because the WI quartet were destroying everybody back in the late70's, 80's and so on. Fact is, WI would have approached the full target over 20 overs differently. Everyone knows that but the Eng just can't accept it. Look we the WI beat you all in the WI for the Stanford Millions. Just accept taht weare better than you. Don't let that fake series fool you all.

  • pragmatist on June 18, 2009, 3:18 GMT

    Duckworth Lewis has been pretty good over the years but it's shocking that it has patently not yet been updated to accommodate T20. My other issue with it is that it just isn't transparent. Why doesn't Cricinfo show the par score on its live scorecards, and offer the calculation table to cricket fans around the world? Maybe then everyone might trust it a bit more.

  • emmwill on June 18, 2009, 2:31 GMT

    When will England fans wake up and smell the cricket coffee? England is not really a strong Twenty20 team. The fact that England just defeated West Indies in their own backyard in tests and ODIs does not logically mean that they would have defeated them in the semi-final if the game had run for the full twenty overs. Please remember that it was the same WI team that defeated England at home earlier this year!! Remember as well that an associate team, the Netherlands, defeated England in the earlier part of the competition. DL or not, England would have lost because of bad decision making on the day. Don't blame DL this time. The reality is that England did not bat that well in the middle overs. There's where they lost the match. Accept it and move on.

  • missone on June 17, 2009, 23:11 GMT

    i agree pycon and i agree with you especially stiab and all of you have several valid points but i said it before and i'm going to say it again i think many of you here are using this issue over D/L trying to downplay the Windies' victory. yes the system does need reviewing but i think 80 off 9 overs was reasonable and defendable as England started to show. 160 ish on the other hand is not defendable at the oval as we have seen before in previous games

  • Kirkstar on June 18, 2009, 20:55 GMT

    I dont know what's wrong with these losers. The WI had 9 over to get half of England's 160 runs. There was no advantage under the circumstances. There were only 3 power play overs. If we were to consider any thing near fair. An evenly balanced system would have asked 160 in 10 overs. Thus from the start WI were the ones under pressure. It is beyond absurd that to even suggest that the WI should also do it with no more than 5 wickets (another half). I concur with the sentiments that when ECB and CA are on the winning side they defend the Duckworth method. India is reknown for its software industry. Maybe the best work around is to design a software that applies a 'sympathy calculation' when ECB or CA are the input variables, and default to a drawn match when the play together

  • cornplus on June 18, 2009, 14:23 GMT

    If T20 was designed to get cricket popular where it isnt (North America & Europe) then the ECB & CA should stop changing/asking for review of the rules/regulations when decisions do not go in their favour. Let's face the fact the chasing 160 over the course of 20 overs requires a totally different approach than chasing 80 in 9 overs. I believe that the D/L target was reasonably fair althought it favoured the chasing team but England need to accept that we have no control over the elements and it could have affected any team. Many of the wickets that England took were due to poor shot selection by the West Indies Batsmen trying to get to 80runs required in about 4-5 overs; my point is that the number of wickets west indies lost should not be an issue. Let me be bold enough to make my June 21st predictions: WI vs Pakistan

  • Nadzzz on June 18, 2009, 11:10 GMT

    When you win D/L is fair when you lose you complain. It's simple as that. Losers always looks for excuses especially underperformers like England & NZ. No big title ever, no big player in last 20 years.(look for figures & numbers, honestly).

  • BryanLara on June 18, 2009, 10:59 GMT

    If the match had gone England way,we would not be reviewing D/L method.Get with England and Austraila fans.Law and regulations are laws and regulation.They are meant to be ammended but not to the ECB and CA liking.

  • aftab38 on June 18, 2009, 10:10 GMT

    I am not even sure that the revised target was correctly calculated. When I type 161 over 20 overs into the D/L calculator on their website, the target required to win after 9 overs is given as 84. Could the ICC/match referee perhaps have messed up? not for the first time ....

  • carthorse on June 18, 2009, 9:13 GMT

    Come on. Duckworth Lewis is fine save the wicket imbalance- Obvious and indisputable. The number of wickets, irrespective of how many overs, should be adjusted in a format relevant to runs required compared to total of team batting first. For example : Windies needed less than half England's runs so should have had four or five wickets only. How can half the runs be chased with the same number or wickets? Currently every team batting first is under a huge disadvantage as results show.

  • nafzak on June 18, 2009, 5:55 GMT

    Usingthe DL method, Eng had to score an easy 135 to win in 20 in a 50 over ODI earlier this year in the WI. No one complained about Duckworth Lewis then. Why now? I am tired of the disrespct that is shown to us in the non-white countries. I am sorry, but that is true. Look at the NZ capt. recently going to the umpire and asking to check on Umar Gul -all because NZ could not play spin. It's the same BS.. change the rules if you can't beat them. Call them for chucking. Pietersen switch hits and it's okay, but the bowler cannot switch from over to around. Isn't that unfair? The rules on bouncers were changed because the WI quartet were destroying everybody back in the late70's, 80's and so on. Fact is, WI would have approached the full target over 20 overs differently. Everyone knows that but the Eng just can't accept it. Look we the WI beat you all in the WI for the Stanford Millions. Just accept taht weare better than you. Don't let that fake series fool you all.

  • pragmatist on June 18, 2009, 3:18 GMT

    Duckworth Lewis has been pretty good over the years but it's shocking that it has patently not yet been updated to accommodate T20. My other issue with it is that it just isn't transparent. Why doesn't Cricinfo show the par score on its live scorecards, and offer the calculation table to cricket fans around the world? Maybe then everyone might trust it a bit more.

  • emmwill on June 18, 2009, 2:31 GMT

    When will England fans wake up and smell the cricket coffee? England is not really a strong Twenty20 team. The fact that England just defeated West Indies in their own backyard in tests and ODIs does not logically mean that they would have defeated them in the semi-final if the game had run for the full twenty overs. Please remember that it was the same WI team that defeated England at home earlier this year!! Remember as well that an associate team, the Netherlands, defeated England in the earlier part of the competition. DL or not, England would have lost because of bad decision making on the day. Don't blame DL this time. The reality is that England did not bat that well in the middle overs. There's where they lost the match. Accept it and move on.

  • missone on June 17, 2009, 23:11 GMT

    i agree pycon and i agree with you especially stiab and all of you have several valid points but i said it before and i'm going to say it again i think many of you here are using this issue over D/L trying to downplay the Windies' victory. yes the system does need reviewing but i think 80 off 9 overs was reasonable and defendable as England started to show. 160 ish on the other hand is not defendable at the oval as we have seen before in previous games

  • gudolerhum on June 17, 2009, 20:22 GMT

    The WI were very fortunate to have to chase just 80 runs with 10 wickets intact. It is absurd for u guys to say that they would have got 160, no one knows that. D/L should not reduce the number of overs each bowler may bowl. If the captain wants to bowl one bowler the full 4 overs that should be OK, the other bowlers make up the remaining overs. The reduction for each bowler severely restricts the fielding team and further loads the odds against them. WI are like the Pakistanis, very inconsistent and tempermental.

  • Saadi69 on June 17, 2009, 19:49 GMT

    Whether WI would have won or not chasing 160 in 20 is a different thing, but the target should have been fair, which in my opinion was not.

  • pycon45 on June 17, 2009, 18:57 GMT

    If west INdies were chasing 161 in 20 overs they would have won.Its because they were chasing 81 in 9 overs that there were 45 for 5.AS mush as you guys have great ponits its because it was West Indies vs England thats some of you want a new system.And how can you guys say West Indies wouldn't win its 20/20 anything can happen after netherlands show that.HAHA salop England.

  • shellgrip on June 17, 2009, 18:04 GMT

    I appreciate that the runs required figure *should* take into account the number of wickets available, but as you say geedubnz, in the England/WI match the total didn't seem to accurately reflect this. The RR difference was actually slightly less than 1 per over (0.84), which is effectively nothing and easily attainable.

    No one is crying over spilt milk here but 161 is a reasonable total - if not as good as England would have liked - and given England's recent performance against the Windies, most followers of the game would have expected an England win from that position. The D/L adjusted game had a 'balance' in entirely the other direction. As soon as it was announced that WI would chase 80 over 9 overs, all of us watching agreed that they'd win. That in itself is probably good evidence that D/L didn't work in this case.

    Jon

  • tjpc25 on June 17, 2009, 14:00 GMT

    West Indies won the game fair and square...BUT if people are honest, surely chasing 162 in 20 overs @ 8.1rpo is harder than chasing 80 in 9 overs @ 8.89rpo. Thats a difference of only .8rpo and the advantage you get from only having to bat 9 overs is massive, as shown by 1) amount of wickets Windies lost but it didnt make a difference to way they batted and also, to an extent the Scotland v NZ match where there were runs galore.

    Also in 50 over cricket a run rate of about 5.5-6rpo is par but when you half the overs (roughly-40%) to 20 over cricket a par score is about 8rpo. This is an increase of 33%. if you applied the same logic from 20 over cricket to 9 over cricket (again about half the overs-45%) then a par score in 9 over cricket would be about 10.5 giving a score of 94 ish.

    I am not saying what I have put above is definitive or anything of the sort just saying surely someone needs to reconsider as the advantage gained in shortened matches is massive.

  • stiab3 on June 17, 2009, 13:24 GMT

    Interesting as I'm pretty confident the WIndies would've done better chasing the 162 over 20 overs. They've consistently scored more runs than 162 . . .

    Scoring at 9 and over for 9 overs puts more pressure on the batsmen as they feel they have to whack every ball tostay with the run rate. This is why the WI top order collapesed - panis. With only 9 overs ever dot-ball is more significant. You can't count on a good over to compensate for a bad one, as there aren't many overs to play with.

    Very often in a 20-over match, the team will score at a high rate at the beginning, it slows in the middle and picks up again at the end. With only 9 overs you have to come out slogging and maintain the rate throughout.

  • lanka_86 on June 17, 2009, 12:28 GMT

    Great suggestion by D.V.C! It's complicated, but it's much fairer than the NRR system ... CricketPissa - but the 'moving target' (which I haven't heard before) would then need to be applied to non D/L matches as well (for consistency). Also target increasing = current total decreasing. It's like indoor you'll need to bat out the overs. Some might like it.

  • Mr_Chablis on June 17, 2009, 12:27 GMT

    I agree with Geoff - over a shorter perod, the difference needs to be greater. In the Eng v WI game, a target of somewhere between 90 and 100 would probably have been fairer (England were 75-2 after 9 overs in their innings). I would disagree with moving targets or a restriction on the number of wickets for the side batting second. The target is x, how you go about it is up to you. As more stats become available, then more accurate calculations can be made, but it can never be perfect - one is trying to to make the best out of unusual conditions, and from time to time anomalies will occur.

  • mandi on June 17, 2009, 12:02 GMT

    Indian vjd system is good for international cricket. Icc always reject it this is there fault

  • missone on June 17, 2009, 12:01 GMT

    in my opinion u guys are just using this whole D/L thing to make the world feel that England was hard done by. Truthfully WI would have won as they would not have thrown their bats at everything. 160 is a below par total at THe Oval and people should just start accepting the fact WI are a better 20/20 side than England and were going to win if it were a a full game. I think England shoud'nt have played 5 bowlers especially 3 fast bowlers. They should have seen that pace on the ball makes run scoring easier at the oval and Dini Mascarenhas would have been a better option in the bowling department and strenthen the already weak batting line up. Persons always criticize WI for sepending on our big four but England is heavily relient on KP and no one says anything. True i have my issues with D/L but stop using it as an excuse to downplay the Windies victory.

  • Air_Wolf on June 17, 2009, 11:42 GMT

    I think the much needed alteration should be the loss of wickets. Like in case of WI vs Eng game. WI should make those 81 runs within 6 or 7 wickets (same as 3 batsmen are allowed in super over). I think this can eradicate such complications.

  • McPiggle on June 17, 2009, 11:02 GMT

    The sooner we all accept that England were no match for the Windies the better for us all. If it had been a full compliment of 20 overs, they still would have won!

    No use crying over spilt milk!

  • geedubnz on June 17, 2009, 11:00 GMT

    Jon -

    The way the D/L method takes account of wickets available in shortened over matches is to set a target higher than the average run rate the team batting first achieved. Ie, England made 160 off 20 overs at 8 per over, the Windies were set 80 off 9 overs at almost 9 per over. So your gripe should actually be that the target they were set was too low, which perhaps it was as one extra run per over doesn't seem a lot over only 10 overs. I'm not sure reducing the number of wickets available is an entirely fair approach to take, because it should be up to the batting team to decide on their tactics to get the score (to go hard from the beginning and risk losing wickets, or to establish a base and then launch with wickets in hand), so I would be more in favour of some sort of exponential factor that increasese the target by a greater amount the shorter the match becomes.

    Geoff

  • MZ00 on June 17, 2009, 10:24 GMT

    I do agree with the number of wickets left when D/L applies. Everything looks fine exept the wickets left. Number of wickets also should be reduced in the same manner number of overs and required runs are reduced. It will bring a balance in L/D method.

  • CricketPissek on June 17, 2009, 10:06 GMT

    i wonder if they'll bring back the 'moving target' which was in the original concept (i remember Sri Lanka losing to SA in such a match in the 'champions trophy') with every wicket that fell, the target became higher! :-o

  • Rabbo on June 17, 2009, 10:02 GMT

    I believe the DL system to be an excellent way of resolving rain affected fixture. Even taking into account the par score increases if the batting side loses wickets could anything be added to reduce the number of wickets the batting side can lose in the reduced innings. When the WI batted to still have 10 wickets gives them a tremendous advantageous to be able to take a much higher risk to throw the bat, where you would normally expect that to chase 80ish off 9 in a normal game may well have seen the loss of several wickets. Messrd Duckworth & Lewis - over to you.

  • lanka_86 on June 17, 2009, 9:54 GMT

    Hopefully they don't need to introduce any new parameters, and can get away with just tweaking a few existing parameters (which I believe they already do from time to time in the current system due to ODIs run-rate changes)

  • D.V.C. on June 17, 2009, 9:48 GMT

    The D/L method might need a little tweaking, but while they are undertaking this review they should look at using the method as a fairer alternative to NRR.

    Consider a team batting second, chasing say 120 for victory, that makes its target with 8 wickets down. Their victory margin is a slim 2 wickets. Yet, if they completed that victory in only 10 overs, then their NRR benefit from the game is huge. Conversely, their oppoonents who came very close to winning the game are penalised for having lost in only 10 overs.

    I would suggest that instead, the D/L target for 10 overs, 8 wickets down be looked up once the (example) target is made and the victory margin converted to a number of runs. In the example I have given the winning margin would be 11 runs. Whereas if the team batting second had reached its target at 10 overs but with only 2 wickets down the winning margin would be 66 runs. This seems a much fairer assesment of the relative success of the two teams in the game.

  • shellgrip on June 17, 2009, 9:41 GMT

    Surely one of the key considerations in T20 is the number of wickets available when the number of overs becomes very small. For example, the target of 80 runs set for the Windies against England seemed reasonable as a figure, but not when they had all 10 wickets available to get that score. There's also the issue of maintaining a high run rate: Asking for 9 an over for 20 overs is much harder than 9 an over for 9. Even with England knocking the wickets over at a good rate it was clear the total was always going to be attainable whereas I'm pretty confident England would have won if the full 20 overs and a target of 162 had been played.

    When T20 in it's FULL form is only 20 overs, I'm not even certain it's possible to get a fair and balanced restricted overs calculation.

    Jon

  • No featured comments at the moment.

  • shellgrip on June 17, 2009, 9:41 GMT

    Surely one of the key considerations in T20 is the number of wickets available when the number of overs becomes very small. For example, the target of 80 runs set for the Windies against England seemed reasonable as a figure, but not when they had all 10 wickets available to get that score. There's also the issue of maintaining a high run rate: Asking for 9 an over for 20 overs is much harder than 9 an over for 9. Even with England knocking the wickets over at a good rate it was clear the total was always going to be attainable whereas I'm pretty confident England would have won if the full 20 overs and a target of 162 had been played.

    When T20 in it's FULL form is only 20 overs, I'm not even certain it's possible to get a fair and balanced restricted overs calculation.

    Jon

  • D.V.C. on June 17, 2009, 9:48 GMT

    The D/L method might need a little tweaking, but while they are undertaking this review they should look at using the method as a fairer alternative to NRR.

    Consider a team batting second, chasing say 120 for victory, that makes its target with 8 wickets down. Their victory margin is a slim 2 wickets. Yet, if they completed that victory in only 10 overs, then their NRR benefit from the game is huge. Conversely, their oppoonents who came very close to winning the game are penalised for having lost in only 10 overs.

    I would suggest that instead, the D/L target for 10 overs, 8 wickets down be looked up once the (example) target is made and the victory margin converted to a number of runs. In the example I have given the winning margin would be 11 runs. Whereas if the team batting second had reached its target at 10 overs but with only 2 wickets down the winning margin would be 66 runs. This seems a much fairer assesment of the relative success of the two teams in the game.

  • lanka_86 on June 17, 2009, 9:54 GMT

    Hopefully they don't need to introduce any new parameters, and can get away with just tweaking a few existing parameters (which I believe they already do from time to time in the current system due to ODIs run-rate changes)

  • Rabbo on June 17, 2009, 10:02 GMT

    I believe the DL system to be an excellent way of resolving rain affected fixture. Even taking into account the par score increases if the batting side loses wickets could anything be added to reduce the number of wickets the batting side can lose in the reduced innings. When the WI batted to still have 10 wickets gives them a tremendous advantageous to be able to take a much higher risk to throw the bat, where you would normally expect that to chase 80ish off 9 in a normal game may well have seen the loss of several wickets. Messrd Duckworth & Lewis - over to you.

  • CricketPissek on June 17, 2009, 10:06 GMT

    i wonder if they'll bring back the 'moving target' which was in the original concept (i remember Sri Lanka losing to SA in such a match in the 'champions trophy') with every wicket that fell, the target became higher! :-o

  • MZ00 on June 17, 2009, 10:24 GMT

    I do agree with the number of wickets left when D/L applies. Everything looks fine exept the wickets left. Number of wickets also should be reduced in the same manner number of overs and required runs are reduced. It will bring a balance in L/D method.

  • geedubnz on June 17, 2009, 11:00 GMT

    Jon -

    The way the D/L method takes account of wickets available in shortened over matches is to set a target higher than the average run rate the team batting first achieved. Ie, England made 160 off 20 overs at 8 per over, the Windies were set 80 off 9 overs at almost 9 per over. So your gripe should actually be that the target they were set was too low, which perhaps it was as one extra run per over doesn't seem a lot over only 10 overs. I'm not sure reducing the number of wickets available is an entirely fair approach to take, because it should be up to the batting team to decide on their tactics to get the score (to go hard from the beginning and risk losing wickets, or to establish a base and then launch with wickets in hand), so I would be more in favour of some sort of exponential factor that increasese the target by a greater amount the shorter the match becomes.

    Geoff

  • McPiggle on June 17, 2009, 11:02 GMT

    The sooner we all accept that England were no match for the Windies the better for us all. If it had been a full compliment of 20 overs, they still would have won!

    No use crying over spilt milk!

  • Air_Wolf on June 17, 2009, 11:42 GMT

    I think the much needed alteration should be the loss of wickets. Like in case of WI vs Eng game. WI should make those 81 runs within 6 or 7 wickets (same as 3 batsmen are allowed in super over). I think this can eradicate such complications.

  • missone on June 17, 2009, 12:01 GMT

    in my opinion u guys are just using this whole D/L thing to make the world feel that England was hard done by. Truthfully WI would have won as they would not have thrown their bats at everything. 160 is a below par total at THe Oval and people should just start accepting the fact WI are a better 20/20 side than England and were going to win if it were a a full game. I think England shoud'nt have played 5 bowlers especially 3 fast bowlers. They should have seen that pace on the ball makes run scoring easier at the oval and Dini Mascarenhas would have been a better option in the bowling department and strenthen the already weak batting line up. Persons always criticize WI for sepending on our big four but England is heavily relient on KP and no one says anything. True i have my issues with D/L but stop using it as an excuse to downplay the Windies victory.