Stats Analysis

Is the DRS a good thing? Yes if you're a fingerspinner, no if you're a left-hand batsman

How have batsmen and bowlers gained and lost under the various eras of the lbw law?

Kartikeya Date
12-Jun-2020
Advantage DRS: David Warner's loss is Shakib Al Hasan's gain  •  Getty Images

Advantage DRS: David Warner's loss is Shakib Al Hasan's gain  •  Getty Images

In the dozen or so years since the DRS first came into cricket, there has been a persistent view that it has resulted in more lbw decisions being given against batsmen. Karthik Krishnaswamy provided an excellent overview about this issue for the DRS era in these pages. In this essay, I present a longer-term picture over the history of Test cricket.
I looked at the dismissal summary (for example, see the record for Jason Gillespie here) for every bowler in Test cricket in four separate eras, organised according to changes in the lbw law. Since the advent of Test cricket, that law has been officially changed in 1937 and 1980. The DRS was introduced in 2008. So the four eras considered here are 1877-1936, 1937-1980, 1981-2008 and 2009-2020.
The lbw law has become progressively friendlier to the bowler on the questions of pitching and impact with each amendment. Until 1937, a batsman could be out lbw only if the ball pitched within the width of the stumps and hit the batsman within the width of the stumps. In 1937, the law was amended to allow lbws to be given even if the ball pitched outside off stump, provided the impact was within the width of the stumps. In 1980 this was further amended to allow lbws to be given even if both the impact and pitching were outside off stump, provided the batsman was not offering a shot. Since the advent of the DRS in 2008, there has been much speculation (see Gautam Gambhir's recent comments for an example) that many lbws that are given under the review regime would possibly have not been given in the era before it. The evidence for this proposition is thin, but it is worth considering how the changes to the lbw law or its adjudication regime sit within larger changes in the game.
A basic way to compare the prevalence of lbw dismissals is to compare it to the prevalence of all other forms of dismissal. Under the highly restrictive pre-1937 lbw law, there were nearly ten non-lbw dismissals for each lbw dismissal. After the 1980 amendments, this figure was halved. Spinners and seamers fared differently under each regime. The 1937 amendments made it significantly easier for seamers to get lbws but not for spinners. Similarly, the advent of the DRS has improved matters for spinners but not for seamers. Compared to the effect of the 1937 and 1980 amendments, which made lbw dismissals significantly more frequent, the DRS has had no overall effect on the frequency of lbw dismissals in Test cricket. The improvement for spinners is balanced by a decrease for fast bowlers.




If we look within spin, fingerspinners have benefited significantly more than wristspinners from the DRS, and they have benefited more in the DRS era than they did from the rule changes in 1937 and 1980.


All this suggests that though the overall claims about the effect of the DRS on lbws are not sustainable, claims such as the one Erapalli Prasanna made, when he said he would have taken many more wickets if he had bowled in the age of the DRS, might well be.
The table below shows how seamers and spinners overall, and the two types of spinners, have fared under each lbw regime. Historically, fast bowlers have been more attacking than spinners, and wristspinners more attacking than fingerspinners. This can be seen especially in the record between 1937 and 2008, when the effects of uncovered wickets were mitigated to a large degree because the rule for covering wickets was changed (unevenly across the world) starting in the 1950s. Fast bowlers conceded quicker runs and took wickets more frequently compared to spinners. In the DRS era, spin and pace have conceded runs at about the same rate (3.1 runs per over). This gap between fingerspin and pace has nearly disappeared (2.6 for fingerspin, 3.0 for pace in 1981-2008; 3.0 for finger spin, 3.1 for pace in the DRS era). This is part of a global trend in scoring in Test cricket that points to a shift in the batsmen's approach. While the average fingerspinner's wicket costs two runs more today than it did under the pre-1980 lbw regime, and two runs fewer than it did in the 1981-2008 lbw regime, it arrives once every 68 balls in the DRS era compared to once every 88 balls in the pre-1980 era.
Run-scoring has not become more difficult over the years (the global batting average has not fallen) though the lbw law has been made progressively more generous to bowlers. The lbw rates suggest that contemporary batsmen are more prepared to attack the ball when it is on the stumps compared to their predecessors. But as the summary below for all types of bowlers shows, overall the increase in lbw rates has come at the cost of bowled dismissals.
Why has the increase in lbws coincided with the reduction in bowled dismissals? What does this suggest about the way in which batsmanship has changed? Each change in the lbw law has expanded the circumstances in which a batsman can be lbw. The increased prevalence of lbw has come at the cost of bowled dismissals. This suggests that by the narrowing the possibilities of pad play, the expanded lbw law has encouraged batsmen to attack more both when the ball is on the stumps and when it is not. The dead bat, we might say, has retreated in Test cricket. This has brought quicker runs and quicker dismissals.
Turning our attention to spinners, another interesting question arises. Are lbws to fingerspinners more frequent in the DRS era because batsmen have become more attacking (for example, are prepared to sweep from the stumps more often than their predecessors), or have the batsmen become more attacking because they think that the DRS takes away their protections if they use their pads as their predecessors did?
One area where the DRS seems to have had a significant effect is with lbws against left-hand batsmen. Lefties have traditionally been something of a novelty in Test cricket. Between 1877 and 1936, 14% of all Test innings were played by left-hand batsmen. This increased to 18% between 1937 and 1980, and 25% between 1981 and 2008. In the DRS era, one in three Test innings is played by a leftie.
The evidence in the table above shows that in all previous eras, lefties were systematically less likely to be dismissed lbw against spin compared to righties. Perhaps this disparity has shrunk because left-handers have become less of a novelty and bowlers have worked out better ways to bowl to them, and because umpires have gained more experience in adjudicating appeals against them. Overall, for all bowlers, the DRS era is the first one in which lefties and righties are out lbw equally often against seam and spin.
Ian Chappell recently proposed in these pages that the lbw law should be altered to eliminate any specific requirements for pitching or impact. His proposed law states: "Any delivery that strikes the pad without first hitting the bat and, in the umpire's opinion, would go on to hit the stumps is out regardless of whether or not a shot is attempted." This would have the interesting consequence of removing any protection the batsman might have on the "blind" (leg) side, and stop cricket from being a side-on game. Further, it would open up a whole new line of attack for bowlers and make spin bowlers even more potent than they are today. As Chappell points out, it would eliminate one avenue of stalemate between bat and ball.
The record shows that the game has been becoming more attacking. Batsmen are scoring quicker and bowlers getting wickets more frequently. Chappell's modification will probably extend this even further. It will probably also reduce overall scores. Its greatest benefit will probably not accrue to wristspinners, but to left-arm seamers bowling to right-handers and right-arm seamers bowling to left-handers. The batsman will no longer be able to cover up and allow the ball to hit the pad if it pitches marginally outside leg stump. This will have consequences for the batsman's ability to cover off stump and the outside edge too.
It should be noted, though, that when the DRS was introduced, there were some early concerns that too many lbws would be given and consequently batting line-ups would be dismissed cheaply too often. These fears have not been realised, in part through the codification of umpire's call. On balance, it may be interesting to try out Chappell's lbw proposals in T20 or the Hundred before introducing it into four-innings cricket.
The major effect of the DRS regime on the implementation of the lbw law has been to benefit fingerspinners and hurt left-hand batsmen. In the era immediately prior to the DRS, fingerspinners performed mostly as a restrictive foil to the more attacking seam and wristspin bowlers. Under the DRS, this restrictive role has largely disappeared from Test cricket. Occasionally, a great bowler like R Ashwin has produced a masterful restrictive performance like his 3 for 92 in 52.5 overs in the fourth innings in Adelaide in 2018, which helped India defend 323 on a very good fourth-innings pitch. But as a norm, fingerspinners have become attacking bowlers, like seamers and wristspinners are.
All this suggests that while the introduction of the DRS has benefited fingerspinners and hurt left-hand batsmen, the contribution of the DRS to these shifts exists within a larger generational shift in Test cricket in which batsmen are scoring quicker and getting dismissed quicker. The DRS has dismantled some prevalent umpiring conventions when it comes to evaluating lbw appeals, but compared to the effect of changes to the actual lbw law, the effect of the DRS on the prevalence of lbws in Test cricket has been marginal. To change the game, its laws have to be changed. Changing umpiring has little overall effect.

Kartikeya Date writes at A Cricketing View. @cricketingview