From whacking it hard to whacking it smart: how new-ball swing has revolutionized the ODI Powerplay
Advanced data points tell the story of the two trends that will define the 2023 ODI World Cup: new-ball swing, and increased batting intent
The cricket ball is made of two hemispheres. Divided by the seam, this characteristic lends itself easily to the creation of uneven airflow around the two halves to generate what is known as “conventional swing”.
The technical explanation is as follows. When a fluid undergoes motion in a system, the motion may be what is called “laminar flow” or “turbulent flow”. Laminar flow is nice, smooth and streamlined - every constituent particle of the fluid moves in one direction, along the same path, without mixing or shifting layers. Turbulent flow, on the other hand, is chaotic and irregular. With its swirling mass of eddies, currents and gyrations, a fluid in turbulent flow exerts a great amount of pressure on bodies it is in contact with. Laminar flow, on the other hand, being the affable host that it is, does not.
And so, bowlers shine one half of the ball, lusting after the geniality of laminar flow, and roughen up the other half of the ball, thirsty for the wrath of turbulent flow - in the hope that the excess pressure on one hemisphere causes the ball to curve laterally through the air. This gives rise to what is called conventional swing.
Take Mitchell Starc, who now stands at the top of his bowling mark, for example.
Starc is now thirty-three years old, that magical age at which the world finally takes notice that you will soon be reduced to a line in a record book, so they might as well celebrate you already. Is Starc one of the greatest? Is he the best? They deliberate.
Starc’s hair is neither as thick and luscious as it was eight years ago when he seized World Cup glory, nor as short yet sleek as four years back when he became the highest wicket-taker in a one-day World Cup. Bags are beginning to form around his face, and a summer in India has rendered his skin the colour of a ripe apple. What a bowler he was in his prime, they reminisce.
But working with the shiny new ball is what Starc does. More precisely, swinging it into the batter’s stumps from a height.
And so he rounds at his mark, rubbing his forefinger on the leather one last time, and sets off in his run-up. Chest thumping, he gets into his stride, as the jaw clenches and the front leg braces. The forefinger standing erect, his non-bowling arm is wrenched backwards. The eyes shut, in fatigue or effort nobody knows, and a bellow is let out. Out comes the ball from his hands, swinging viciously into the right-hander’s pads, doing exactly what fast bowlers across the world have been doing for the past two years.
Remember the seam that divided the ball into two halves? It’s hard and new, by the way.
*****
Our story starts at---well, we don’t quite know where. The closest there exists to evidence of Kookaburra’s altered manufacturing process is the astute Jarrod Kimber’s documentation of the new Kookaburra balls and the missing 38 runs. These new balls may or may not have been trialled out in Adelaide in 2020 during India’s 36-all-out Test, though we cannot say for sure. The closest there is to an official source comes from Snehal Pradhan’s 2018 The Cricket Monthly article: “For their part, Kookaburra plan to develop a new, more cost-effective white ball for T20Is, with a ‘more bulky seam that gives the bowlers something to work with.’”
This is the conjecture. The fact is that the two years leading into the 2023 ODI World Cup have seen the lowest batting average and run rates of the past ten years in the bowling Powerplay.
And yet, as Jarrod speculates in his video, the new ball has not swung more in ODI cricket in the last two years. But what it has done is swing and seam more in the top ten overs in some highly specific terrains - India and England (and possibly, Australia too, for which we do not have the data).
So what has driven this change? The answer to this question is as clear-cut as any info about Kookaburra’s ball change itself. But we can make a few guesses. Maybe the reinforced Kookaburra has come into effect only in the Big Three nations, where a higher percentage of matches are played and more balls are in use. Maybe the rest of the world, having witnessed a sudden drop in the number of matches they play since Covid, hasn’t got around to using the new batches of Kookaburra just yet.
It is on this admittedly shaky ground that the founding planks of our story rest. But little is shaky about the fact that the ball is swinging more in England and India. As the 2023 ODI World Cup looms, the best way to reacquaint ourselves with the one-day international, this forgotten child of the last three years, is to study its new avatar in shape and form. Think of it as a father’s pursuit to fathom his eldest teenage son, whom the age of smartphones and porn has swept away from his cheery eight-year-old self, while the father himself was too busy tending to his newborn darling daughter. And to do this, he decides to study a set of similarly-minded teenage individuals – just as we shall do when we combine England and India to say something about the top ten overs of the ODI in India.
First things first. The ball is doing more in the air and off the pitch. And what’s more, this effect is driven chiefly by the top five overs of the Powerplay rather than the bottom five. When each new ball is between two and three overs old, it’s a wholly different gravy.
And batters have been responding to this. Before 2021, against balls which swung away from the right-handed batter by approximately 1.5 degrees or more1 in the top five overs, the second- and third-most common shots were the defence and the push, respectively. Since then, batters have been shouldering arms and defending instead. The incidence of the defence, similarly, has gone up against balls that seam in, while the one that seams away is being driven less.
A somewhat analogous effect is seen in overs 5-10. Balls which swing in are flicked and glanced fine more - an interesting result we will explore in further detail soon - while those that go away are being left alone. While the effect of the outswinger stays constant across the top and bottom halves of the Powerplay, the inswinger’s effect is more volatile. Since the ball swerves in and therefore threatens the stumps, it fetches more respect from the tentative batter; since it angles into his pads, the set batter despatches it.
*****
Mitchell Starc is in his stride, jaw clenched and chest thumping. The forefinger standing erect, his non-bowling arm is wrenched skyward, as a bellow is let out. Out comes the ball from his hands, swinging viciously into the right-hander’s pads, doing exactly what fast bowlers across the world have been doing for two years.
Suryakumar Yadav, batting on 0, hoists his bat towards the point fielder, notwithstanding the tell that the ball is pitched on his off-stump. SKY averages a measly 11 in his last ten one-day games, and only once in these ten innings, of the times when he has gotten off the mark, has he been out with a strike-rate of less than 100. India are 32-2 in 4.5 overs, and the dressing room has issued for calm. But what do we know of an entertainer that the entertainer does not know of himself. SKY brandishes his bat in preparation for a lascivious on-drive, mouth agape in concentration - but before he can sense it, in an abrupt clatter of sounds, the ball has crashed into his pads, and the Australians have nested themselves in the umpires’ ears in appeal.
SKY is given out lbw on zero, trying to attack a swing bowler at the peak of his powers. Yet he is only the starkest manifestation of what is a global trend.
If we define an “attacking shot” as a drive, cut, flick, pull, sweep, scoop, or slog, and define a “T20 shot” as a slog or a scoop, the data reveals that the proportion of attacking shots played in one-day internationals in the first five overs has increased by a statistically significant two percentage points since 2021. More interestingly, the incidence of T20 shots has increased by 25%. The trend is marginally easier to see in the bottom five overs of the Powerplay: the occurrence of attacking shots has increased by almost three percentage points, while the T20 shots are being played 42% more often since 2021.
Remember, this is despite the fact that the ball is swinging and seaming more, and that overall run rates have never been lower since 2013.
So, what explains this increased attacking intent? The answer lies in history: between 2015 and 2019, England revolutionized white-ball batting, scything the ball from over number one, introducing the world to a batting strategy so openly superior to the traditional one. Like all innovation cycles, it takes time for this idea to catch on - England must even win a World Cup to prove its utility - but somewhere down the lane, the world was bound to recognize its usability. Now they have.
And that is not all. Batters are being clever in picking and choosing the kinds of balls they attack. The below chart shows the attacking intent of batters split by delivery type before and after 2021. It may be seen that the increased attacking intent of the 2020s is driven largely by a disdain of stock balls, which make up the bulk of the sample, and to a lesser extent, slower balls. In the rare instance that the ball is not swinging, batters across the world are being especially harsh at the delivery, going at a run rate of over six-an-over. The trendline is clear if you are a bowler: if the new ball is swinging, do not attempt a slower ball.
Together, it is these two trends that characterize the Powerplay dynamics of the ODI format in 2023: enhanced lateral movement courtesy of a hardened Kookaburra, and improved intent thanks to England’s innovation.
Its effects can be broadly categorized into three, as shall be presently seen.
*****
Of the three effects I discuss, two are driven by the rise in lateral movement, while the third is a child of aggressive batting.
1. Popularization of the Outside Off-Stump Line This is somewhat of a counterintuitive effect. Conventional wisdom suggests that when the ball swings, and is predicted to do so for a short period of time, bowlers might be better advised to target the stumps on a fullish length. Nathan Leamon, the English analyst, suggests this here and Kartikeya Date has stumbled upon evidence of this in his ball-tracking dataset.
For the uninitiated, the following are the numbers for balls which swing by 1.5 degrees or more in either direction (according to the same calculation referenced above) in one-day internationals and Tests since 2018. When the ball is outside off-stump, batters exhibit a control rate of almost 73%; whereas when it is pitched on the stumps it falls to 70%.
But empirically, something seems to be stopping bowlers from leveraging this advantage. In Tests and ODIs since 2018, when the ball swings by 1.5 degrees or more, bowlers pitch 81% of their deliveries outside the off stump as opposed to 76% when it swings less. And this is exactly what we see in the one-day game since 2021. In an era in which the ball has been swinging and seaming more than it has in the ten years before, pace bowlers are opting for the comparatively ungainly channel outside the off stump, when they might as well be reaping more rewards off the stumps. Before 2021, in the first five overs of the game, 33% of balls from pacers used to end up outside off stump; since 2021, it’s a marginally hiked 36%.
What explains the determined pursuit of a suboptimal ploy? Surely, bowlers must be aware that bowling at the stumps offers greater rewards? They probably are, but there is one important thing to consider here: bowling outside off-stump, with the ball swinging, produces a greater probability of picking a wicket via the highly sought-after caught, lbw, and bowled modes of dismissals.
This is seen in the data. 99.38% of wickets when bowling outside off with the ball swinging come from these three modes of dismissal, but when the line is on the stumps, this figure is down to 97.86%. The corresponding numbers when the ball is not swinging, on the other hand, are 93% for both.
Not only are these wickets the dearest possessions of bowlers; they also present a safer modus operandi in the advent of increased swing. The outside-off channel is a diversification of resources: should the ball swing in, the stumps are brought into play, and should the ball swing away, your slip fielders are in interest. By attacking less, you attack more.
In an era in which the wobble ball, with its unpredictable post-bounce trajectory, has come to be the centrepiece of bowling success, not only is such diversification of prime importance to the bowler; it is also paramount in a time of increased batting intent that run-scoring is curtailed when the field is in. Bowling at the stumps, a minute inaccuracy in release angle can cause the ball to slip into the pads. From a defensive point of view, there are few greater sins.
2. The Rise of The Hard Lengths Unless of course, you can bowl back-of-length. Before 2021, 14% of the balls bowled in the 5-10 overs phase of the ODI Powerplay were labelled as back-of-length. Since 2021, there’s been an uptick to 18%.
This might seem marginal at first - and it is - but remember two things. Firstly, the decrease in incidence of the other lengths is even smaller, occurring barely by a few half percentage points. The expansion of the back-of-length figure is what all these individually tiny reductions sum up to. Secondly, this ratio is calculated from a large sample space of almost forty-three thousand deliveries, so any marginal change visible here is likely to be indicative of a shift in patterns in the overall population of interest.
So, what explains the rise of the hard lengths? Two causal streams spring to mind. On the one hand, the explosion of T20 that had relegated the one-day format to the side-lines may have played with bowlers’ minds, and introduced to the one-day game a defensive strategy seen most prominently in the shortest format.
On the other hand - and this is probably the more apparent explanation - the war that batters have waged against bowling sanity through aggressive intent has caused bowlers to opt, at least sparingly, for more defensive lengths. This is perhaps why we see that the inflation of back-of-length is a more prominent feature of the first five overs of the Powerplay than the last five: in the same time frames, the incidence of the hard lengths rose by 13% in the 0-5 over phase and by 21% in the 5-10 over mark.
3. There is a lot of waiting around to do if you are a spinner This is not so much an effect as it is an observation - but perhaps the most statistically significant of the three discussed here. Before 2021, the average point of introduction of any spin for the first time in an innings occurred at 12.1 overs. Since then, the first signs of spin are observed not until after 16.5 overs. That is a gap of twenty-eight deliveries, or almost 10% of the innings.
Pace bowlers are ruling the roost, while the tweakers are being happy to sit back and watch.
*****
So, who does it best? Which batter is best poised for holding the fort down when the ball is swinging and charging the bowler when it is not?
There is an easy way to answer this question, and a tough way. Sure, we may glean the best Powerplay batters of this era by simply looking at the list of the highest-averaging and fastest-scoring players since 2021, but this limits our sample space and includes no mention of the more objective measure that is the control rating. So, I came up with a better way.
Since getting your eye in against the swinging new ball and smacking the stock ball seems to be the way to go forward in the brand-new ODI Powerplay, one way to identify the best batters of the current era is to look up the best defenders of swing and the best hitters of stock deliveries by control percentage. It’s a gross simplification - it is certainly not the case that batters attack only stock balls or only play a defensive shot against ones that swing - but it’s a simplification that allows us to better understand the world than any simple method does.
A couple of general inferences first. The top 10% of defenders against swing in the database show a mean 63.79% control, whereas the bottom 10% show a control rate of 60.66%. This apparent closeness between the best and the worst is replicated while attacking stock deliveries: the top decile of attackers averages a 77.57% control rate, while the bottom decile is at 73.79%. If ever you needed anything to tell you how high the standards are or how marginal the gaps are in elite-level cricket, then that is it.
So, if we rank batters in this period by control rate against swing when defending, and we rank them again by control rate against stock balls when attacking, and sum the control ratings thus formed using equal weights, we get a list of 230 batters that have been, or have the potential to be, prolific in the Powerplay. After this, I manually removed players who have recently retired, like Aaron Finch and MS Dhoni, or players who are unlikely to feature in the 2023 World Cup, like Shikhar Dhawan and Shoaib Malik, or full-time bowlers, like Shadab Khan and Boyd Rankin. This resulted in the following Top 20 list:
If you are surprised, that is because you should be. While the likes of Kane Williamson and Rassie van der Dussen are expected entries, Alex Carey and Dhananjaya de Silva are not. Nor does the ranking change all that much if instead of equal weights, you assign 3:2 weights to batting pedigree versus swing and hitting skills versus stock, respectively. A few placement changes do take place, but the top twenty more or less remain the same.
There is one essential lesson to take away from all this. The presence of only a single English batter in this list is alarming, but not entirely surprising. Jonny Bairstow is a Test mainstay, but not all others in the English one-day department are. In a period in which bowling conditions have evolved to resemble the Test format as closely as ever, England’s hyper-specialized white-ball pipeline, reared on a wave of crazies birthed by its domestic 40-over tournament, may not entirely be an advantage - for the optimal Powerplay batting strategy has shifted from whacking it hard to whacking it smart.
As the English think-tank shuttles between Jason Roy and David Malan at the top of the order, this is the second thing to consider, apart from their forms. The new ball is going to swing in India this World Cup, and Malan - the second-highest placed Englishman on this list - may be better equipped to handle its challenge than Roy.
Needless to say, this is a limited analysis. The most obvious shortcoming of the findings presented in this piece is that results for the post-2021 era are produced out of a biased sample space comprised of ODIs in India and England alone. The reasoning for doing this, as mentioned previously, has been that unless we do this the swing-friendly nature of this era is not appropriately fleshed out, but doing so stands the chance of returning conclusions that are seemingly supported by the data but in reality hold only for Indian and English batters. It is hoped, however, that this is kept from being the case by lumping both these countries together, rather than studying solely India, the venue of the 2023 World Cup, alone. It is a lesser-of-two-evils kind of approach, one through which I hope the statistical bias is minimized.
A second limitation is that the data for line and length is obtained through manual tagging of the footage, rather than a programmable process through, say, ball-tracking data. The most accurate one can expect such observations is about four in five, and that itself is a long shot. Lastly, the dataset does not include ODIs played in 2023 which, from observation, may have offered comparatively reduced lateral movement for fast bowlers.
The seam is hard, the new ball is swinging, and a generation of cricketers are in their heyday. The batting is aggressive, the bowlers have their tails up, and the ten teams are evenly matched. The 2023 ODI World Cup promises to be a cracker, and when it does, let the mind to wander back to that fateful day in Vishakhapatnam where Mitchell Starc and Suryakumar Yadav joined hands like two godsent cameo artists to give you a teaser of what is to come.
The dataset used for this analysis manually tags balls as having “swung” or “seamed”, but we know from Nathan Leamon and Ben Jones’ seminal “Hitting Against the Spin” that the probability distribution of swing is approximately normal with a mean centred around zero. This piece of information is combined with the fact that around 12% of balls from pacers in the dataset are classified as having “swung” to arrive at the 1.5-degree mark.