What does it take to nail batting roles in T20 cricket?

The demands of T20 cricket, arguably the most fickle format in the world, are ever-evolving. Yet data analysis and data science are slowly taking shape in the industry, helping players and teams adjust to the constant changes in the landscape. One of the key ways in which analytical work can aid decision-making is by understanding and demonstrating the primary driving factors behind success.

So what differentiates the most successful IPL batters from their peers? By looking at the three phases of a T20 innings separately and mining the data on each one, we start to see some of the characteristics that have allowed certain batters to stand out from the crowd, and what skills might set them apart in the future.

The PowerPlay - a story of intent and length

The powerplay is statistically the most important phase in T20 cricket. Since 2015, the team that has had the higher impact in the first six overs (calculated via the DLS method) has come out on top 68% of the time. Historically speaking, a team that loses three wickets inside the powerplay has a 70% chance of losing. Maximizing the powerplay, which is a combination of frequent boundary hitting and wicket preservation, becomes an integral cog in the wheel of T20 strategy.

A lot has changed in the shortest format over the years, but a few overarching trends have stood the test of time. For example, the lengths fast bowlers bowl in the powerplay have remained largely constant over the years, with nearly 71% of deliveries in the good (6-8m from the stumps) and hard (8-10m) lengths.

Bowlers have continued to operate in this area because lengths on either side of 6-10m are hit fairly easily in the powerplay. Not only is it tough to set a field for full and short deliveries, but it also feeds into most batters' go-to attacking shots. On average, batters strike at around 160 outside the 6-10m length and around 120 within it.

After length, lateral movement is one of the most crucial factors in determining the difficulty of hitting a ball. Swing lasts less than three overs in a T20 innings. After that, it drops below one degree, a figure that remains constant until the 16th over, when a tinge of reverse creeps in. And interestingly, the higher the count of boundaries in the first two overs, the sooner the ball stops swinging. "Hitting the shine off the ball" is a real phenomenon even in the shortest format of the game.

Good powerplay batting is a combination of two main skills - high intent - which means that the bowlers never get a chance to work their way into the game and the most important phase for batting is maximized - and the ability to convert 6-10m deliveries into run-scoring opportunities.

There are two main ways to implement the latter skill. The first, adopted by batters like Virat Kohli and Abhishek Sharma, is to come down the track and get closer to the pitch of the ball, opening up areas down the ground and over the infield. The second, used by Travis Head and Sunil Narine among others, is to play the pick-up shot over the leg-side after creating room by moving around in the crease. Batters like Phil Salt and Faf du Plessis, meanwhile, combine these two approaches.

These seven openers ranked in the top 10 in both total runs scored as well as highest impact per ball in the powerplay in IPL 2024. Impact, calculated via the DLS method, is an excellent way to establish a relationship between wickets lost and runs scored. It quantifies the net runs added by a batter to the projected score and is ideal for determining the risk-reward factor of a particular approach.

The approaches are different, but as mentioned earlier, the results follow a distinct pattern. The best powerplay batters in the IPL are not necessarily good all-round batters - for example, Kohli, Du Plessis and Narine had strike rates below the league average outside the 6-10m range. But all seven of them showed distinctly above-average performance against the length bracket that they faced the most often.

The below two graphs illustrate that clearly:

There's no way around it - an elite T20 opener needs to have elite length-conversion ability, especially in the first half of the powerplay when swing is a factor.

And, of course, there's intent.

Among the above seven openers, in IPL 2024, only Narine had a strike rate under 150 in his first 10 balls. Fraser-McGurk, Head, Salt and Abhishek motored along at astonishing strike rates above 175 in their first 10 balls, which was much higher than the league standard of 135.24. Needless to say, their initial boundary percentages were well above the league average of 18.97%, with Narine's 24.32% being the lowest and Fraser-McGurk's 42.86% being the highest.

In modern T20, there isn't any time to "get set". A batter who takes time to "get set" simply cannot open the batting, given how valuable the powerplay is as a resource. And a batter whose natural instinct is to defend good and hard lengths - Ruturaj Gaikwad and KL Rahul, to name a couple - isn't doing enough to get their team ahead of the game early on.

The Middle Overs - where dip, height, and short balls reign

The middle overs are, in a sense, more conditions-dependent than the other two phases. The powerplay is dictated by the field restrictions, while the death overs are dictated by the number of balls remaining. But the intermediate phase has flexibility, meaning that the pitch is usually the most influential factor in teams' style of play.

In IPL 2024, pitches on the whole were significantly flatter compared to previous years. Average pace lost after pitching and average turn reached all-time lows among seasons in the previous auction cycle, making it easy for batters to trust the trajectory and hit through the line. Spinners bore the brunt of this, suddenly getting less than two degrees of turn in the middle overs for the first time in years.

Despite that, wrist-spinners managed to stand out, as they have been doing over the years in the IPL. Without any assistance from the pitch, they were reliant on deceiving batters through the air and were able to do so thanks to dip. Dip, combined with length accuracy, proved near-impossible to hit even amid the flat tracks, high intent, deep batting lineups and short boundaries.

The 5-6m length is statistically the best one for a spinner to bowl, with strike rates dropping under 110 - well below 4-5m (132.2) and 2-4m (163.6). Spinners who pitched the ball in the 5-6m range in the middle overs in IPL 2024 had a bowling average of 16.1 and an economy rate of 6.21 when the ball had dip on it. When it didn't, these figures shot up to 37.56 and 6.85 respectively.

Wrist-spinners, who get considerably more dip than finger-spinners thanks to the overspin they impart, were able to stay afloat in the bat-ball battle, relatively at least. Nine wrist-spinners managed more than seven wickets in the season, while only four finger-spinners did, three fewer than the 2023 campaign. In a notable example, Varun Chakaravarthy has admitted to having turned his career around by focusing on overspin.

So how can a batter counter dip? Height could possibly be a significant factor. Batters are protected by the insurance of their reach to adjust to balls that dip and therefore pitch shorter than they initially anticipate, with the added benefit of having the long levers to rock back and pull deliveries that others may not be able to power.

Countering dip isn't easy, and the non-elite spin-hitters who aren't among the best in the world off primary attacking shots like the slog-sweep and coming down the track need every advantage they can get. Among the 26 batters who faced more than 30 dipping spin deliveries in IPL 2024, only six averaged more than 35 while striking at above 150.

Faf du Plessis, who averaged 46.33 and struck at 146.3 against spin last year, isn't an elite spin hitter. However, against balls that dipped, he averaged 61 and struck at 179.41. Axar Patel's strike rate increased from 147.9 to 182.9 when the ball had dip. Shashank Singh, Cameron Green and Sai Sudharsan weren't dismissed by a single dipping delivery. All five are listed at above six feet tall.

The shorter batters, in contrast, didn't have it easy. Head wasn't dismissed by a single non-dipping delivery but averaged only 10.6 against deliveries with dip, with his strike rate dropping by 10 points as well. Rishabh Pant's average plummeted from 42 to 20 when dip was involved, while for Prabhsimran Singh the fall was even steeper from 42 to 11. Ishan Kishan and Sam Curran also witnessed drops in their averages.

The below graph illustrates how averages fluctuate based on dip.

There are exceptions, of course. Heinrich Klaasen, KL Rahul and Shubman Gill, all of whom are taller than the average batter and capable against spin, averaged under 20 against dip. The elite spin hitters, including Abhishek Sharma, Rajat Patidar and Nicholas Pooran, have been perfectly comfortable despite being under six feet tall.

It isn't a given that a taller batter will be able to play dip and wrist-spin better, and it isn't practical to correlate these numbers with the heights of all batters involved given the small sample sizes and lack of access to the relevant data. But it's certainly something worth thinking about from roster construction and bowling strategy perspectives.

The middle overs aren't all about spin, though. Teams have grown increasingly fond of deploying their hit-the-deck pacers through that phase, and splitting the phase into overs 7-11 and 12-16 is natural. Pacers bowled nearly 60% of overs in the 12-16 phase in IPL 2024, compared to 48% in 2023. Even in the 7-11 phase, the number of pace overs rose from 27% to 36%.

Without swing and not much seam, fast bowlers have become reliant on bounce. Close to 50% of pace deliveries bowled in the middle overs in IPL 2024 were shorter than 8 metres. Some of the league's best spin hitters, despite dominating their frontline matchups, were caught out by the increased amount of hit-the-deck bowling.

Rajat Patidar, Yashasvi Jaiswal, Jitesh Sharma, Sunil Narine and Nicholas Pooran, all of whom are known to be deadly against spin, averaged under 20 against deliveries shorter than 8m in the middle overs - well below the league average of 26.77. Pooran and Narine were narrowly above the league's overall strike rate of 141.8 while Patidar was just below, but Jaiswal and Jitesh were found wanting on that front by big margins. India's two main T20 finishers, Rinku Singh and Hardik Pandya, averaged under 10 while striking at under 110.

Among the 36 batters who faced more than 20 balls of this type, only two - Ruturaj Gaikwad and Rilee Rossouw - struck at more than 180. This is despite the fact that the pull, the hook, the upper cut and the ramp, which are the four main shots played against short balls, all produce almost two runs per ball on average.

It's simply not sustainable to rely on the aforementioned primary attacking shots against short balls through the middle overs. Playing the cut well is an important factor, and those who generate more than the overall strike rate of 125.3 while attempting the shot are at an advantage. But the key is being able to negotiate the short balls faced without losing a clump of wickets, and capitalizing on the other run-scoring opportunities.

Shreyas Iyer, who has long been criticized for a perceived inability to play the short ball, did this well in IPL 2024. The title-winning skipper wasn't dismissed by a single delivery shorter than 8m in the middle overs, collecting 67 runs off 55 balls at a strike rate of 121.8. Other batters who aren't the greatest pullers can afford to adopt a similar strategy, assuming they have a partner at the other end who's better equipped to tackle fast bowling.

An example of a data-driven change would be to move Gaikwad into the middle order. The CSK captain may not have the intent and length-conversion ability to be an elite T20 opener, but he can certainly be the ideal foil for Shivam Dube given his prowess against hard lengths and spin through the middle overs.

The Death Overs - combatting pace-off and even spin?

To understand what a death-overs hitter needs to do, looking at the strategies bowlers employ is a logical first step. The yorker has been discussed at length when it comes to death-bowling plans, and even today, most teams utilize strategies that revolve around their fast bowlers being able to hit the blockhole.

But as teams become more data-conscious, they are bound to realize that variance (or unpredictability) could be more important than execution, even of a yorker. Moreover, the attributes of a bowler - release height, amount of backspin imparted on fuller balls, amount of dip generated on slower balls, pace, and so on - could also be more important than execution. The days of "back your yorker at the death" without taking the attributes of the bowler into consideration should be long gone.

To illustrate that, here's a simple example. Contrary to popular belief, Jasprit Bumrah isn't the most accurate yorker bowler in the world. Dwayne Bravo (43.2%), Bhuvneshwar Kumar (42.5%) and T Natarajan (37.3%) are among those who have had higher yorker success rates than Bumrah (36.7%) over the last three IPL seasons. But while Bumrah goes at an economy rate of 7.02 and Bravo at 7.54 on yorker attempts (successful or not), Bhuvneshwar and Natarajan go at 8.57 and 10.36 respectively.

The difference lies in the revolutions Bravo and Bumrah put on the ball. While Bravo generates considerable dip (caused by overspin) on his slower balls, which form the vast majority of his death-bowling plans, Bumrah has the dual advantage of dip on his slower balls and appreciable lift (caused by high backspin) on his on-pace deliveries.

Considering that a quality death bowler employs pace variations, picking and hitting slower balls automatically becomes an extremely important skill for a death-overs batter. Over the last three IPL seasons, 36.7% of the balls delivered by pacers at the death (overs 17-20) have been below 128 kph, which is the rough threshold value for a slower ball. If roughly one in three balls is a slower ball, then irrespective of whether it's an off-cutter, a leg-cutter, a knuckle-ball or a back-of-the-hand slower ball, it needs to be treated with the same attacking intent.

Tristan Stubbs and Dinesh Karthik were two of the four batters with the highest impact per ball at the death in IPL 2024. Their strike rates against slower balls were 285.7 and 227.17 respectively. The others in the top four, Heinrich Klaasen and MS Dhoni, had strike rates well under 200. As things currently stand, even the best death hitters in the league have clear weaknesses that can be exploited if oppositions have the right personnel and employ intelligent data-driven tactics.

Rahul Tewatia is among the better finishers in the league and ranks eighth in impact per ball at the death, but his strike rate drops by nearly 25 points when pace is taken off. The same goes for Ravindra Jadeja, who strikes at 140 under 128 kph and at 184 above 128 kph.

Reduced speeds producing a drastic difference in attacking output isn't restricted to the realm of fast bowling. In no surprise, as many as four of the above six batters are not known to be elite players of spin. There's a serious case for oppositions to backload spin against batters of the ilk of Dhoni and Tewatia, much like then-PBKS captain Sam Curran used Rahul Chahar to stifle the CSK legend in a decisive 19th over last year.

The death-overs hitter of the future, perhaps even in IPL 2025, will be required to combat various kinds of spin at the death. The Paarl Royals took this to the next level in the SA20 league, where a slow wicket at Boland Park allowed them to use well over 15 overs of spin at times. Teams like CSK, who have three quality spin options and a favorable home venue, will certainly consider this approach. The data backs it up, with spin at the death having a better average, strike rate and boundary percentage than pace.

So a quality death-overs hitter, or a bowler for that matter, will need to become significantly more well-rounded if they are to keep up with the rate at which modern T20s are evolving. A batter will need to be capable of handling all variations to a decent degree. If they have a notable weakness, such as Dhoni against spin or Jadeja against slower balls, it will be ruthlessly exploited by teams that pay attention to the numbers.

Sai Krishna is a freelance cricket data analyst and data scientist who is currently working with international cricketers and coaches