Data and tech integration is exploding in the sport, beyond just the limelight of international cricket and leagues like the IPL, but also at the grassroots level. Rahul Iyer explores the phenomenon.
Most data in cricket relates to the recording of results. Averages, strike rates, and similar measures tell us what happens in the end, after the ball is bowled and hit. Peel back one layer, and more advanced data such as ball-tracking, as seen on most broadcasts throughout the world, can help contextualise these measures.
This still leaves two open ends in the understanding of cricket; one, on the processes that lead to the results we see and two – from the players’ perspective – on how they can improve. These two, naturally, are linked to each other and after all, a coach’s eye is only human and can only do so much.
“Right now the data which we have is primarily ball tracking from Hawkeye, but that explains the outcome of the ball. That doesn't explain the why, and why it happens,” Sharan Mangrulkar, founder of CricProcess, tells Wisden.com.
A fast bowler in Mumbai school cricket himself, Mangrulkar says frustration from not getting enough feedback on why he wasn’t improving at training led to the idea behind providing players with a biomechanics analysis tool. When it comes to biomechanics, fast bowling is the easiest starting point thanks to a wealth of research on the topic.
“How it works is quite simple. So someone sends us a video that goes into our system, and then we apply the computer vision tech on it that gets all the data, and then we've built a layer of biomechanics on top of that, which is backed up by scientific principles of what actually makes a bowler fast.”
CricProcess measure a wealth of data related to fast bowling; the generation and transfer of momentum, direction of the bowler’s jump, the time of foot contact on the ground and more.
Mangrulkar says they look at the ‘primary limiter’ of a player’s action. This can be an aspect that limits their pace, or their line and length. This specificity is important, in order to nail down exactly why a bowler may be struggling.
“Realistically you have to see for a player and for a coach, what is the actionable data? What do they need to work on after that? Otherwise, it's just a report, which is nice to have, and we didn't want to be there.
“We wanted to be a tool, used in an essential part of the coaching process. How it works right now for our coaches is that they onboard a player, first thing they do is they send the video, and then we analyze the video, and then we give them the report back. So, for that, we have a summary, and we have the entire longer report, which has it all.”
But batting, he admits, has been more of a challenge. “We are first creating the data set, because that data itself doesn't exist.
“The batting [side] is more like being like scientists, rather than building a product, because you're looking at independent variables and dependent variables and trying to find correlations. In a sense that's good for us, because there's more opportunity, there's much more of a gap which needs to be filled.”
Some of the batting insights rather confirm conventional wisdom, indicating that at least some of the existing knowledge has sound foundations. For instance, the fact that all elite batters play the ball right under their eyes. This is a known conclusion, but Mangrulkar says the backing from data enables us to positively call it a way to distinguish elite batters from non-elite ones.
“The other thing is that they track the ball very well,” he adds. “So we were looking at [Yashasvi] Jaiswal, Abhishek [Sharma], and Vaibhav [Sooryavanshi], and for all of them, you saw that even if the ball was on the leg side, then the bat swing would go in the correct direction where it has to go from, and if it's on the offside, it'd go from the opposite direction, and before the ball lands. So just knowing where the ball lands before it does, that's something which is very high in elite batters.”
He elaborates on the process: “The head position is one system, the bat swing is one system, and then the weight distribution on the feet is another system. Then you're looking at that for each batter. At the moment batting is more qualitative. So the analysis we've done so far is more qualitative. Our aim is to make that quantitative, to try and standardize that into, ‘These are the factors, these are the things to do.’”
Mangrulkar also worked with former India men’s team analyst Himanish Ganjoo on examining Sooryavanshi’s biomechanics. From their discussions, he says that the future of biomechanics data is bright, as it ties directly into the work coaches do with players, on tweaking movements and timings. “Sometimes they don't trust all the other data, because they don't know why it happens or how it happens.”
Also read: Rapid hands, split-second reactions: What makes Vaibhav Sooryavanshi a batting tornado
-----
Even outside biomechanics, the wealth of information available to today’s cricketers on what and how they perform in training is lightyears ahead of the scene even a decade ago.
“Some of the coaches at the highest level call this a conversation starter,” Rahul Nagar, co-founder of the company Str8bat, tells Wisden.com. “Data is bridging the gap between player and coach.”
Now ten years old, Str8bat’s sensors, which have now evolved into stickers no thicker that fit onto the back of a cricket bat, were one of the first evolutions in going beyond measuring what happened, into how it happened, both in training and matches.
“The bat knows exactly what’s happening,” Nagar says about the genesis of the company. “We started discussing, if there's a way to understand what the bat is saying, then we will have a good technology on hand. So there are motion sensors – we call them IMU sensors, which are accelerometer, gyroscope, magnetometer. They capture the motion of any moving object.
Also read: Virat Kohli looks as dangerous as ever - how has he adapted his T20 game?
“We capture the motion and convert the raw movement into interpretable motion signatures using our proprietary intelligence platform, right from the speed of the bat… or the speed of the bat when it actually hits the ball, because that's more important. You might be swinging the bat at 100 kilometers per hour, but when it's actually impacting the ball, that's where the energy is getting transferred.
“Players fundamentally want to understand their own game and improve. Coaches want clarity. Parents want measurable progression. We built around that. str8bat works directly on the bat without altering the weight of it. There are no cameras, markers or complicated setups. A player can train naturally and instantly receive objective feedback. That matters enormously in grassroots environments.”
Timing ratings and how well batters hit the ‘sweet spot’ of the bat are also measured based on the energy dissipated through the bat after contact.
Nagar explains how cricket teams and players have made use of this technology: “One IPL player, in his stance before attacking a ball, was vigorously moving the bat. If you put a camera in the front, it was not very easy to capture, but when the coach saw that [the data], that this is how he's loading when he's just waiting for the ball to come, he was losing a lot of stability. So then the coaches worked on making this more stable and less vigorous.”
Str8bat is used extensively by IPL franchise Rajasthan Royals and Cricket Australia, to name two organisations at the top level. While CA mainly use it during high-performance camps, it did come in handy during the Covid-19 lockdown, enabling a centralisation of training data while the cricketers practiced individually, and could not be physically present together.
Nagar says RR also use this technology for player scouting. Team analyst Giles Lindsay told Cricbuzz last year about its role in shortlisting Vaibhav Sooryavanshi based on his bat speed, and his subsequent work on that aspect of his game with batting coach Vikram Rathour at the franchise’s high-performance centre.
The democratisation of information is Str8bat’s primary goal, Nagar says, and their recently-inked partnership with KL Rahul excites him. Similar to how Mangrulkar and CricProcess are creating the database on ideal biomechanics for batting, Nagar and Str8bat think the ability to monitor the metrics of a top-class player like KL Rahul can have incredible knock-on effects for the formative education of batters, right down to the grassroots.
“With KL Rahul, we are building a system where one of the world's best batters can effectively influence and guide millions of learning journeys through data and AI. If we can work out a way for the other cricketers, especially the grassroots and budding cricketers, to get these details of him, then that's also a very good legacy he will leave.”
Also read: Anatomy of a fall: Has the Varun Chakravarthy mystery finally been solved?
-----
If CricProcess work off video and Str8bat listen to the bat, so to speak, Arc Simulations track the ball itself.
Lawrence Booth first reported for the Daily Mail on their chips inserted into cricket balls at training, used by three counties in England. Co-founder Michael Armenakis spoke at length on the technology, on The Scoop, Wisden’s new YouTube show.
“[The tags] allow us to track each stage of the sporting event. And in a cricket context, that is the bowler’s run-up, release point, bounce point, contact point. And just before the ball hits the parameters of the net, we take the trajectories, angles, velocities of the ball and feed that into a physics simulator to play out the result, effectively, of whether you’ve been caught at point or if the ball’s gone over cow corner for six.
“Launch angle, shot power, contact point… not yet bat speed, but that’s something that we will employ in our development. From the bowler’s perspective as well, we get really accurate delivery speeds, lengths, pitch maps, release heights, swing curvature, all of the information that bowlers and batters both need.”
Teams can access this information through the company’s app. Having access to this information in training, gives them two advantages – a larger sample of data with which to judge players (compared to just in-match information), and the ability to tailor and tweak training sessions to be more focused. A video on Arc Simulations’ Instagram page shows how Kent’s Mikey Cohen walked through a session with analyst James Tomson with the help of the app.
What is far more tantalising though, is the plan for Arc Simulations’ next phase. The company are working on measuring RPMs (revolutions per minute), with a component of direction, so that players can experiment better in the nets with overspin, undercutting the ball, or sidespin. This same tracking technology can also be adapted to track fielders – the fielding and catching side of cricket remains grossly understudied, mostly due to the paucity of hard data.
While all these initiatives are new in cricket, they are long overdue. “It’s proven that data and simulation is part of other athletes’ processes,” Armenakis says.
“So we really feel cricket has some catching up to do, and hopefully we can bridge that gap, really, and supplement the work that coaches already do.”
Follow Wisden for all cricket updates, including live scores, match stats, quizzes and more. Stay up to date with the latest cricket news, player updates, team standings, match highlights, video analysis and live match odds.
