How data analytics could help India get back on track v Australia
Travis Head inflicted pain on India in the Adelaide Test; numbers shows the short ball on a channel could stop him
Brisbane: There is a misconception doing the rounds that India did not bowl enough short deliveries at Travis Head during the Adelaide Test. According to data supplied by CricViz, 24% of the deliveries bowled at the left-hander as he stroked a match-winning 140 (off 141 balls) at the Oval were short.
Now, 24% is almost 34 deliveries, and that is a fair amount. But with Head the key isn’t just bowling short, rather the key is where you are bowling them.
Head’s average while facing up to short deliveries in Tests is telling. When the short ball is wide outside off, he averages 173.0. When it is in the channel, he averages 83.5. When it is on the stumps, he averages 60.7. But here is the kicker: when it is down leg (into the body), that average drops to 20.6.
And the number of deliveries India bowled into Head’s body? Five. To be fair, there is no guarantee that a ball in that direction would have got his wicket. But there is a higher chance — that is what the data was telling us, and isn’t cricket at the end of the day about taking your chances?
“With the kind of data we have — we not only have line and length but we can also measure things like deviation and drift, seam and swing, and these things matter a lot in Test cricket,” said former India data analyst Himanish Ganjoo. “For example, if I have a batter who has a tendency to get out leg before wicket to balls that seam in. Now, a video person would have to remember this or watch hours of video, but with the data I can segregate very clearly that if I show that balls that move in more than half a degree do the damage. Test cricket is about planning dismissals and that is where data can help a lot.”
Ganjoo, who was with the Indian team between 2022 and 2024 (he left after Rahul Dravid’s tenure as coach ended), makes an important point about how data is used differently based on the format you are crunching the numbers for.
“The framework of the analysis shifts between different formats,” said Ganjoo. “In T20 you want to restrict runs or you want to shift the distribution of them making errors while trying to hit certain shots. So, data can help with that. But in Test matches you want to pick wickets. You want to zero in on locations or kinds of deliveries which cause the batters to lose control.”
The CricViz data shows where Head tends to lose control but the Indian team, for whatever reason, was unable to target him in that specific area. Bowling coach Morne Morkel and skipper Rohit Sharma weren’t very happy with the execution of the plans and perhaps that was the problem.
There is an outcome bias here too. In his book ‘Football Hackers, the science and art of a Data Revolution’, Christoph Biermann defines it well: “If the outcome is positive, we assume that the plan or pre-game decisions were right. If the result is negative, however, we are quick to believe that his tactics must have been bad.”
But the whole point of data is to give the captain options, to slow down the thinking and hopefully draw better conclusions.
“For a given batter,” said Ganjoo, whose PhD research focuses on early universe cosmology, “I can paint a grid of beehives where I can tell you the exact zone where he is weak. Now, a video analyst would have to look at hours of video, but I can tell you this based on five seconds of code that if you bowl at this line and length, at this pace to this batter, maybe he will get out.”
We have, of course, seen this happen before. At the World Test Championship in June 2023, India waited far too long to put the short-ball plan into motion against Head. They finally got him that way, but by then he had hammered 163 off 174 balls.
“The idea is that whatever insight you have for whatever data you can get, which is done best if the video and the data guys work in conjunction, you hand it over to the captain and then the captain uses that in combination with his own thoughts to make decisions that help the team. It is all about informing the captain the best you can,” Ganjoo explained.
There are captains or players who, on the face of it, choose not to rely on data. MS Dhoni, Nathan Lyon, and even Jasprit Bumrah, fall into that category but the only difference is that instead of relying on data that someone else had crunched for them, they instead fall back on their experience.
“Very often that aligns with what the data says,” said Ganjoo. “These cricketers have played a lot of cricket, so most times what the data says is what they know. But sometimes, it doesn’t match and those are the times you have to press on. So, people like Dhoni have understood cricket in a way that the data will align. It is not like he doesn’t use data; he just has his own data, stowed away somewhere in his brain.”
But while data may not have the same value for the experienced pro, it can be a tool that younger players can use to get up to speed quickly. There may be immense value there and one that perhaps Indian cricket needs to use even more going forward.
Recommended Topics 