The Indian Premier League auction is a funny place at times. Millions of dollars are spent on players in a spectacle that draws eyeballs from around the world. Sometimes, the money is spent on veterans who seem to be well past their sell-by date. Sometimes, big money is spent on a rookie who has never even played at the level. And it might be right to say that much of that expenditure doesn’t quite make sense. In fact, it seems ridiculous.
But is it always?
India’s current U-19 and ‘A’ team coach Rahul Dravid was once heading the Rajasthan Royals team in the IPL and during a press conference in Mumbai, he recounted a tale from one of the auctions.
“Taking things back to my time with the Rajasthan Royals, Brad Hodge is a good example. Maybe it was the 3rd or 4th IPL and we at Rajasthan Royals at that stage were very much a budget team and we were looking for a middle-order batsman to fill the gap. And Brad Hodge was in the auction list,” said Dravid at a function to launch ESPNcricinfo’s new metrics, Superstats.
Now, everyone knew that Hodge was hugely successful at Shield cricket level. But he had never quite managed to cut it at the IPL. Still, as RR dug deeper, they discovered something interesting.
“We just looked at his numbers vis-a-vis the IPL, at that stage he had played at KKR and Kochi… and saw that one of things he really struggled with… where his strike-rate really came down was left-arm spin and his early IPL teams were KKR, where the ball turned square, and Kochi, which also turned a fair bit. So we looked at those numbers and said that we could do something with this. Hodge was also a brilliant player of quality fast bowling. He had a lot of success because of his ability to face and take down quality fast bowling. So we picked him in the auction and decided to bat him only in the last five overs of a game,” said Dravid.
The former India skipper added: “A lot of the people said ‘why are you doing this, look at his average for T20 cricket… look at his numbers…’. When we first told him what we wanted him to do, Hodge looked at me and said: ‘Let me bat at No 3 and I will score 450 runs for you.’ I said, ‘I don’t want the 450 runs, I want 200 runs at 160 SR against Dale Steyn, Mitchell Johnson, Malinga at the back end.’ He didn’t buy into the idea initially but he did win us 2-3 games. He then played for Australia in a T20 World Cup batting at No 6.”
Using technology to help coaches and scouts
This conversation would have been a perfect fit in ‘Moneyball,’ the book which chronicled the cash-strapped Oakland Athletics’ pioneering use of sabermetrics in their attempt to compete with Major League Baseball’s wealthiest clubs. Unlike the movie which seemed to suggest, at times, that statistics could replace scouts, Oakland’s real goal was to build a squad… an affordable squad… that could compete with the best in the business. And it worked. Oakland reached the playoffs five times in an eight-year span that ended in 2006. In that period, the Athletics had winning records every year while other teams with similar financial limitations struggled.
In a way, it is the dream of every IPL team to do the same — get a bargain in the auction, make it work in the league and win. Role players can have a huge impact in T20 cricket because they can come out and play very specific match-changing roles. The same players might struggle in ODIs or Tests but in the IPL, they might do the job.
In the United States, professional baseball, basketball and hockey are among the sports that are now using Artificial Intelligence to supplement traditional coaching and scouting.
Baseball scouts in particular have long used statistics to evaluate players. But now teams are starting to use AI to analyze an ever-expanding set of player data — including radar gun data (throwing speed and spin), video tracking (how players move around the field) and swing speed and mechanics from sensor-studded bats.
Some teams analyse this wealth of data using machine learning to identify things like how effectively players execute pick and rolls and other plays. And some are using the data to model how teammates interact to determine which players were key to successful plays.
Heat Maps in football will tell you exactly the kind of areas that your players have operated in during a game, how many touches they have had on the ball, in what areas and what their work-rate was. If a coach ever wanted to make an argument during a player during half-time, he can now fall back on data and show just where his team is going wrong.
“For example, from a coach’s viewpoint, if I wanted to point out that a particular’s batsmen strike-rate fell between the 8-12th over… when it needed to be higher given the match situation, it is easier to rely on data to make my point. The batsman may turn around and say that his overall strike-rate was 140 but if it fell during a particular period, it could impact the overall result,” said Dravid.
Not just about numbers
Sanjay Manjrekar, who was also part of the panel, pointed out just how useful data can be.
“While giving commentary I noticed that Suresh Raina was now actually playing the short ball well. It was one of the major weaknesses in his game but because teams were bowling short to him a lot, he had perhaps got a hang of it now,” said Manjrekar.
But Dravid added a caveat.
“Suresh Raina is bad against the short ball and he may be scoring more off the short ball but that doesn’t mean bowling the short ball isn’t a good idea because bowling full might see him score at a SR of 200 instead of 100. So data can play an important role in understanding that. How you read the numbers and what you get out of these numbers is what is important. You will get a lot of numbers thrown at you but the guys who can makes sense of them is the key.”
The big change in how cricket looks at data is a direct result of the T20 game. The statistics that are being used are outdated in the sense that one can’t employ the same data that one collates for Tests and ODIs to T20s as well.
The average length of an innings, ESPNCricinfo revealed in a presentation, in the top-order in T20 cricket is 15 balls now. But those 15 balls could be enough to win one game or lose it. In a game of fine margins, the right kind of data could be key but Dravid also warns against a totally data-driven approach.
“Primarily looking at numbers is probably a mistake but ignoring them is also a mistake. I don’t like to use numbers to look back, I am more interested in numbers that help me take decisions. I think there is a human element to cricket also. So you cannot look at numbers blindly. You can’t go blindly to a player and say, ’this is what the numbers, so this is what you’ve got to do.’ You have to look at the whole picture. So that is why I think geeks won’t ever completely take over because teams don’t run like that,” Dravid said.
He added: “Even as a coach you sometimes want captains who just don’t go by numbers alone. Sometimes, you want captains to do their own things. Sometimes, the numbers might say that X needs to bowl a certain over but sometimes the captain might get a feel that this bowler is not up for it today. Numbers can quantify someone if he is carrying a niggle. Numbers can’t quantify whether someone is not up for it. You can sense it but you can’t put a number to it.
“So as a coach, if a captain takes a gut call on the field, you back them especially if you know they aren’t calling it blind. That is something that Dhoni… Dhoni especially does very well. I think he gets that. I think he gets the feel of that.”
Cricket isn’t quite data driven as many other professional sports at the moment but more money flows into the sport, we will see teams looking for an edge, as tiny as it may be, and who knows... data might be it.