ANALYTICS: EVERY BALL COUNTS
Cricket has long been a game of numbers. But the T20 format, which arrived at about the same time as big data exploded, has embraced analytics in the sport like never before,
Cricket loves its numbers, but T20 has embraced data like never before, writes Jonathan Northall.
When there are just 120 balls per innings, each delivery represents an opportunity. For the bowling team, restricting the scoring is paramount. For the batting team, the opposite is true; runs are the currency and maximising each opportunity is crucial. The antithesis between each discipline makes T20 such a fascinating form of cricket. As the format develops, whether it be the Big Bash League or the T20 World Cup, data analytics are the catalyst.
Technology-aided data collection and analysis has been around in cricket since the late 1990s, right in time for the advent of the 20-over format. Apart from historical data from scorecards, ball-by-ball and ball tracking information also became available. These sources give cricket data analysts vital intelligence to use for multiple purposes.
Patrick Noone is one of those analysts, working for expert company CricViz. He explains how data might be used: “Ball-by-ball analysis allows us to break down players’ records, for example how a batsman has performed against different bowler types. Ball-by-ball data allows us to break that down further by providing data about shot type and footwork data, so now we can see not just what a player is doing, but how they’re doing it.
“Finally, the ball-tracking data allows us to draw further conclusions, based on exactly what the ball has done. For example, we are able to see how successful Steve Smith is at playing leg-spinners on the front foot when the ball is on a good length and spinning more than 2.5 degrees.”
The question that underpins the usefulness of cricket data is “what if?” Having abundant amounts of information gives confidence in drawing inferences. Coaches have access to data sets about venues, players and teams that can help shape a game plan. “We build packs for each match that illustrate features of the venue – whether it spins or seams, etc – as well as analysis of opposition players, their strengths and weaknesses and recommendations of who should bowl when,” Noone says.
Ask English white-ball specialist Harry Gurney, though, and he is not totally convinced of the efficacy of data analytics. It is an interesting take from a player who has featured in T20 leagues around the world such as the Big Bash and the Indian Premier League, as well as in Pakistan and the Caribbean.
Gurney was part of the Melbourne Renegades’ winning campaign last BBL season, and he’s been a member of championship sides in the PSL and CPL. He explained his reasoning: “If you sit in a meeting with an analyst and seven
“THERE’S A BIT OF A FASHION OUT THERE FOR ANALYSTS, AND CAPTAINS, TO BE QUITE PRESCRIPTIVE TOWARDS A BOWLER IN TERMS OF HOW THEY WANT THEM TO BOWL AT A CERTAIN BATSMAN.”
coaches before a game and you come up with a different plan for every batsman and every scenario, you’re just not going to remember them when you get you there, and you’re in the heat of the moment, because it is a pressure cooker.
“There’s a bit of a fashion out there for analysts, and captains, to be quite prescriptive towards a bowler in terms of how they want them to bowl at a certain batsman. I don’t play into that, which probably rubs some captains up the wrong way.”
Data analytics is not a panacea; the purpose is to highlight potential strengths or weaknesses so coaches and players can formulate a plan accordingly. Noone is quick to warn of “paralysis by analysis”, and reminds the objective is to inform correct decisions. In the book Cricket 2.0: Inside the T20 Revolution, authors Freddie
Wilde and Tim Wigmore give the example of how England used data about Chris Gayle’s weakness against off-spin to inform their in-game decision. Eoin Morgan used the tactic of bowling Joe Root to Gayle in the second over, and it paid off with two wickets in three balls, including the big-hitting West Indian opener.
In that tournament, it wasn’t only England using the work of analysts. West Indies’ coach Phil Simmons believed in utilising any information available. Cricket 2.0 recounts: “Data informed the West Indies’ decisions during the tournament. They knew that South Africa struggled against spin, so played an extra spinner and used Chris Gayle’s off-spin for three overs against their left-handers; the West Indies spinners bowled ten overs in this match.”
It was not only bowling decisions that were improved upon. “Andre Russell’s poor economy rate at the death led to him bowling earlier in the innings. When batting, Russell was told to anticipate the slower ball, which had dismissed him 15 times in T20 in 2015; he did so spectacularly during the epic semi-final run chase against India.”
Cricket has always been a mind game – trying to put your opponent into a position, perhaps several overs in advance, where they succumb to a trap. What data analysts have done is to add a forensic element to game management. In
T20, this is vital because of the lack of balls available to execute plans. Critical moments in a match are magnified – in Tests and one-day matches, single knockout blows are rare, but in
T20 there is always a coup de grâce close at hand. Having a small competitive advantage can be the difference between winning and losing. T20 cricket is still a human game but is enhanced by the plethora of data available to teams.
Whether it was the West Indies in the 2016
T20 World Cup, or the Melbourne Renegades taking out BBL08 final, data analytics have informed decision-making. It is interesting to attempt a retrospective view. Jim Laker once said: “As I ran up to bowl, Bradman seemed to know where the ball was going to pitch, what stroke he was going to play and how many runs he was going to score.” Had data analysts been around in the time of Bradman, they might have been the reason why.
DATA ANALYSTS HAVE ADDED A FORENSIC ELEMENT TO GAME MANAGEMENT. IN T20, THIS IS VITAL BECAUSE OF THE LACK OF BALLS AVAILABLE TO EXECUTE PLANS.