In cricket, a few numbers can easily decide the fate of the game. Know How

Across the world, sports teams, investors, and fans take their sports very seriously. They would do anything to up their game and get a result in their favour.

In the heat of the moment, where split-second decisions can decide whether you finish first or second, Big Data provides a calm and clinical aid to decision making. Big data in the sports industry has been gaining momentum since 2007.

Coaches and players are using Big Data to make more-informed decisions that could have a big say on how much prize money they take home. With the corporatisation of sports, teams can select the best players, field the most effective teams and make smarter decisions on the field or court.

Lets take the example of cricket; here are three ways data analytics can improve efficiency, accuracy and profitability in sports.

Cricket is a very popular game in India and one of the richest games. Cameras, sensors and wearables record every aspect of player performance. Managers, coaches and athletes are using data to dictate calorie intake, training levels in the chase for better performance on the field.

While playing the game, both teams have to make very important decisions while the game is in progress.

The team captain, along with important players of the team, decide who will bat first. They also determine the entire batting sequence with the team coach and the senior players.

So what are the variables? There could be thousands of them!

The inputs of the problem will be the available data set which includes, players track record in various field conditions, player form, opponent skill set, opponent’s track record against players on the pitch where the game is being played, other weather parameters, wind speed and its direction, humidity and current temperature and its forecasts, new ball effect, rate of ball deterioration, country where the match is played, spectators and their number, spectator behaviour, day or night match, required run rate, current run rate, umpires judgment record, ground condition, pitch record and several other static and dynamic data factors.

The design and development of this algorithm will use concepts from game theory, machine learning which includes both supervised and unsupervised learning and some aspects of artificial intelligence (AI).

Based on all the processed inputs, the algorithm can dynamically decide which player should bat (from the batting perspective or which player should bowl from the bowling perspective) and in real time as the game progresses. This will assist in making right decisions to maximise the chances of winning a match.

Brad Pitts 2011 film Moneybag had him posing as Peter Brand, a Yale graduate with a degree in economics, who believed that baseball isn't about instinct and intuition and hunches; but it’s all about the numbers. His selection of players and the theme of the film, indicate that exacting statistical analysis is here to stay in sports.

The coach and his computer are now inseparable!

(About the Author: This article has been contributed by Dr. Mahendra Mehta. He is the Program Director for Big Data at the SP Jain School of Global Management.)