How analytics makes football more fun

It is said that data is the most valuable resource of the 21st century. It has proven to be a key factor in decision making process of businesses in almost every industry. As a result, many organisations are turning towards data-driven framework of operations and have established a data management structure to make the best use of available data. Over the last few years, businesses in all the major industries have transformed their operating framework by adopting this approach with education, healthcare, travel and tourism, finance, retail, media and entertainment to name a few. Sports industry has also seen a revolution in terms of using data for performance assessment and pre-match preparations. Basketball and Baseball were the sports that first made use of the statistics to prepare for matches and assess performance of players after every match. Football has also followed this trend, with many football clubs and national teams across the globe now relying on data to be successful both on and off the field.

Football clubs these days have a dedicated data analytics team to monitor training schedule and fitness of players and assess their performance after every match. This team is also responsible for studying the pattern of play and identifying strengths and weaknesses of the opponent before every match using data from the previous matches played by them and their results. This is a key factor in preparation for the match as the tactical approach for every match is decided by considering the insights gained from statistical analysis. This method of preparation has proven to be very effective. There are many examples of teams with limited financial resources that have defeated other teams superior in terms of capital and quality of players by leveraging the power of data analytics.

Monitoring Training Schedule and Fitness

Modern football is demanding due to increasing number of games being played every season. Apart from club football, players also take part in international tournaments and friendly matches with their respective national teams. To be able to play consistently at the highest level, players need to be fit throughout the season. As each player has different traits and body type, training schedule and intensity has to be managed differently for each player and customised as per their playing position on the field. Due to this, football teams are moving towards data-driven monitoring and scheduling of training sessions for players.

In training and during matches, players’ movement is tracked by using data collected by wearable chips. This enables analysts to analyse data such as number of sprints and distance covered by each player. This data is used to determine which players are in good shape physically, which players are exhausted and need to be rested, and which players need a different training schedule or intensity to be more effective. When players are monitored continuously over a period of time, it reveals physical traits of players and if this data is analysed considering players’ playing position on the field, their diet can be adjusted accordingly to get the best out of each player. Knowing which players need to be rested or need change of intensity in training can help in preventing injuries. This makes sure players are available all the time playing in peak physical conditions

Germany’s national team used SAP Match Insight tool to help players make adjustments to position, ball handing and speed during 2014 FIFA World Cup Final. Image and Caption Credits: Steven Norton’s article in The Wall Street Journal

Player Scouting and Recruitment

Traditional way of collecting information and evaluating players involves monitoring their performances and progress, by actually watching them play over a period of time. Every club has a scouting department that includes player scouts who are responsible for monitoring players and generating detailed reports on them. Scouts are assigned to work at various locations across the globe and this helps clubs to widen their reach and unearth special talents that are hard to find otherwise. Players are monitored over a period of time to gain understanding of their traits and judge their suitability for the club. The scouts then report their findings to the management.

The traditional approach of finding new players can be combined with modern data gathering techniques to make recruitment process more precise. Over the course of a season, huge amount of data is generated about each individual player, including their overall performance, their contribution to team’s success and their level of influence on how the team plays. This data can be termed as performance data. If this data is combined with players’ personal data such as their age and contractual information, their net worth or market value can be estimated when players switch clubs in the transfer market. This method of evaluating players’ worth can be very useful when clubs negotiate with other clubs for buying or selling a player. There is no standard process to decide value of a player and hence, value estimation using data analysis can be used as a starting point for these negotiations. Some research has already been carried out in this area and researchers have proposed solutions for estimating market value of players using machine learning techniques. There is a lot of potential for research on this topic and latest machine learning and deep learning approaches can improve the performance of such predictive models.

Understanding the Context of Statistics

The statistical analysis of players’ performance data can sometimes mislead analysts when they are trying to identify potential new recruits for a club and understand if that player is suitable for the team, without considering the context of the statistics. For instance, if a player has good passing statistics, but he plays in a team that tends to keep the ball for the majority of the game, then it doesn’t necessarily mean he is a good passer. If a player scores a lot of goals playing in a league that is not highly competitive, then there is no guarantee that he would be able to perform at that level in a highly competitive league. Hence, statistics don’t always tell the whole story on their own. In some cases, intuition and judgement of analysts or managers also plays a key role in assessment. Statistics provide a direction for analysis and when combined with human instinct, can be very effective in sports analysis.

Final Word

Data analysis has proven to be a decisive factor in football and sports industry in general. The insights gained from data can make a huge difference in key moments during the game. Sports science has evolved with the use of data as keeping players fit and ready for each game is vital. Also, getting the best value for players when selling them, and identifying potential new recruits who represent good value for money is key to be successful both in business and football perspective. Attention to detail is key in competitive sports. Using data analysis for preparation and assessment enables clubs and players to get competitive edge over their opponents. This often turns out to be the difference between a player scoring a goal and getting shrugged off the ball by a defender, a team winning the championship and settling for a runner-up spot, a player getting a move to his dream club and being rejected by the club for failing the medical test at the last minute. Put simply, it can be the difference between success and failure. These are fine margins and hence, if the available data is used appropriately considering the context of application, it has the potential to revolutionise sports science and influence results in the modern game.