Analytics - Football Simplified

The newer avenues of  the analytics are opening every day. This is the era of data. Wherever you can collect data you need to analyze that and your tool will be analytics. Sports become the new application of analytics. Various sports like cricket, football, basketball started applying analytics in their respective sports to build the proper strategy for the game.

The question is how a data analyst can help to a football coach to build his strategy for the next game. Let’s discuss it with a simple example. The below image is the passing image  of the match between Manchester city and Wolverhampton in the year 2010. By the help of  Image Processing technology we can convert the image to the  set of parameters with detailed description. In this image we indicate correct passes with blue line and the wrong passes with  red line . So what kind of data you can generate from this image. Let’s have a look.

• No: of passes each team plays
• Percentage of hit/miss passes from each team
• No: of hit /miss passes from each player etc.

Suppose Real Madrid  has the next match against Manchester city in the champions’ league, the strategist of Real Madrid can have the historical data about each player of Manchester city Data about strikers

• Demographic variables like age,height,weight,
• Playing factor data like

ü  percentage of shot from each foot ü  Shot rate ü  Shot on target rate ü  Percentage of diagonal runs and straight runs ü  Running speed with ball and without ball

• etc.

He also has the data about his team defender.

• Demographic variables like age,height,weight
• Playing factor data like

ü  Running speed with ball and without ball ü  percentage of shot from each foot ü  Ball clearance percentage ü  Counterattack  percentage ü  No: of Goal line save Etc. Just for giving example how the analytics help I am just siting a simple example. You can run the simple correlation test to find out association between opposition defender’s height and the shot on target rate of that striker from the previous matches. We have made an imaginary data on the basis of defender height & shot on target of a striker. The strategist of the Real Madrid coach can easily get this data from previous matches. After running correlation we have found the following correlation table. Here we can see both are negatively correlated. So if Real Madrid put any tall defender in front of that striker the chances of his good playing  rather shot on target will come down. Now after having your data for the long run that means if you have this kind of data about couple of matches then you can find the pattern about some particular player as well as about the team that will help to build your strategy. Analytics is becoming so popular in the field of sports that lots of courses & conferences are being organized by various universities. If  one is interested in this field he/she can easily check these out. ü  MIT Sloan analytics conference ü  Sports analytics course in Columbia Business School ü  Georgia tech university Sports analytics course ü  Manhattanville Sports Analytic Institute