Usage of Analytic Data and Devices in Soccer

Introduction

The game of soccer or football has historically been reluctant to accept changes of any kind. It is one of the least quantified team sports, it is not a game of numbers; it’s low-scoring, features few individual statistics and lacks the figures that many sports fans crave. Many view it from a purist perspective and do not want to see the simple game changed with all the modern advancements in technology. So when the movement to use new technology that analyzes all aspects of each major sport began to be implemented across the globe, like always, soccer was hesitant to adapt. Soccer has been known to be a sport where one’s ‘gut feeling’ and experience/knowledge of the game is often still valued above statistical evidence.

Over time, even the sport of soccer has begun to accept Big Data and analytics to help each team try and get ahead by equating all aspects of each player’s performance into data that can be reviewed in decision making. The use of all this new data is all about finding the extra 1% that can exploit even the slightest weakness in a team’s opposition and hopefully make the ultimate difference in one’s results; increasing their number of wins and decreasing their number of loses. The better performances lead to more than just new silverware for a team’s trophy cabinet; with the hundreds of millions of dollars being invested into the sport each year, a team’s financial standings can be changed forever the higher in the standings they finish and this will affect everything from the ability to add better players to their team as well the better financial standing for the organization, its owners, and its investors. Rather than simply applying data to tactical performance, objective information is now used throughout almost all leagues and clubs across the globe to enhance efficiency and develop processes that enable these teams to be as well-prepared as possible for their games on the pitch as well as in the owner’s boardroom meetings (LinkedIn).

Modern Uses of Analytics in Soccer

Today, all 20 Premier League football stadiums in the United Kingdom and in almost every team in the top leagues throughout the world are equipped with a set of 8-10 digital cameras that track every player on the pitch. Around 10 data points are collected every second for each of the 22 players (11 per team) on the field, and they are generating 1.4 million data points per game. This is capable because of one of the companies that pioneered analytics in soccer, Prozone. Prozone analysts will then code the data that they take from every match played to identify every tackle, shot or long pass in order to enable managers and performance analysts to gain insights of what exactly happened in each game, on and off the ball (LinkedIn).

Another company, 21st Club, has created PIRLO (Predictive Intelligence Research and Learning Outputs), an analytics engine with information on 150,000 players. Named after former Italy midfielder Andrea Pirlo, the consultancy company uses this program with similar machine-learning algorithms as Netflix and Amazon to offer smart insights and recommend players to look at in upcoming transfer windows who may not already be on club’s radars (Forbes).

There is a problem when looking at through all of this data though. Every league in each country has its own playing style and overall skill level. To make a comparison to American sports, a baseball player may be doing very well in the lower divisions in the minor leagues, but that does not mean his stellar performance will translate over to the highest level of Major League Baseball. So now to translate this back to soccer, a player may be posting excellent numbers in the Eredivisie (the Netherlands) may look like a superstar against opposition in his country, but how would they perform if moved to The Championship in England or to a team battling for the top positions in the Premier League where the play styles are greatly different (Forbes)? Quantifying the strength of shot data and other key numbers players produce in Holland needs to be correctly weighted, taking into account strength of opposition.

Some forms of these newer technologies include multiple camera semi-automatic systems, Local Position Measurement (LPM) systems, and GPS systems, each capable of quickly recording and storing data about a team’s players. GPS systems is a navigational system that uses information sent by multiple satellites to calculate the geographical position of a receiver (i.e., longitude, latitude, and altitude). The modern GPS receivers have integrated triaxle accelerometers, which measure the acceleration in three planes to produce a composite vector magnitude known as G-Force. The integrated accelerometers can be used to quantify the forces acting on the player, known as body load, and can also be used to measure impact of players with other objects and surfaces (Chelsea Academy). Unlike GPS systems, where devices are passive receivers of signals from overhead satellites, LPM systems work by having the wearable emit signals to local receivers, which do the actual triangulation and have shown to result in better accuracy than GPS-based systems (Pettersen).

Wearable Technology in Training

Due to recent rulings made by Fédération Internationale de Football Association or FIFA, wearable technology has been made legal for teams to track players’ performance in both matches and in practices and will help analytics teams study all aspects of a player’s performance. Although, achieving optimal physical performance and minimizing risk of injury can be conflicting, improving physical performance often necessitates intensive exercise regimes which may increase the likelihood of player fatigue and thus risk of injury. Managing training loads is crucial in enabling players to physically perform at an optimal level across the duration of a playing season (Chelsea Academy).

Visualizing Results

After analytics team gather all their data from each player’s performance, they are now expected to visualize the data and make it presentable so any manager or player can look at this data and be able to easily understand it. To help analyze player positioning, GPS technology is used to quantify a player’s location and speed at all times. Research and practical applications resulting from positional data are much focused on the descriptions from the activity profiles, such as quantifying the physical and physiological demands of soccer players.

When analyzing player positioning, key metrics were established in order to best calculate the overall results. Centroid is the average position of the outfield players and is calculated by taking the mean lateral and longitudinal position of all outfield players' coordinates at a specific time, each of them contributing equally to the computations. The centroid of the attacking team moves forward. The defending team tries to prevent the attacking team from scoring by moving backward. It is employed to measure intra and inter team coordination in a group of players and it reveals important facts about how a team sets tactical and other valuable data for the analyst (Frontiers). Centroid can then be used to analyze scoring opportunities based off player positioning. It can also be used to measure your team’s average centroid where you can then try and discover your how well your team is working together and identify potential outliers for those whose centroid is in line with the rest of our team’s.