The Bundesliga is collaborating with AWS to organize moments during the live game to broadcast the “Match Facts” real-time program.

As the most dazzling match in the Bundesliga, the match between Dortmund and Bayern Munich has completely exceeded the scope of sports events. For most people, this “national derby” may become the only Bundesliga event they pay attention to each year.

This year, the two most successful teams in German modern football have brought new meaning to the duel, and are even expected to attract a wider viewing group than in previous years.

The Bundesliga is the only football league in Europe that has been restored after the mandatory isolation of the Coronavirus epidemic. However, considering the rapid spread of the COVID-19 virus, the game can only be played in an empty stadium where fans cannot attend.

So far, the Bundesliga has safely completed two rounds of competition. Due to the suspension of various sports events, spectators who are bored can only choose to watch football games, and the attention of the International Sports Federation to the Bundesliga events has also begun to rise.

But as other leagues resume their games in the next few weeks, the Bundesliga’s monopoly will soon be challenged. In order to maintain the current advantage, the Bundesliga organizers hope to make full use of this rare window of opportunity.

Digital plan         

At present, the technical solution has become a confrontation weapon between the major European leagues. The organizers hope to use this to expand their business scope, cross borders and open up new markets. People combine digital services with application products to help fans who are thousands of miles away from the arena, and who have never even had a live viewing experience, bring an “immersive” experience.

Data-driven insights are increasingly seen as an effective way to increase audience participation and improve the quality of broadcast programs.

To this end, the Bundesliga cooperated with Amazon Web Services (AWS) to organize exciting moments during the live game to play the “Match Facts” real-time program. In this partnership, AWS will provide cloud infrastructure and artificial intelligence (AI) tools to track the number of TV viewers and produce tailored content for fans on the online platform. Considering the huge influence of young audiences and mobile/social media in developing markets, this kind of content customization is indeed very meaningful.

This cooperation agreement was signed in January this year. Attentive viewers may have discovered that the AWS logo has begun to appear on the game screen and the overlay analysis interface. The long-term plan for the Bundesliga is to collect 3.6 million data points and 10,000 event archives based on each game, and create a set of advanced statistical platforms accordingly.

Data-driven functions       

The first phase of “Match Facts” launched with the support of AWS chose this year’s national Derby as the stage to debut. This time, the cloud giant brought two unprecedented innovations to the audience-the average formation and expected goals (xGoals, referred to as xG).

In terms of average formation, AWS can track the player’s position data in real-time, providing fans with insights on the changing trends on the field, helping the audience understand that the team is currently more offensive or defensive, how to determine tactical changes, and analyze the replacement players will be on the court. What changes it brings to the game.

Over time, the audience can even slowly understand how a team can change its skills and tactics based on available players. For example, fans may ask a substitute to play as soon as possible because he can play more threatening and aggressive combinations on the left side while he is on the court. The plan of the Bundesliga organizers is that the higher the participation of fans in the event, the greater the possibility of attracting new fans and retaining old fans.

As a second innovation, the expected goal represents a statistical model that predicts the likelihood of the team scoring a goal in a specific area on the court based on real-time and historical information. In traditional analysis, the duration of ball control, effective shots and field goals are regarded as the highest judgment indicators to measure the team’s level.

But what are the specific differences between these indicators? Obviously, it is more difficult to score from a distanceless shot outside the penalty area than to approach the goal with a light nudge. It is expected that the target model will analyze these data and provide an xG score (out of 1) for each goal opportunity. By summarizing this data, each team and broadcaster can continue to analyze the results of the game before, during, and after the game.

In the UK, the expected target model has been enthusiastically supported by many journalists who recognize the value of data. However, many conservative voices pointed out that the emergence of such indicators means that football is abandoning traditional analytical ideas and falling into the quagmire of excessive analysis. Despite some objections, in the past few seasons, the broadcast program will release xG statistics at the end of each game and has been widely recognized by the audience.

Make expectations a reality     

In the Bundesliga, the expected target model is also used to generate the “goal rate” percentage. Amazon’s SageMaker platform will host a variety of machine learning models for real-time calculation of xG results through data such as player position, distance from the goal, angle to the goal, current running speed of the player, number of goalkeepers and defensive players.

The audience can also understand the specific role played by certain actions (such as passing the ball) in a successful goal. This will increase the fans’ understanding of football, and also allow fans who already know the rules of football to be appreciated by others.

Leave a Comment

Scroll to Top