Introduction
In the course so far, we have seen various networks used to represent a wide range of relationships and situations such as friendship bonds, email ties, and website linkages with the help of graph theory. The concept of network graphs might seem very simple, yet is such a powerful tool for multiple fields to help understand or analyze much more complex real world problems and cases. One interesting problem in which graph theory can be used to analyze and even increase performance is sports. Sports is something that has become almost a part of the global culture. No matter where you are in the world, there is always some sport that people care about a lot, and has even become a part of the culture. The largest one globally is, of course, soccer.
With over 3.5 billion fans worldwide, soccer has consistently been the biggest sport in the world, with the world cup as the biggest sporting event that gets held every 4 years. Nowadays, soccer at club-level is arguably just as, if not even more popular than international-level soccer. With the advancements in technologies in all areas of the sport like training, fitness, conditioning, and many more, club-level soccer (especially in Europe) has reached new levels in terms of competition and entertainment.
Research
In 2019, a group of researchers performed a study, with the purpose of analyzing what made the FC Barcelona team under Pep Guardiola arguably the best soccer team of all time, using a Network Science approach. The researchers created a “passing network” of each match this team played in a single season in the league (38). A passing network in this case is a described as a graph where the nodes are players in the team, summing up to 11, and the edges between them are the passes performed in a particular soccer match. Figure 1 below is an example of such a network for a particular match.
Some noticeable characteristics in this particular graph is that of course, most of the edges occurs in the middle of the field, where most of the ball is in a soccer match. There are players that have a significantly lower number of edges like the goalkeeper (Valdes) and central defenders (Puyol). Most of the edges occur in the central part of the field (parallel to the goal) instead of wider areas, which shows a signature style of this football team where they play passes through the lines centrally more than focusing on wider areas. These are all good observations at first glance, but what does the numbers and quantitative data tells us about the team?
From all the data gathered, more analysis is performed such as the clustering coefficient, average shortest paths, and more, in order to study how the average “passing network” of this particular team compares to those of the rest of the league (shown in Figure 3).
Results
The results that the researchers got from the data was that most of the metrics used and taken from these “passing networks” were indeed “better” in the Barcelona team compared to the rest of the league. Higher clustering in the team showed the low expectancy to lose the ball when making a pass from the players compared to other teams, lower average shortest path length showing that it takes less amount of passes (and therefore presumably time) to get the ball between any two players, and more. After obtaining results and excluding as many external influences on the data like how much of the ball the teams actually have during a match, a few significant findings were that this particular team had a very connected style of play, in which they are the team that takes the least amount of time to reach 50 passes, highest “advance ratio” (i.e., the team that plays the most horizontal to the opponent’s goal) , instead of going wide with their players, and the shortest-path average is very low meaning the team moves the ball between players at a very fast pace.
Conclusion
In conclusion, network science showed that the FC Barcelona team under Pep Guardiola, known most for its passing plays, indeed had very good passing networks as their central system of play, and the data from the research supports this premonition people have regarding this team with solid numbers using network science. This study shows that network science can extract significant conclusions, though not perfect by any means, that can analyze quite a complex problem specifically in the world of soccer or sports in general.
Sources
Buldú, J.M., Busquets, J., Echegoyen, I. et al. Defining a historic football team: Using Network Science to analyze Guardiola’s F.C. Barcelona. Sci Rep 9, 13602 (2019). https://doi.org/10.1038/s41598-019-49969-2