Soccer is one of the most popular sports in the world and is increasingly welcoming data analysis and quantitative approaches to search for advantages in the soccer pitch. In this direction, the analysis of different playing styles is promising. Accordingly, this paper proposes a non-traditional application of clustering techniques on data from soccer matches in worldwide championships. It uses the data to describe how each team behaves in terms of defense, progressions, laterality, and efficiency in different matches. The approach revealed distinct and recurring playing styles which helped to understand how the teams match with one another, and their chances of obtaining positive results depending on the combination of their playing styles. Our experimental results pointed to the importance of the efficiency in the fundamentals of the sport and show that this approach can bring benefits in terms of team strategy.