Basketball, in its traditional conception, has categorized players into five well-defined positions: point guard, shooting guard, small forward, power forward, and center. This classification, based mainly on height and general on-court responsibilities, has served for decades as a framework for understanding a team's structure. However, the game evolves.
Traditionally, during a basketball player's development, they would specialize based on their physical size. The taller and heavier they were, the more likely they were to play near the basket. Consequently, the specific skills they were taught were of a particular type (post-up game, mid-range shooting, strength to gain position, etc.). This has been changing in recent years, as development now often focuses on the "innate" technical characteristics of the players, allowing us to enjoy players who are 2.24 meters tall playing facing the basket and shooting three-pointers (highlights of Victor Wembanyama).
This change necessitates a different approach. For this reason, in recent years, advanced analytics and a deeper understanding of performance have driven a paradigm shift: from rigid "positions" to fluid "roles." This new approach focuses on the technical and physical characteristics a player can bring to the team, regardless of their traditional positional label.
Among the various proposals, the National Basketball Association (NBA) itself has made its own suggestions based on the knowledge generated in recent years, identifying roles such as scorers, playmakers, players with skilled hands, athletic players, or defenders, among others [1].
Extraído de Fadeaway World.
Extraído de Fadeaway World.
Extraído de Fadeaway World.
More than a decade ago, some authors highlighted the need to approach the study of positions from a different perspective. Studies like the one by Lum et al. (2013) were the first to import principal component and cluster analysis to derive performance profiles in the NBA [2]. More recently, Kalman and Bosch (2020), using data from 10 NBA seasons (NBA.com/stats), proposed redefining basketball positions based on cluster analysis of player statistics, identifying up to nine distinct roles [3]. These roles were: high-usage guards, stretch forwards, three-point shooting guards, traditional centers, versatile role players, floor generals, mid-range bigs, skilled forwards, and ball-dominant scorers.
<aside> 🔖
Consulta el glosario de la estadística de la NBA aquí: NBA Glossay
</aside>
Kalman y Bosch (2020). NBA lineup analysis on clustered player tendencies: a new approach to the positions of basketball & modeling lineup efficiency of soft lineup aggregates.
A recent study by Çene et al. (2024) in the EuroLeague delves deeper into this line of work, using multivariate statistical analysis to differentiate roles among players [4], emphasizing that focusing solely on position hides the real diversity of functions on the court.
Çene et al. (2024) How do European and non-European players differ: Evidence from EuroLeague basketball with multivariate statistical analysis.
While statistics in sports provide an incomplete picture by not capturing many nuances of the game, they do allow us to understand certain events. Therefore, it is important to note that they should be studied with great caution and without the intention of providing absolute answers to our questions. That being said, the ideas proposed in this publication focus on describing the characteristics of these new roles and generating a practical tool to help identify what type of player we are dealing with.
If you are familiar with basketball, you will quickly realize that a role-based classification offers clear advantages. Some of these are: