Abstract

Determining the constellation of football players determines a team's success when competing on the field. Disassembling players is an option that must be made considering performance history and costs. This research experiments with K-Means to automate the search for groups of players based on performance and price history. Grouping can achieve a constellation of players with high-performance characteristics but at an affordable price. The dataset used in this research is 580 football players for the 2022/2023 season from Sofifa, Fbref, and SofaScore. The data is divided into four player positions: goalkeeper, defender, midfielder, and attacker. Data for each position is grouped into 3 clusters. Each cluster is analyzed to obtain dominant performance indicator values and determine the characteristics of the cluster. Experimental results using K-Means show that cluster 1 is a team with medium player prices but low performance. Cluster 2 has the cheapest price but with the best performance. Meanwhile, cluster 3 is the most expensive but performs similarly to cluster 2.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call