Sim2Win: Team-Agnostic Football Prediction System
A new AI system called Sim2Win predicts football match outcomes without using team names or identity features, enabling generalization to unseen teams. Developed using StatsBomb open event data from 11 competitions, 178 teams, and 1,411 team-match records, the framework constructs five-match rolling tactical profiles and engineers four interpretable tactical feature ratios. It clusters team behaviors into eight playstyles via K-Means and trains 13 classifiers to estimate win, draw, and loss probabilities. The system reframes match outcome prediction as a tactical decision-support problem, operating without subjective expert analysis or identity-based scouting.
Key facts
- Sim2Win is a team-agnostic, event-based pre-match tactical recommendation framework
- Uses StatsBomb open event data from 11 competitions, 178 teams, and 1,411 team-match records
- Constructs five-match rolling tactical profiles
- Engineers four interpretable tactical feature ratios
- Clusters team behaviors into eight playstyles via K-Means
- Trains 13 classifiers to estimate win, draw, and loss probabilities
- Operates without team names or identity features
- Reframes match outcome prediction as a tactical decision-support problem
Entities
Institutions
- StatsBomb
- arXiv