NTDevLab / KHL
KHL Prediction — KHL match predictions and analytics
An independent ML model for Kontinental Hockey League matches: outcome probabilities, period totals, an open prediction history and season accuracy reports.
- Python
- Flask
- ML
- PostgreSQL
- Active project
- Interface languages: RU
The problem it solves
KHL match outcomes depend on many factors — form, schedule, lineups. KHL Prediction converts season statistics into probability estimates for each game.
Data and ML
The model is trained on season statistics: results, period scores, team rankings. Accuracy is validated at the end of each season.
The NTDevLab approach
The prediction is published before the event and is never rewritten afterwards: the open history keeps both hits and misses, so model accuracy can be checked independently.