Building a Player-Focused Data Team
summary
Describes concrete systems: player-level models for matchmaking and personalization, a champion recommendation engine, positional-play detection and segmentation, and monitoring for frequent biweekly patches. Uses Python, PySpark, R, Databricks, AWS and Tableau and snapshots/version control for models and data. Claims cover scale and impact with 100 million monthly players and 500 billion data points per day and 140+ champions, stresses separate regional models, mixed central and embedded teams, automated data quality checks, and production deployments that serve real-time play.
AI-written summary · about
chapters (8)
- 0:00Welcome
- 1:10Robotics to data science
- 4:18Riot Games and League
- 8:55Player-focused decisions
- 13:58Data science use cases
- 19:57Balanced data team
- 23:10Future data priorities
- 24:35Q&A
related talks
03mThe Ludic Potentials of Player Data
16mGame Data for All: How AI Is Reinventing Game Analytics
41mEnhancing Event Analysis at Scale: Leveraging Tracking Data in Sports
17mWhat Moves Players? Visual Data Exploration of Twitter and Gameplay Data
48mBecoming a Data Head, a reflection on the trade offs and decision
55mBuilding The Data Dream Team
38mBuilding Effective Data Teams: Strategies for Success
19mData & Analytics; Shaping the Future of Gaming and Maximizing ROI