Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

PandaSkill - Player Performance and Skill Rating in Esports: Application to League of Legends

Maxime De Bois, Flora Parmentier, Raphaël Puget, Matthew Tanti, Jordan Peltier

TL;DR

PandaSkill tackles the challenges of evaluating player performance and skill in esports by first deriving a Performance Score (PScore) from per‑game statistics using role‑specific, calibrated ML models tied to win probability, then updating player skills within a Bayesian OpenSkill framework in a free‑for‑all setting. A dual contextual/meta rating system mitigates cross‑region rating isolation, enabling global comparisons. Across five years of professional League of Legends data, PandaSkill's PScore and OpenSkill variants outperform traditional approaches in outcome forecasting and align more closely with expert judgments, while providing interpretable results via SHAP analyses. The work advances esports analytics by delivering a model‑agnostic performance measure, region‑aware skill estimation, and publicly available tools for broader adoption and further research.

Abstract

To take the esports scene to the next level, we introduce PandaSkill, a framework for assessing player performance and skill rating. Traditional rating systems like Elo and TrueSkill often overlook individual contributions and face challenges in professional esports due to limited game data and fragmented competitive scenes. PandaSkill leverages machine learning to estimate in-game player performance from individual player statistics. Each in-game role is modeled independently, ensuring a fair comparison between them. Then, using these performance scores, PandaSkill updates the player skill ratings using the Bayesian framework OpenSkill in a free-for-all setting. In this setting, skill ratings are updated solely based on performance scores rather than game outcomes, hightlighting individual contributions. To address the challenge of isolated rating pools that hinder cross-regional comparisons, PandaSkill introduces a dual-rating system that combines players' regional ratings with a meta-rating representing each region's overall skill level. Applying PandaSkill to five years of professional League of Legends matches worldwide, we show that our method produces skill ratings that better predict game outcomes and align more closely with expert opinions compared to existing methods.

PandaSkill - Player Performance and Skill Rating in Esports: Application to League of Legends

TL;DR

PandaSkill tackles the challenges of evaluating player performance and skill in esports by first deriving a Performance Score (PScore) from per‑game statistics using role‑specific, calibrated ML models tied to win probability, then updating player skills within a Bayesian OpenSkill framework in a free‑for‑all setting. A dual contextual/meta rating system mitigates cross‑region rating isolation, enabling global comparisons. Across five years of professional League of Legends data, PandaSkill's PScore and OpenSkill variants outperform traditional approaches in outcome forecasting and align more closely with expert judgments, while providing interpretable results via SHAP analyses. The work advances esports analytics by delivering a model‑agnostic performance measure, region‑aware skill estimation, and publicly available tools for broader adoption and further research.

Abstract

To take the esports scene to the next level, we introduce PandaSkill, a framework for assessing player performance and skill rating. Traditional rating systems like Elo and TrueSkill often overlook individual contributions and face challenges in professional esports due to limited game data and fragmented competitive scenes. PandaSkill leverages machine learning to estimate in-game player performance from individual player statistics. Each in-game role is modeled independently, ensuring a fair comparison between them. Then, using these performance scores, PandaSkill updates the player skill ratings using the Bayesian framework OpenSkill in a free-for-all setting. In this setting, skill ratings are updated solely based on performance scores rather than game outcomes, hightlighting individual contributions. To address the challenge of isolated rating pools that hinder cross-regional comparisons, PandaSkill introduces a dual-rating system that combines players' regional ratings with a meta-rating representing each region's overall skill level. Applying PandaSkill to five years of professional League of Legends matches worldwide, we show that our method produces skill ratings that better predict game outcomes and align more closely with expert opinions compared to existing methods.
Paper Structure (33 sections, 6 equations, 5 figures, 6 tables)

This paper contains 33 sections, 6 equations, 5 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Player Performance in a Game
  4. Player Skill Rating
  5. Position of Our Work
  6. Background on League of Legends
  7. Game Layout
  8. Players
  9. General Game Strategy
  10. The Esports Ecosystem
  11. PandaSkill
  12. Measuring the Performance Score of Players in a Game
  13. Estimating the Skill Rating of the Players
  14. OpenSkill
  15. Free-For-All (FFA)
  16. ...and 18 more sections

Figures (5)

  • Figure 1: Annotated minimap of League of Legends.
  • Figure 2: Calibration plots with histograms of predicted probabilities for each role.
  • Figure 3: Relative importance of features for each role-based XGBoost model using SHAP values. The features are ordered in descending order based on their average importance across roles.
  • Figure 4: Distribution of players' skill ratings per region using the Meta_FFA_OpenSkill rating model with PScore.
  • Figure 5: Majority and unanimity concordance between models and experts per region.