League of Legends (LoL) has been a dominant esport for a decade, yet the inherent complexity of the game has stymied the creation of analytical measures of player skill and performance. Current industry standards are limited to easy-to-procure individual player statistics that are incomplete and lacking context as they do not take into account teamplay or game state. We present a unified performance model for League of Legends which blends together measures of a player's contribution within the context of their team, insights from traditional sports metrics such as the Plus-Minus model, and the intricacies of LoL as a complex team invasion sport. Using hierarchical Bayesian models, we outline the use of gold and damage dealt as a measure of skill, detailing players' impact on their own-, their allies'- and their enemies' statistics throughout the course of the game. Our results showcase the model's increased efficacy in separating professional players when compared to a Plus-Minus model and to current esports industry standards, while metric quality is rigorously assessed for discrimination, independence, and stability. Readers might also find additional qualitative analytics which explore champion proficiency and the impact of collaborative team-play. Future work is proposed to refine and expand the SIDO performance model, offering a comprehensive framework for esports analytics in team performance management, scouting and research realms.

翻译：《英雄联盟》（LoL）作为统治电竞领域十年的竞技项目，其游戏内在复杂性阻碍了玩家技能与表现分析指标的建立。当前行业标准局限于易于获取的个体玩家统计指标，这些指标既不完整又缺乏情境性——既未纳入团队协作因素，也未考虑游戏实时状态。我们提出了一种统一的英雄联盟表现模型，该模型融合了玩家在团队语境中的贡献度量、传统体育指标（如正负值模型）的洞察，以及LoL作为复杂团队入侵类运动的特性。通过分层贝叶斯模型，我们论证了将金币与造成的伤害作为技能衡量指标的方法，详细解析了玩家在比赛进程中对自身、队友及敌方统计数据的多重影响。结果表明，与传统正负值模型及当前电竞行业标准相比，本模型在区分职业选手方面展现出更优效能。我们采用区分度、独立性和稳定性三大指标严格评估了模型质量，同时提供了额外的定性分析以探索英雄熟练度与团队协作效能。未来研究方向将致力于完善并扩展SIDO表现模型，为团队绩效管理、选手发掘及学术研究领域的电竞分析构建综合性框架。