Author: Nikolay Shmakov
Journal: SN Computer Science, 2025
Abstract: Player evaluation is a crucial aspect of many real-world managerial applications. Strict budgetary constraints on annual salary funds in ice hockey intensify competition for high-quality players and narrow down the influence of budget factors on a title pursuit. This forces clubs to carefully assess player contracts and continuously compare them to the transfer market to identify potential reinforcements. This article shows a pipeline of creation of a transparent and verifiable framework for player evaluation based on Expected Goals methodology. The XG model calculates the probability of each shot being a goal by solving a binary probabilistic classification task. The final rating is calculated using the explainable and auditable Expected Goals on Target model (XGoT), which complements the classical Expected Goals model by adding information about points on the gate plane for the shots on goal. This article expands the discussion of the XG methodology application in ice hockey and proposes a new way of player evaluation. The traditional approach of rating creation in hockey has been enhanced by introducing an opposition level that balances the game in equal strengths, power play, and penalty kill to assemble an overall rating. The resulting data-driven analytical tool can be applied to business processes in transfer operations and can provide intriguing insights into player performance levels, stimulating a higher level of maturity in business processes and increasing operational revenue.
Journal: SN Computer Science, 2025
Abstract: Player evaluation is a crucial aspect of many real-world managerial applications. Strict budgetary constraints on annual salary funds in ice hockey intensify competition for high-quality players and narrow down the influence of budget factors on a title pursuit. This forces clubs to carefully assess player contracts and continuously compare them to the transfer market to identify potential reinforcements. This article shows a pipeline of creation of a transparent and verifiable framework for player evaluation based on Expected Goals methodology. The XG model calculates the probability of each shot being a goal by solving a binary probabilistic classification task. The final rating is calculated using the explainable and auditable Expected Goals on Target model (XGoT), which complements the classical Expected Goals model by adding information about points on the gate plane for the shots on goal. This article expands the discussion of the XG methodology application in ice hockey and proposes a new way of player evaluation. The traditional approach of rating creation in hockey has been enhanced by introducing an opposition level that balances the game in equal strengths, power play, and penalty kill to assemble an overall rating. The resulting data-driven analytical tool can be applied to business processes in transfer operations and can provide intriguing insights into player performance levels, stimulating a higher level of maturity in business processes and increasing operational revenue.