On the Issue of Analyzing and Optimizing the Salary Budget of Professional Hockey Clubs

Abstract

The purpose of the study. The aim of the study is to use the methods of correlation-regression analysis as a tool to determine the fair share of a professional hockey player’s salary in the total salary budget of a hockey club. In modern conditions of functioning of professional hockey clubs, in which the total costs of the players’ salaries are strictly regulated by the “salary cap”, the availability of a tool that allows an objective assessment of the player’s contribution to the overall team result can increase the effectiveness of management decisions. In this paper, a regression model is proposed that allows us to determine the impact of individual characteristics of players, using the example of team defensemen, on the share of their wages in the salary cap.Materials and methods. To select a multiple regression model, the statistical indicators of the defensemen of the National Hockey League in the 2018-2019 season were analyzed. At the substantive stage to design a model, a list of those statistical indicators was determined that, in our opinion, allow us to conclude above that a particular player is useful for the team, and also meet the requirements for factors to be included in the multiple linear regression model (quantitative assessment, close relationship with the result, lack of multicollinearity). In order to obtain the highest quality regression model, a posteriori approach was used in the selection of factors that should be included in the final version of the regression model. As a result of the step-by-step selection of factors, the factors were excluded from the model, t-statistics’ values of which made it possible to draw a conclusion on their statistical insignificance.Results. As a result, a statistically significant model was obtained that describes the dependence of the share of wages in the salary cap of the 2018-2019 season. Comparing the salary shares predicted using the obtained model and their actual values, it was possible to determine the most overrated and underestimated defensemen of the National Hockey League in the 2018-2019 season.Conclusion. The proposed regression model is an example of how econometric methods combined with hockey statistics allow us to quantify the pricing patterns of a professional hockey player’s contract. In our opinion, the obtained model of multiple linear regression is an affordable tool that allows us to give an adequate assessment of the value of a professional hockey player’s contract and help in solving one of the most urgent tasks in sports management – the formation of a competitive team in the presence of a salary cap.