Home // DATA ANALYTICS 2015, The Fourth International Conference on Data Analytics // View article

League Adjusted Salary Model using Local Polynomial Regression

Authors:
Shinwoo Kang

Keywords: Local Polynomial Regression

Abstract:
Since the 2012 National Hockey League (NHL) Lockout, there have been many economic trends in the league that one might argue inconsistent. While many players’ salaries were significantly altered as results of buy-outs or extravagant contract signings, the salary cap has fluctuated dramatically in the following years due to these chaotic activities. To understand the seemingly contradicting NHL economic trends, in this paper, we discuss League Adjusted Salary Model (LASM) applying Local Polynomial Regression Modeling to properly gauge a player’s monetary vs. production feasibility value. The League Adjusted Salary Model is a approach that is dependent on a player’s League-Relative Salary Percentages and his Individual Production. The League Relativity is emphasized to account for the different payrolls of all 30 NHL teams and to understand the year-by-year economic trend. The Individual Production is a user flexible element of the individual level model that can be improved with utilizations of “Enhanced Statistics” such as Unblocked Shot Attempt Relative Percentage values. Combining these two data sets, we apply the Local Polynomial Regression Modeling to compute the feasibility of cost and production.

Pages: 59 to 62

Copyright: Copyright (c) IARIA, 2015

Publication date: July 19, 2015

Published in: conference

ISSN: 2308-4464

ISBN: 978-1-61208-423-7

Location: Nice, France

Dates: from July 19, 2015 to July 24, 2015