The study tests the relative importance of different player metrics in determining an NBA player’s salary. Using each year’s data across 2006-2025 of every league athlete, a dataset comprising 8,798 unique player rows and 26 distinct variables was compiled. Multi-variable linear and polynomial regression models were constructed using Ordinary Least Squares (OLS) estimation after checking for multicollinearity. Points Scored per game and Age turned out to be the most dominant factors in the polynomial regression model which had the highest adjusted R² of 0.72. Game’s evolution and salary trends were also analysed from perspective of a player’s playing position and his lifecycle. Over the study period, an NBA game has clearly turned more offensive and higher scoring due to sharp rise in 3-Point shots leading to greater value in the team for forward players. A player’s golden age for earning is between the age of 30-34 when he earns the maximum value per point scored. Average player salary growth has also beaten inflation by a wide margin in the study period.
This study identifies the critical game statistics that determine the regular-season win-loss percentage of National Basketball Association (NBA) teams. Utilizing box-score statistics from the past 20 years, a dataset comprising 600 team-season observations and 25 distinct variables was compiled. Multiple linear regression models were constructed using Ordinary Least Squares (OLS) estimation to evaluate predictive power and forecasting accuracy. In the concurrent descriptive model, point differential (PD), effective field goal percentage (eFG%), and turnover percentage (TO%) emerged as the most critical statistics, accurately explaining team performance (Adjusted R² > 0.92). However, predictive accuracy dropped significantly when forecasting a team’s future win-loss percentage using prior-season data (Adjusted R² = 0.44), with a substantially higher standard error. This inherent predictive volatility underscores the league's competitive balance, contributing to rising viewership, escalating league revenues, and growing team payrolls over the past decade. Finally, individual historical team performance trends are evaluated alongside a novel 'Return on Salary' metric, revealing diminishing value of salary as a variable to improve team performance.
The literature review paper highlights the impact on socio-economic development parameters under different economic models (Capitalism, Socialism and hybrid models like China). Capitalism leads to greater focus on innovation and a self-perpetrating cycle of growth in income levels, wealth and quality of life, eventually culminating into overall improved country happiness scores as visible in studies. Socialism, on the other hand, leads to inefficient resource allocation and wealth inequality which is against its basic principle. However, the side-effects of capitalism cannot be ignored and thus hybrid development models could provide the desired output. The paper highlights various alternate growth models which are gaining prominence.
