Analysing shooting efficiency
I present evidence, based on play-by-play data from the 2009/10 to the 2018/19 season of the National Basketball Association (NBA), that basketball scoring can be modeled by mathematics. I use logistic regression to test hundreds of different in-game scenarios that could affect the likelihood of a player making a basket, such as the amount of time left in the game, the score differential, and the defensive coverage. I demonstrate that certain shots are more efficient than others, and show that the playing tendencies of teams and players have drastically changed in the last 10 years.
Below you can download the full paper. There you will find more shot charts, graphs, as well as a logistic regression study where I mathematically explain why this change happened.
We see a significant increase in points scored per game. My hypothesis is that such an increase in points scored is given by a change in shot selection by teams and players.