This is a review of the basketball analytics research by Justin Kubatko, Dean Oliver, Kevin Pelton, and Dan T. Rosenbaum
Within standard basketball statistics there are a variety of basic variables used. Understanding these variables helps everyone, the knowledgeable and the amateur, better understand the meaning behind the statistics.
The first variable is possession. In basketball a possession starts when a team gains control of the ball and ends when they lose control of the ball, either by scoring or losing possession of the ball to the other team. Analysts look at the number of possessions each team has during the course of the game. Typically, both teams have a fairly equal number of possessions throughout a game.
Possessions are used to evaluate the efficiency of teams and players by looking at the number of points scored per possession. Points scored per 100 possessions is the offensive rating and points allowed per 100 possessions is the defensive rating. A higher offensive rating and lower defensive rating is how teams win games. Offensive and defensive ratings are not related to each other. Teams with high offensive ratings are not necessarily more likely to have a better defensive rating.
Another variable is plays. Plays are similar to possessions except that now rebounds are taken into account. A team can shoot, miss, and rebound the shot many times without losing possession of the ball. The result is multiple plays within a single possession. So, while possessions are family equal between teams during a game, the number of plays is not.
When statistics are calculated on a per-minute basis they tend to be consistent across players even if those players play different amounts of time. These statistics allow players at different levels to be compared with each other.
There are three types of field goal percentages. Field goal percentage (FG%) does not include points gained by three pointers or free throws. Effective field goal percentage (eFG%) includes three pointers and true shooting percentage (TS%) includes both free throws and three pointers.
Rebound rate is determined by looking at the number of shots a player rebounds while they are playing. This can divided into offensive rebounding percentage and defensive rebounding percentage. These are useful tools as players are typically not equally skillful in both offense and defense.
Plus/minus statistics are determined by subtracting defensive points from offensive points while that particular player is in the game. They are determined on a per-minute or per-possession basis. The net plus/minus is found by taking the plus/minus statistic for a given player and subtracting the plus/minus statistic for the team when that player is not in the game. Adjusted plus/minus statistics take into account the skill levels of the player’s teammates and their opponents.
Individual possession rate looks at how many field goal attempts, free throw attempts, assists and turnovers a player has per possession.
Linear weights involve putting weight or importance on different statistics. Linear weights tend to be subjective and consequently are typically not as reliable as other statistics.
These are just a few of the basic statistical ideas used in basketball. Further analysis can be done using the Pythagorean winning percentage which is based on the idea that team winning percentages, points scored, and points allowed are all closely related or the bell curve method which is based on the idea that points scored for and against are normally distributed.
Analysts, teams and fans alike can use these statistics to compare players and teams, determining strengths and weaknesses.
How mature is your team’s analytics program? Take the Sports Analytics Maturity Assessment.
Learn about the Groundbreaking Sports Analytics Model coaches and sports analysts are talking about!
Learn all about Sports Analytics here.