Calculating Basketball Bpm: A Guide To Understanding Player Performance

how to calculate bpm basketball

Box Plus/Minus (BPM) is a metric in basketball analytics that provides a snapshot of a player's overall impact on the game relative to the league average. Created by Dan Myers, it aims to capture a player's performance by quantifying how much better or worse a team performs with that player on the court. BPM is calculated using box score stats from an individual and team level, factoring in elements like minutes played, rebounds, assists, and the player's position and role estimates. The result is a raw BPM, which is then adjusted at the team level to account for the team's shooting context and overall performance. BPM has evolved over time, with versions like BPM 2.0 incorporating more play-by-play data to enhance accuracy, particularly on the defensive side.

Characteristics Values
Creator Dan Myers
Purpose To capture a player's overall impact on the game, given the lack of defensive statistics in traditional box scores
Calculation Sum of different player's contributions to obtain a raw value, which is then calibrated on team performance to obtain the real BPM
Components Offensive Box Plus Minus (OBPM) and Defensive Box Plus Minus (DBPM)
OBPM Calculation Points differential (positive or negative) spread over 100 possessions
DBPM Calculation Total BPM – Offensive BPM
Regression Coefficients Vary linearly between position 1 and position 5
Team Adjustment Added to the raw BPM of all players on the team so that the team's total sums to the team adjusted efficiency
Relevance Provides valuable insights for coaches, analysts, and fans

shunwild

Calculating a player's raw BPM

Box Plus/Minus (BPM) is a basketball box score-based metric that estimates a basketball player's contribution to the team when that player is on the court. It is based on the traditional basketball box score—no play-by-play data or non-traditional box score data (like dunks or deflections) are included. BPM uses a player’s box score information, position, and the team’s overall performance to estimate the player’s contribution in points above league average per 100 possessions played.

To calculate a player's raw BPM, you need to determine their position and offensive role, generate position-specific coefficients, and adjust for team shooting context.

First, determine the player's position and offensive role. This can be done by using the entire season's worth of data and calculating the player's estimated position and offensive role. For example, a player's position regression may estimate them to be position 2.3 with an offensive role of 1.0 (pure creator).

Next, generate the player's coefficients to be used in the raw BPM calculation. Some of the coefficients will vary based on the player's position and offensive role, while others will not.

Then, adjust the points scored by the players on the team to account for the team shooting context. Calculate the team-average points per adjusted shot attempt and compare this to the baseline points per adjusted shot attempt used by the regression. Adjust the points scored by the players on the team up or down by adding or subtracting a constant points per adjusted shot attempt.

Finally, calculate the raw BPM for the player by summing up the adjusted points scored by the player.

It is important to note that BPM has some limitations. It relies on box score data, which may not capture all critical defensive elements like positioning and communication. Additionally, the stat can be influenced by "statistical noise" as it uses team performance as a baseline but applies it to individual players.

shunwild

Offensive and defensive BPM

Box Plus/Minus (BPM) is a basketball metric that measures a player's overall impact on the game relative to the league average. It was created by Dan Myers and prioritises historical usefulness over improved accuracy.

To calculate BPM, the first step is to generate the player's coefficients based on their position and role estimates. The next step is to calculate the team-average points per adjusted shot attempt and compare this to the baseline points per adjusted shot attempt used by the regression. The points scored by the players on the team are then adjusted up or down by adding a constant per adjusted shot attempt to account for the team shooting context. The raw BPM for the player can then be calculated. To get the total BPM, the raw BPMs for the team are summed up (weighted by the percentage of minutes played) and compared to the team's adjusted efficiency per 100 possessions. Finally, a team adjustment constant is added to the raw BPM of all the players so that the team's total sums to the team adjusted efficiency.

BPM can be split into offensive and defensive components. The defensive BPM is calculated by subtracting the offensive BPM from the total BPM. The offensive and defensive BPM use the same variables as the full BPM but with different coefficients. The coefficients vary linearly between position 1 and position 5. For example, if a player is a small forward (position 3), the coefficients halfway between the first and second columns are used.

While BPM is good at measuring offence, defensive numbers should be taken as a guide and can be discounted if a player is well known as a good or bad defender. BPM also has limitations due to the choice to limit it to stats that were tracked as of the 1973-74 season.

shunwild

BPM and historical player data

Box Plus/Minus, Version 2.0 (BPM) is a basketball box score-based metric that estimates a basketball player’s contribution to the team when that player is on the court. It was created by Dan Myers and prioritises historical utility over accuracy. BPM is calculated using a player’s box score information, position, and the team’s overall performance to estimate the player’s contribution in points above league average per 100 possessions played.

BPM does not consider playing time, focusing instead on a player's rate. League average is defined as 0.0, meaning 0 points above average or below average. Above-average players play more minutes, resulting in a greater number of below-average players in the league at any time. BPM also assumes that every player on a team has contributed equally, and if the team is good, all players are assumed to be equally good.

To calculate BPM, the first step is to generate the player’s coefficients to be used in the raw BPM calculation. These coefficients are based on the player’s position and offensive role, and some vary while others do not. The team-average points per adjusted shot attempt are then calculated and compared to the baseline points per adjusted shot attempt used by the regression. The points scored by the players on the team are adjusted up or down by adding a constant per adjusted shot attempt to account for the team shooting context.

The raw BPM for the player can then be calculated. The raw BPMs for the team are then summed up (weighted by the percentage of minutes played) and compared to the team-adjusted efficiency per 100 possessions. Finally, a constant ("team adjustment") is added to the raw BPM of all players on the team so that the team's total sum matches the team-adjusted efficiency.

BPM can also be split into offensive and defensive components using the same regression style as the full BPM calculation, with different coefficients. Offensive BPM is calculated, and defensive BPM is then obtained by subtracting offensive BPM from total BPM.

shunwild

BPM and player coefficients

Box Plus/Minus (BPM) in basketball is an advanced metric that evaluates a player’s overall impact on the game by estimating their contribution to the team’s point differential per 100 possessions. It is a valuable tool for estimating a player’s performance relative to the average and can be used to inform sports betting decisions.

To calculate BPM, you need player box score stats per 100 possessions, team-adjusted efficiency per 100 possessions, and estimated player positions. The process involves determining a player’s position and offensive role, generating position-specific coefficients, adjusting for team shooting context, and applying a team adjustment to align with team-adjusted efficiency.

The coefficients used in the calculation vary based on the player’s position and offensive role. For example, if a player is in position 1, the coefficients for position 1 are used. If a player is in position 3 (small forward), then coefficients halfway between the first and second columns are used, as position 3 is halfway between position 1 and position 5.

The regression coefficients were developed to maximize the fit for both offense and defense concurrently. Offensive BPM is outputted directly by the regression, and defensive BPM is calculated as Total BPM minus Offensive BPM.

Some specific stats, such as MPG, ORB%, DRB%, STL%, BLK%, and AST%, are multiplied by their corresponding coefficients to add to the player’s overall value. The interaction term between rebounding and assists, sqrt(AST%*TRB%), is also included and helps the overall fit of the regression.

In summary, BPM in basketball is a valuable metric for evaluating player performance, but it has limitations, such as its reliance on box score data and the influence of "statistical noise" due to the use of team performance as a baseline for individual players.

shunwild

BPM and team adjustments

Box Plus/Minus (BPM) is an advanced metric in basketball that evaluates a player’s overall impact on the game by estimating their contribution to the team’s point differential per 100 possessions. BPM was created by Dan Myers and is closely akin to a stat called Statistical Plus/Minus (SPM). Myers aimed to create a solid metric that could capture a player’s overall impact on the game, given the lack of defensive statistics in traditional box scores. BPM is calculated using player box score stats per 100 possessions, team-adjusted efficiency per 100 possessions, and estimated player positions.

To calculate BPM, you must first determine a player’s position and offensive role, then generate position-specific coefficients, and adjust for team shooting context. The coefficients are developed based on the player's position and offensive role estimates. The regression coefficients are developed to maximize the fit for both offense and defense concurrently. The offensive role adjustment is much smaller for offensive BPM than overall, indicating that much of the value not captured in the box score is on defense.

After calculating the raw BPM for the player, the next step is to add a "team adjustment" to align with team-adjusted efficiency. This involves summing up the raw BPMs for the team (weighted by % of minutes played) and comparing this to the team-adjusted efficiency per 100 possessions. A constant is then added to the raw BPM of all the players on the team so that the team's total sums to the team-adjusted efficiency. This team adjustment is critical as it allows the regression to assign credit to other players on the floor besides the player who actually posted the stat.

While BPM in basketball is a valuable metric for evaluating player performance, it has some limitations. One issue is its reliance on box score data, which may fail to capture critical defensive elements like positioning and communication. Additionally, the stat can be influenced by "statistical noise" as it uses team performance as a baseline but applies it to individual players. Despite these shortcomings, BPM remains a valuable tool for estimating a player’s performance relative to the average and understanding their contributions in a detailed and nuanced way.

Frequently asked questions

Written by
Reviewed by

Explore related products

Share this post
Print
Did this article help you?

Leave a comment