At its core, basketball is the management of possessions.
Every possession begins with control of the ball and moves through a simple chain of events. The offense attempts to produce a scoring attempt. That attempt may produce points, it may miss and continue through a rebound, or the possession may end early through a turnover.
The structure of the game can therefore be expressed as a sequence:
possessions → shots → points
Six-Factor RAPM measures how players influence each stage of this process. Offensive impact and defensive impact can each be decomposed into three mechanisms. Together they form total player impact.
RAPM = oTS + oTOV + oREB + dTS + dTOV + dREB
Each factor captures a different way players change the expected value of possessions while they are on the floor.
oTS — Offensive Scoring Efficiency
This factor measures how much a player increases the points their team generates per scoring attempt.
The primary driver of offensive scoring efficiency is the combination of usage and true shooting percentage. Players who take a large share of their team's attempts and convert them efficiently have the strongest direct impact on this factor.
However, scoring efficiency is not created through shooting alone. Players also influence it through playmaking, ball movement, spacing, and transition offense, all of which affect the quality of shots their teammates receive. Passers who generate open looks, movers who keep the offense flowing, and players who create advantages in transition can raise team scoring efficiency even when they are not the one finishing the play.
Drawing fouls is also critical, since free throws increase the points generated per scoring attempt.
oTOV — Offensive Turnover Control
Before a scoring attempt can occur, the offense must maintain possession of the ball.
Turnovers end possessions without producing a shot. This factor measures how a player influences the rate at which their team loses possessions before reaching the shot phase.
The primary drivers of offensive turnover impact are offensive load and turnover rate. Players who carry a large share of the offense—through scoring attempts, passing volume, and time on the ball—naturally have greater influence on this factor, since turnovers can only occur when a player is directly involved in the play.
Ball security, decision-making, and the ability to handle defensive pressure all contribute here. Players who maintain control of the offense under heavy responsibility allow their team to reach scoring attempts more consistently.
oREB — Offensive Rebounding Impact
Even when a shot misses, the possession may not be finished.
Offensive rebounds extend possessions and create additional scoring attempts that would not otherwise exist. Players influence this factor by securing rebounds themselves and by creating opportunities through positioning, timing, and physical presence near the basket.
Second-chance opportunities increase the total number of attempts a team has to convert possessions into points.
dTS — Defensive Scoring Efficiency Suppression
This factor measures how much a player lowers the points opponents generate per scoring attempt.
Defenders influence scoring efficiency through several mechanisms. Contesting shots, navigating screens, and containing isolation attacks reduce shot quality on the perimeter. At the rim, defenders protect the basket through positioning, verticality, and shot blocking.
Interior defenders also create rim deterrence, forcing offenses to avoid high-value attempts at the basket altogether. Foul discipline is also important. Because scoring efficiency includes free throws, defenders who avoid unnecessary fouls—or commit smart fouls that prevent easy layups—can reduce the points opponents generate per attempt.
Together, these actions lower the efficiency of opponent scoring attempts.
dTOV — Defensive Turnover Creation
Turnovers immediately end possessions before a scoring attempt can occur.
Defenders generate turnovers through steals, deflections, and disruptive pressure on the ball and in passing lanes. Pressure can also force rushed decisions that lead to mistakes by the offense.
Another important source of defensive turnovers comes from offensive fouls drawn, including charges and moving screen violations. Players who position themselves effectively to draw these calls can end possessions without allowing a shot.
By forcing turnovers, defenders prevent scoring attempts from occurring in the first place.
dREB — Defensive Rebounding Control
Even strong defensive possessions remain incomplete if the offense secures the rebound.
Defensive rebounding ends the possession by securing the miss and preventing second-chance opportunities. Players influence this factor by grabbing rebounds themselves and by boxing out to allow teammates to collect the ball.
Strong defensive rebounding ensures that missed shots actually end the opponent's possession.
The Full Structure of Impact
The entire game can be decomposed into three mechanisms applied on both offense and defense:
Shot Value: oTS · dTS
Possession Survival: oTOV · dTOV
Possession Extension: oREB · dREB
Six-Factor RAPM estimates how much each player shifts these components relative to league average while they are on the floor. Instead of collapsing player value into a single number, the model reveals where that value is created within the lifecycle of a possession—through improving scoring efficiency, protecting or ending possessions, and controlling what happens when shots miss.
NBA advanced impact metrics are designed to quantify a player's overall contribution to their team's success. These metrics go beyond traditional box score statistics to capture the nuanced ways a player influences the game, attempting to measure aspects of performance that may not be immediately apparent.
Impact metrics aim to answer questions such as: How many more points does a team score (or prevent) with a particular player on the court? How does a player's presence affect the overall efficiency of their team's offense or defense? These metrics are particularly valuable for identifying players who contribute significantly to winning in ways that might not be reflected in traditional statistics.
RAPM is a foundational impact metric in basketball analytics that estimates a player's contribution in points per 100 possessions. It considers both offensive and defensive impact and is designed to be relatively unbiased regarding playstyle.
RAPM is particularly noteworthy for its ability to capture the value of players who contribute to winning in ways that aren't always reflected in traditional statistics or even other advanced metrics. It does this by focusing solely on how a player impacts the scoreboard, regardless of their individual stat accumulation. RAPM ignores the boxscore.
RAPM uses ridge regression, a statistical technique that helps address the "multicollinearity" problem arising from players frequently playing together. It analyzes how the score changes when players enter or leave the game, while accounting for the quality of teammates and opponents.
Time Decay RAPM: This variation applies an exponential decay function to a player's entire history, giving more weight to recent performance. It's particularly useful for capturing a player's current level while still considering historical data.
3 Year Recency RAPM: This approach uses the last 3 years of data, with heavier weight on more recent possessions. It balances capturing established performance levels with current trajectory.
Lifetime RAPM: A comprehensive 28-year calculation incorporating an age curve to account for typical player development and decline. It's valuable for historical player comparisons and understanding career arcs.
EPM is an advanced impact metric that combines box score data with play-by-play data to estimate a player's impact per 100 possessions. It aims to provide a more stable and predictive metric than pure RAPM while still capturing the depth of a player's impact.
EPM was developed to address some of the limitations of pure RAPM while maintaining its strengths. It incorporates individual player actions to provide a more granular view of player impact.
EPM uses a two-step process:
For more detailed information and current EPM rankings, visit Dunks and Threes.
DARKO is an advanced player projection system that uses machine learning techniques to estimate player impact. Unlike many other metrics, DARKO updates its projections daily, providing a real-time view of player performance and potential.
DARKO uses a Kalman filter, a statistical technique often used in time series analysis and control systems. This allows the model to update its estimates of player skill in real-time as new information becomes available.
DARKO is particularly useful for player evaluation, team construction, identifying undervalued players, and projecting player development. Its daily updates make it valuable for applications like daily fantasy sports.
For up-to-date DARKO projections and more detailed explanations of the methodology, visit DARKO.app.
LEBRON is a comprehensive impact metric that aims to capture a player's total contribution to winning. It combines box score data with on/off impact data and incorporates various adjustments to provide a nuanced view of player value.
LEBRON uses a multi-step process:
LEBRON values are expressed in points per 100 possessions above average, with 0 representing an average player. The metric also includes Wins Added calculations to estimate a player's total impact over a season.
For a more detailed explanation of LEBRON methodology and current player ratings, visit the BBall Index LEBRON Introduction.
DFG metrics are advanced defensive statistics that measure a player's impact on opponent shooting and rim protection. These metrics provide insights into a player's defensive effectiveness beyond traditional box score stats.
For most of these metrics, negative values are better as they indicate that the player's presence reduces opponent shooting efficiency or frequency of attempts at the rim. The contest metrics (per 100) are exceptions, where higher values indicate more active defensive engagement.
Passing metrics provide insights into a player's playmaking abilities and efficiency in distributing the ball.
Higher values are generally better for most metrics, except BadPass% where lower values indicate better ball security. These metrics together provide a comprehensive view of a player's passing ability, frequency, and efficiency.
Defensive playmaking metrics measure a player's ability to disrupt the opponent's offense and create turnovers.
Higher values are generally better for all these metrics. STOP% is a comprehensive metric that captures overall defensive playmaking ability. It represents the percentage of possessions where the player directly causes a change of possession through steals, drawn offensive fouls, or recovered blocks.
The Four Factors are key components that contribute to a team's success. They are:
These factors are calculated using on/off court data. By comparing how these factors change when a player is on the court versus off, we can estimate their career impact in each area on both ends of the court.
Importantly, a player's Four Factors can predict their non-age adjusted RAPM with a 0.95 r^2. This means the Four Factors are extremely valuable in explaining why a player may have a high or low RAPM.
Read these statistics as percentage increases or decreases to the team's four factors Effective Field Goal %, Turnover %, Offensive Rebound %, Free Throw rate These RA-four factors reveal the drivers behind a player's RAPM Positive numbers indicate positive impact in that area.
I've carefully chosen these stats to provide insight into a player's passing impact. Here's what they mean:
28-Year Lifetime RAPM is a stat that measures how much a player impacts winning over their entire career, relative to their age, using data from 1997 to 2024.
The Four Factors are key components that contribute to a team's success. They are:
These factors are calculated using on/off court data. By comparing how these factors change when a player is on the court versus off, we can estimate their career impact in each area on both ends of the court.
Importantly, a player's Four Factors can predict their non-age adjusted RAPM with a 0.95 r^2. This means the Four Factors are extremely valuable in explaining why a player may have a high or low RAPM.
By considering both the overall RAPM and the Four Factors, you can get a more complete picture of how a player contributes to their team's success over their career, taking into account their performance relative to their age and specific areas of impact.
This section explores the intriguing relationship between two metrics that measure a player's impact on opponent turnovers:
A measure of a player's ability to disrupt opposing offenses, representing the number of turnovers they force per 100 possessions above or below the league average. This metric is derived from box score statistics, specifically combining steals and offensive fouls drawn.
The impact on opponent turnover rate derived from RAPM (Regularized Adjusted Plus-Minus) analysis of lineup data, independent of box score statistics.
By comparing these metrics, we gain insight into the value of lineup-based analysis versus traditional box score statistics. This comparison reveals potential hidden contributions to forcing turnovers that may not be captured by steals and drawn fouls alone. It's worth noting that while RA-DTOV may capture indirect ways of forcing turnovers, it is also subject to the inherent variability in RAPM and lineup analysis. Nevertheless, it shows how well lineup-based analyses can largely agree with what the boxscore based analysis would suggest.
The significance of this analysis is underscored by the substantial impact of turnovers on game outcomes. In terms of plus-minus value, forcing an additional turnover per 100 possessions is roughly equivalent to +1 points per 100 in value, representing a crucial aspect of a player's defensive impact that should not be overlooked.
Note: For players with over 10,000 defensive possessions, rFTOV predicts RA-DTOV with an R² of 0.70, demonstrating a strong correlation between box score-derived and lineup-based turnover impact metrics. This predictive power increases with larger sample sizes, reaching an R² of 0.77 for players with over 40,000 defensive possessions, despite potential underestimation due to mismatched time periods in the datasets.
No historical data available for this dataset.