AUG 13, 2026 • NBA

Which NBA Players Are Better Than Their Box Scores Suggest?

Which NBA Players Are Better Than Their Box Scores Suggest?

An original possession-based model for estimating value that traditional counting statistics leave behind.

An original possession-based model for estimating value that traditional counting statistics leave behind.

PLAIN-LANGUAGE GUIDE

Each statistic is explained in context. Rates compare events with opportunities, rankings compare the subject with its peers, and model estimates describe likelihood—not certainty. The article states whether a higher or lower value is better.

The box-score blind spot

The box-score blind spot

Points, rebounds, and assists record visible endpoints. They incompletely describe spacing, screen quality, defensive rotations, advantage creation, possession preservation, and role difficulty. A useful model separates what happened from how much a player changed the probability of a good possession.

Hidden Impact Score

Hidden Impact Score

Define HIS = 0.24Z(advantage creation) + 0.20Z(spacing gravity) + 0.18Z(defensive disruption) + 0.15Z(possession preservation) + 0.13Z(role difficulty) + 0.10Z(lineup lift). Every component is standardized and winsorized at plus or minus 2.5 standard deviations so one extreme feature cannot dominate the score.

Synthetic possession study

Synthetic possession study

Consider 800 generated possessions for a fictional Connector Wing. The player scores only 11.2 points per synthetic game but creates an advantage on 14.8 percent of touches, draws 1.34 defenders on modeled drives, commits a turnover on 7.1 percent of creation attempts, and improves synthetic lineup efficiency by 4.6 points per 100 possessions. The standardized components yield HIS = 0.91, indicating substantially more modeled impact than the scoring total suggests.

Advantage creation

Advantage creation

Credit the first action that forces a defensive rotation, not only the final pass. Advantage Creation Rate = possessions creating a rotation divided by on-ball opportunities. To prevent inflated credit, subsequent actions receive decaying weights of 0.60 and 0.30. A generated sequence worth 1.20 expected points would assign 0.72 to the initiator, 0.36 to the connector, and 0.12 to the finisher context.

Spacing gravity

Spacing gravity

Spacing value can be expressed as the change in teammate driving-lane width and help-defender distance when the player occupies a location. A synthetic gravity score of plus 0.42 means the player increases the standardized quality of teammate driving environments by 0.42 deviations. It does not require the player to take a shot.

Defensive disruption

Defensive disruption

Blocks and steals are rare endpoints. Defensive Disruption Rate combines denied actions, forced redirects, contested decisions, and recovery success. Weighting them 0.35, 0.25, 0.25, and 0.15 creates a repeatable index. Each event must be coded from an analyst-owned sample before the formula represents real play.

Role-adjusted efficiency

Role-adjusted efficiency

Efficiency should be evaluated against assignment difficulty. Define RAE = observed possession value minus expected value given usage, shot-clock pressure, opponent quality, and teammate spacing. In a synthetic case, 1.06 points per possession against an expectation of 0.99 produces plus 0.07 role-adjusted value.

Shrinkage and sample size

Shrinkage and sample size

Small samples exaggerate extremes. Use Adjusted HIS = n divided by n plus k times observed HIS, where k is a reliability constant. With n = 300 possessions, k = 500, and HIS = 1.20, the adjusted score is 0.45. At n = 1,500 the same observed score becomes 0.90.

Interpretation

Interpretation

A player can be better than the box score suggests when several low-visibility contributions point in the same direction and survive role adjustment, shrinkage, and lineup checks. HIS is an auditable framework for finding that pattern. It is not a claim about any real player until applied to independently collected observations.

THE FINAL PULSE

Players outperform box scores when repeatable low-visibility actions improve possessions after role, lineup, and sample-size adjustments.

Players outperform box scores when repeatable low-visibility actions improve possessions after role, lineup, and sample-size adjustments.