Lead Dominance Index

Interactive signed score-differential area chart for measuring sustained game control in point-time units.

Game Setup

Regular season game logs from nbastatsv3.

Total LDI
0
Select a game
Average LDI
0
Select a game
Final Differential
0
Select a game

Score Differential Over Time

Team A lead area Team B lead area Differential

Computed Intervals

Start End Team A Team B Diff Duration Control Signed Contribution
Season Leader
-
Waiting for dataset
Hidden Winner
-
Team whose LDI most outperformed their record
Hidden Loser
-
Team whose LDI most underperformed their record

Wins vs Avg LDI Margin

Team Rankings

Ranked by Total LDI in point-minutes. Select a column to sort.

Team

Regular season team LDI from nbastatsv3, ranked by signed point-time control. Net Rating is from Basketball-Reference’s Advanced Team table.

Methodology

Waiting for season data

Lead Dominance Index (LDI) is treated here as a continuous-game measure of scoreboard control rather than as a box-score efficiency statistic. For each interval between observed score changes, the score differential is held constant and multiplied by the interval length; the team-season value is the signed sum of these point-time areas. Formally, for interval j, LDI = sum(D_j * delta_t_j), where D_j is the team's score differential and delta_t_j is elapsed game time. The displayed predictor, Avg LDI Margin, divides each team's season LDI point-minutes by the actual minutes it played, including overtime, so extra periods are included in both the numerator and the denominator. This construction follows the same premise that motivates score-process models in basketball: the evolving score differential contains information about team strength and win probability, not merely the final score. Stern's Brownian-motion model of sports scores is the statistical precedent for treating the game margin as a time-indexed process, and Kubatko, Oliver, Pelton, and Rosenbaum provide the broader basketball-analytics basis for evaluating team performance through normalized, possession-aware and efficiency-oriented summaries.

The predictive test is intentionally out of sample. All available regular seasons are ordered chronologically. For a held-out season s, the model fits a one-variable linear regression on prior seasons only: Predicted Win% = alpha + beta * Avg LDI Margin. The fitted equation is then applied to each team in season s, predicted win percentage is bounded to the interval [0, 1], and predicted wins are obtained by multiplying by games played. The table reports season-level accuracy, mean absolute error, root mean squared error, R2, correlation, and the largest team miss. Accuracy is defined as 1 - total absolute win error / total games, so it expresses the share of scheduled games not lost to absolute prediction error. For the earliest seasons, where no prior-season training set exists or the prior set is too small for a stable regression, the model trains on all other seasons and flags that fallback in the coverage statement. This design follows the out-of-sample validation logic emphasized by Sill in adjusted plus-minus work: a metric intended to explain winning should be judged by its performance on games or seasons not used to estimate the model, because in-sample fit can reward noise.

LDI should not be read as a proven replacement for plus/minus. If plus/minus means team point differential or net rating, LDI is usually expected to be less directly predictive of season wins than plus/minus, because plus/minus is the realized scoring margin that most closely determines wins and losses. NBA Stats defines net rating as team point differential per 100 possessions, and Basketball-Reference's Pythagorean wins framework estimates expected wins directly from points scored and allowed; both conventions reflect the empirical strength of aggregate scoring margin as a win predictor. LDI adds a different signal by preserving when the margin occurred: a team that leads by double digits for 40 minutes and wins by one will rate very differently from a team that is even all night and wins by one, even though both games have the same final margin. That temporal information can describe control, pressure, and game script, but it also creates reasons for weaker win prediction: late-game fouling, garbage time, deliberate clock management, and comeback volatility can make sustained control diverge from final margin. In short, LDI is best interpreted as a complementary measure of dominance shape. It is more descriptive of how a team controlled games, while team plus/minus or net rating remains the cleaner benchmark for predicting the number of games won.

If plus/minus instead refers to raw player plus/minus, the comparison changes. Raw player plus/minus measures the score change while a player is on the court, but Sill notes that unadjusted plus/minus is confounded by teammates, opponents, lineup collinearity, and overfitting; adjusted plus-minus and regularized adjusted plus-minus were developed precisely to correct those problems. LDI in this project is a team-season statistic, so it avoids individual lineup attribution but also does not identify player value. The appropriate conclusion is therefore limited: team LDI may be more stable and interpretable than raw single-player plus/minus for describing team control, but it should be expected to trail team net rating or point differential as a pure season-win predictor unless a direct head-to-head model shows otherwise.

Sources: Hal S. Stern, A Brownian Motion Model for the Progress of Sports Scores; Justin Kubatko, Dean Oliver, Kevin Pelton, and Dan T. Rosenbaum, A Starting Point for Analyzing Basketball Statistics; NBA Stats, Stat Glossary; Basketball-Reference, Glossary and Pythagorean Wins definition; Joseph Sill, Improved NBA Adjusted Plus-Minus Using Regularization and Out-of-Sample Testing.

Overall Accuracy
-
All tested team-seasons
Overall R2
-
Predicted wins vs actual wins
Average MAE
-
Mean season-level missed wins
Seasons Tested
-
Team-seasons tested

Season Results

Each row is one held-out regular season. Accuracy follows the article-style predicted-wins comparison.

Season Train Seasons Teams Formula Accuracy MAE RMSE R2 Correlation Largest Miss

Formula

Full Formula
Single game LDI_team = sum_{i=1}^{n} [(Score_team,i - Score_opponent,i) * minutes_i]
Season average Avg LDI Margin = total season LDI point-minutes / total minutes played
i
One interval in the game. Intervals start and end at score changes, period boundaries, or the final buzzer.
n
The total number of intervals being summed.
Score_team,i
The selected team's score during interval i.
Score_opponent,i
The opponent's score during interval i.
Delta t_i
The length of interval i. In point-min LDI, this is measured in minutes.
total season LDI
The sum of every game LDI for that team, measured in point-minutes.
total minutes played
The team's actual season minutes, including overtime minutes.

About LDI

Lead Dominance Index is designed to evaluate how a team performs qualitatively over the course of a game, not just the outcome of a game.

To explain, let’s look at the following two games:

On November 4, 2025, the Bulls played the Sixers. The game featured one lead change, and the largest lead was 24 points by the Sixers. The Sixers led the game for 47 minutes and 41 seconds. The Bulls won the game by two points.

Bulls–Sixers LDI Bulls 113, Sixers 111
Loading game data…

On New Year’s Eve, the Magic played the Pacers. The game featured 10 lead changes, and the largest lead was 10 points by the Pacers. The Magic led the game for 24 minutes and 58 seconds. The Magic won the game by two points.

Magic–Pacers LDI Magic 112, Pacers 110
Loading game data…

It’s obvious from this information that one of these games was close from wire to wire and the other was an amazing come-from-behind victory. But point differential and net rating tell us nothing about the path to victory—two points is two points.

So how can we account for this qualitative difference? How can we measure this qualitative difference across the season?

That’s what Lead Dominance Index (LDI) attempts to do. The solution is simple: don’t just measure the final score of the game. Instead, measure how much a team leads or trails by and multiply that margin by the amount of time it lasts.

The stat is calculated on a per-minute basis:
LDI = score differential × elapsed minutes

If the Bulls had a lead of exactly 10 points for an entire minute, then the Bulls’ LDI for that minute would be 10: 10 points multiplied by one minute. If the Bulls were losing by 10 points for two minutes, then their LDI would be -20.

This gives you Total LDI. In the Bulls–Sixers game, the Bulls had a Total LDI of -634. On its own, that number is confusing and does not mean much.

That is why Average LDI exists. It takes the total and divides it by the minutes in the game—48 in a regulation game—to produce the game’s average LDI. Let’s return to Bulls–Sixers. The Bulls had a Total LDI of -634 and an Average LDI of -13.21. The Bulls were losing by an average of 13.21 points in that game. In Magic–Pacers, the Magic had a Total LDI of 20.7 and an Average LDI of 0.43.

We have taken two games with the same point differential and demonstrated how the quality of those games was radically different.

Anomaly Finder

Games where scoreboard control and the final result disagree—or where one side never let go.

Dominated but lost

Waiting for the game index

Matchup

Select a row to open that game in the Game tab. Team LDI, Avg LDI Margin, and Time Leading belong to the highlighted team.