NBA Unveils ‘Leverage’ Metric to Quantify Player Impact Beyond Traditional Stats
Table of Contents
The NBA is redefining how player contributions are measured with the launch of “leverage,” a new metric designed to pinpoint the moments and players that most significantly influence game outcomes. this innovative system, powered by machine learning and AWS AI, goes beyond conventional statistics to assess impact, even in seemingly minor plays.
The outcome of a basketball game is rarely simple. While dramatic buzzer-beaters – like Nikola VucevicS game-winner for the Chicago Bulls against the Orlando Magic on November 29 – capture attention, many plays subtly shift the odds of winning. Leverage aims to quantify a player’s impact by comparing what did happen in a game to what could have happened. The system utilizes a machine learning model to predict win probability based on the game state – score, time remaining, and other factors. Instead of simply measuring the difference between the start and end of a possession, Leverage compares a made shot to the projected outcome of a missed shot with a defensive rebound.
For example, if a team has a 60% win probability before a shot and a player makes a 3-pointer increasing that probability to 75%, Leverage doesn’t focus on the 15% increase. Instead, it compares the 75% win probability to the estimated 55% win probability that would have resulted from a missed shot and defensive rebound. the resulting 20% swing determines a portion of the player’s Leverage score.
A higher Leverage score indicates a greater positive impact on a team’s chances of winning. generally, a score of 1.0 correlates with improving win probability by 10%. Crucially, Leverage scores can be positive or negative, reflecting both beneficial and detrimental actions.
Beyond the Shot: A Holistic View of Impact
The NBA’s new metric doesn’t solely credit the player taking the shot. Leverage distributes credit across all players on the court, utilizing existing NBA tracking metrics like Expected Field Goal percentage (xFG%) and Defensive Pressure Score. This means that assists, screens, and even strong defensive rotations all contribute to a player’s overall leverage score.
“Not all credit is given to the shot-taker, but it is distributed across all players on the court,” one analyst noted. The system also accounts for whether a shot was self-created or assisted, and adjusts credit based on the difficulty of the shot as resolute by xFG%. Furthermore, leverage recognizes the importance of sequences – a steal leading to a fast break dunk, as an example, awards credit to the player who forced the turnover.
Diving Deeper: Subtypes of Leverage
The NBA has broken down Leverage into several subtypes to provide a more granular understanding of player impact:
- Offense: Leverage generated thru offensive actions.
- Defense: Leverage generated through defensive actions.
- Shot Making: Leverage directly tied to prosperous and unsuccessful shot attempts.
- Creation: Leverage earned by players who create scoring opportunities.
- Rebound: leverage attributed to securing rebounds.
- Turnover: Leverage assigned for committing turnovers (negative leverage).
- On Ball Defense: Leverage earned through blocks, steals, and contesting shots.
Leverage represents a meaningful step forward in basketball analytics, offering a more nuanced and comprehensive way to evaluate player performance and identify the true game-changers. It highlights how players influence their team’s chances of winning across every facet of the game, not just in scoring, and underscores the moments that truly swing the outcome.
Learn more about the NBA’s Leverage metric: https://www.nba.com/inside-the-game/player/leverage.
