Most basketball stats are designed for a season-long story. Points Per Possession is designed for the next timeout.
PPP is the cleanest single measure of how efficient an offense is. It strips out pace, removes scheme noise, and reduces a 24-second possession to a single number any coach can read between plays. If you only learn one efficiency metric, this is it.
The PPP Formula
The math is simple:
``` PPP = Total Points Scored / Total Possessions ```
A possession is one trip down the floor. It ends in exactly one of three ways:
1. A made or missed field goal (after which the other team rebounds) 2. A turnover (live-ball or dead-ball) 3. A made or missed final free throw
The standard estimator for possessions, used by every public analytics source:
``` Possessions ≈ FGA + 0.44 × FTA + TO − OREB ```
The 0.44 coefficient is empirically derived. It approximates the share of free throws that end a possession — a two-shot foul ends a possession on the second shot, but a one-and-one only ends on the second shot when the first is made, and an and-1 doesn't end the possession at all. Across decades of NBA data, 0.44 is a tight average.
Multiply PPP by 100 and you have Offensive Rating — the same metric, scaled for league comparisons.
The Four Inputs That Move PPP Most
In any 100-possession sample, four variables explain about 80% of PPP variance:
1. Effective Field Goal Percentage (eFG%). Adjusts for the fact that a three-pointer is worth 1.5x a two. `eFG% = (FGM + 0.5 × 3PM) / FGA`. Every percentage point of eFG% above league average adds roughly 1 point to per-100 PPP.
2. Turnover Rate (TOV%). Turnovers / Possessions. A possession that ends in a turnover is worth zero points and frequently produces ~1.3 PPP for the opponent in transition. Reducing TOV% by 2 percentage points typically adds 2-3 points per 100 to PPP.
3. Offensive Rebound Rate (OREB%). Each offensive rebound effectively extends a possession. The second-chance points have a higher PPP than first-chance shots (~1.20 vs the league baseline ~1.13), so OREB% is leverage.
4. Free Throw Rate (FTR). Free throws to field goal attempts. Free throws are the highest-PPP shot type in basketball (a 78% league free-throw shooter is generating ~1.56 PPP per trip).
These four inputs are called the "Four Factors" in basketball analytics — a framework Dean Oliver introduced in *Basketball on Paper* (2004) and that Basketball-Reference maintains as the standard analytical lens for offensive efficiency. The 12-lens system treats them collectively as the Analytics lens; the scouting report evolution piece covers how staffs weight them differently across a series.
Why Box Scores Lie About Offense
A team can score 120 points in a game and have a PPP of 1.05 — average. A team can score 95 points in a game and have a PPP of 1.18 — elite. The difference is pace.
Box scores measure totals. PPP measures efficiency. In a fast-paced game (110 possessions per team), scoring 120 points is mediocre. In a slow-paced game (88 possessions per team), scoring 95 points is excellent.
This is why coaches read PPP and almost never read total points scored. A coach making a defensive timeout call cares whether the opponent is scoring at 1.20 PPP, not whether they've scored 18 points in the quarter. Those are very different questions.
How NBA Staffs Use PPP in Real Time
Three live uses, in order of how often they get called:
1. The 5-possession rolling window. Analytics staffs track PPP across the last 5 possessions on the bench. When the rolling number crosses a threshold (typically below 0.80 for the defense, above 1.30 for the opponent's offense), the head coach gets a flag.
2. Lineup-specific PPP. Every five-man lineup has a PPP both ways. When a lineup is below 1.00 PPP across 10+ possessions in a game, the coaching staff is debating substitutions in real time.
3. Possession-type PPP. PPP by play type (PnR ball-handler, transition, post-up, isolation, spot-up). A defense yielding 1.20 PPP on transitions specifically is going to see ATO sets that force transition off live-ball turnovers.
The deepest use is the one fans never see: PPP forecasts. Staffs run game-plan simulations against the opponent's actual coverages and tag the expected PPP for each set in their playbook. When the actual game starts diverging from the forecast — when the in-game PPP on a set is below 0.85 against an expected 1.10 — the staff knows the opponent has scouted the set and the call list needs revision.
Building Your Own PPP Tracker
You don't need a stats firm to do this for your own team. Three columns on a spreadsheet:
- Possession count (the trip number)
- Points scored on the possession
- Outcome (FG / FT / TO / OREB)
Every game, that's it. The rolling PPP across the last 10 possessions tells you whether your team's hot or cold. The PPP by lineup tells you who actually generates offense. The PPP by play call tells you which of your sets the defense has figured out.
This is the same workflow scouting reports evolve through across a playoff series — at the pro level, with more granularity. At the amateur level, three columns and a clean possession count is the difference between coaching from feel and coaching from evidence.
PPP isn't the most advanced stat in basketball. It's the most useful. Learn it, track it, and you'll read the game differently in a week.