Modern NBA analytics is intimidating from the outside and surprisingly simple on the inside. Front-office discourse, advanced-stat Twitter, and high-end coaching staff conversations run on a relatively small vocabulary: seven numbers that explain about 90% of what "advanced basketball stats" actually measure.
This piece is that vocabulary. No math beyond addition and percentages. By the end you should be able to read an NBA front-office quote about "their defensive Net Rating dropped to 110 in the playoffs as their opponents' eFG% climbed past 55%" and translate it into plain English.
Aimed at four audiences: coaches who want to use analytics without a graduate degree, bettors and DFS players who want a real edge instead of a coinflip, journalists who want to write smarter, and front-office aspirants who want to talk the talk in interviews.
Number 1: Points Per Possession (PPP)
What it is: average points scored per offensive possession used. The foundation metric — every other modern basketball stat is a refinement of this one.
The math: Total points / total possessions. A possession ends when the team scores, turns it over, or has a defensive rebound recovered against them.
The league context (2026): league-average offensive PPP is roughly 1.15. Elite offenses run 1.20+; struggling offenses run 1.10 or worse. The Knicks won the 2026 title at about 1.18 PPP across the playoffs.
Why it beats points per game: PPG is a function of pace. A team that plays fast can score 120 points on bad efficiency; a team that plays slow can score 100 on great efficiency. PPP normalizes for pace and lets you compare any offense to any other.
Use case: when you see a team's offense get "shut down" to 105 points in a playoff loss — was that bad efficiency or just slow pace? Convert to PPP. If they were at 1.18 PPP, the offense was fine; if they were at 0.95 PPP, the offense actually got shut down.
Our PPP deep-dive covers the full math and league context.
Number 2: Effective Field Goal Percentage (eFG%)
What it is: the shooting-efficiency stat that correctly weights 3-pointers (worth 1.5x a 2-pointer because they're 50% more valuable).
The math: (FGM + 0.5 × 3PM) / FGA.
The league context (2026): league average eFG% is roughly 54%. Above 56% is elite shooting; below 50% is poor.
Why it beats raw FG%: raw field-goal percentage punishes 3-point-heavy teams unfairly. A team shooting 42% on a lot of 3s might be more efficient than a team shooting 48% on all 2s — eFG% catches this; raw FG% doesn't.
Use case: evaluating shooting performance. "He shot 45%" — was that good or bad? Calculate eFG%. If it's 58%, he was elite; if it's 48%, he was below average.
Number 3: Turnover Percentage (TOV%)
What it is: the percentage of offensive possessions that end in a turnover.
The math: TOV / (FGA + 0.44 × FTA + TOV). The 0.44 weight accounts for the fact that not every free throw attempt ends a possession.
The league context (2026): league average TOV% is roughly 13%. Elite ball control is 11% or lower; sloppy teams run 15%+.
Why it matters: turnovers are roughly 1.2x worse than missed shots, because they often produce transition opportunities for the opponent. A team with great eFG% but terrible TOV% is still a bad offense.
Use case: evaluating a primary handler. Brunson's 2026 Finals MVP run had a TOV% under 10% — top-tier ball security at high volume.
Number 4: Offensive Rebound Percentage (OREB%)
What it is: the percentage of available offensive rebounds a team or player grabs.
The math: OREB / (team OREB + opponent DREB while team was on offense).
The league context (2026): league average OREB% is roughly 26%. Elite offensive-rebounding teams run 30%+; weak ones run 22% or lower.
Why it matters: an offensive rebound creates a second-chance possession with elevated PPP (typically 1.20+). OREB% is one of the cleanest ways to find a team's hidden offensive value that doesn't show up in shooting stats.
Use case: explaining why a poor-shooting team wins. Sometimes the answer is "they get every miss back" — that's high OREB%.
Want to study OREB% by player and lineup with NBA-staff tagging? Start a HoopBrief plan — the Analytics lens surfaces OREB% on every possession across the 12-lens framework.
Number 5: Free Throw Rate (FTr)
What it is: how often a team or player gets to the free throw line per field goal attempt.
The math: FTA / FGA.
The league context (2026): league average FTr is roughly 0.22 (22 free-throw attempts per 100 FG attempts). Elite foul-drawers like SGA and Brunson run 0.35+; non-foul-drawing perimeter shooters run 0.15 or lower.
Why it matters: free throws are the most efficient shot in basketball — about 1.5 PPP for an average shooter. A player who reliably draws fouls has a built-in offensive floor that survives cold shooting nights.
Use case: evaluating scoring efficiency. A 22-PPG scorer with a 0.10 FTr and a 22-PPG scorer with a 0.35 FTr are completely different players — the second one has scoring resilience the first one doesn't.
Number 6: Net Rating
What it is: the difference between offensive PPP and defensive PPP allowed, scaled to per-100 possessions.
The math: Offensive Rating - Defensive Rating, where each is points per 100 possessions.
The league context (2026): the NBA champion typically has a Net Rating of +5 or higher across the season. +10 is a historically great team; -5 is a bottom-quarter team.
Why it matters: Net Rating is the cleanest single number for team quality. A +8 team will beat a +2 team on a neutral floor 70% of the time; a +12 team will beat a +0 team 80% of the time.
Use case: predicting series outcomes. A 7-game playoff series between a +9 team and a +4 team usually ends in 5 or 6 games. Net Rating is the closest thing to a single quality grade in basketball.
Number 7: RAPM (or BPM)
What it is: the gold-standard impact metric. RAPM (Regularized Adjusted Plus-Minus) uses regression to isolate each player's on-court effect from teammates and opponents.
The math: statistical regression on possession-level data — not computable by hand.
The league context (2026): elite players run +5 to +10 RAPM (impact per 100 possessions). Rotation players at +0 to +2. Replacement-level at -2 or worse. MVPs typically have RAPM in the +8 to +12 range.
Why it matters: raw plus-minus is noisy because it's contaminated by who you played with and against. RAPM strips that out and tells you the player's actual impact independent of context.
Use case: comparing players whose box-score stats are similar. Two players might average 22-8-5 but have RAPMs of +6 and +1. The +6 player is dramatically more valuable; the box score didn't show it.
Want RAPM-style impact numbers on every NBA player without doing the regression yourself? HoopBrief Pro ($999/mo) surfaces impact metrics across the league with plain-English interpretations.
How to Apply the 7 Numbers
For each of the seven, the practical workflow:
- Coaches: use PPP + eFG% + TOV% to evaluate your team's offensive performance; OREB% + Net Rating to find hidden weaknesses; FTr to know which players have built-in scoring floors.
- Bettors / DFS: track recent-form Net Rating differentials, fatigue-adjusted defensive PPP, and individual player FTr/eFG% drift. Closing-line edges come from applying basic stats faster, not from finding magic metrics.
- Journalists: when writing about a team's performance, lead with eFG%, TOV%, and Net Rating. Skip the raw point totals — they tell readers less.
- Front-office aspirants: master RAPM/BPM interpretation. The biggest skill gap between aspiring and actual NBA front offices is the ability to read impact metrics and translate them into player evaluation.
The single biggest analytics mistake at every level: chasing exotic stats before mastering these seven. A coach who can fluently read PPP, eFG%, TOV%, OREB%, FTr, Net Rating, and RAPM outperforms a coach who knows 30 stats but lacks fluency in any of them.
The HoopBrief Approach to Analytics
HoopBrief's 12-lens framework includes a dedicated Analytics lens that surfaces all seven of these numbers — plus their league context and a one-sentence tactical read — on every NBA possession, every matchup, and every game state.
Starter tier ($9.99/mo): Analytics lens with the 7 core numbers + plain-English reads on the NBA full season library.
Pro tier ($999/mo): adds possession-level data export, RAPM-style impact metrics across the league, and lineup-by-lineup Net Rating decomposition.
Team / Enterprise: custom data integrations for front offices and analytics staffs.
The framework is built specifically for the non-stats-degree audience: you see the number, the league context, and what it means tactically — without computing anything yourself.
Want to apply NBA-staff-grade analytics to any team or player you're studying? Start a HoopBrief plan today — $9.99/mo Starter, no math required.
Where to Go Next
Foundation reading: points per possession (PPP) explained for 2026, moneyball for basketball — how data is changing the NBA, why the mid-range still matters in today's NBA.
Tool comparisons: the basketball scouting software stack in 2026, Hudl vs Synergy vs HoopBrief 2026 comparison.
Hub: Playoff Prep Hub.
