A team has a 35-25 record. Another sits at 32-28.
Which one is better?
The first one, obviously. They have 3 more wins.
Not so fast. The first team played an easy schedule and won a lot of close games. The second faced tough opponents and lost several overtime games.
Net Rating tells you the truth: the second team is probably better.
This metric measures the difference between points scored and allowed per 100 possessions — removing the influence of pace and revealing a team's true efficiency.
What Is Net Rating?
The Definition
Net Rating (NRtg) is the difference between a team's Offensive Rating and Defensive Rating.
Net Rating = Offensive Rating - Defensive Rating
Offensive Rating (ORtg)
Offensive Rating measures how many points a team scores per 100 possessions.
Offensive Rating = (Points Scored / Possessions) x 100
| ORtg | Interpretation |
|---|---|
| > 118 | Elite offense |
| 115-118 | Very good offense |
| 112-115 | Good offense |
| 109-112 | Average offense |
| 106-109 | Below-average offense |
| < 106 | Struggling offense |
Defensive Rating (DRtg)
Defensive Rating measures how many points a team allows per 100 possessions.
Defensive Rating = (Points Allowed / Possessions) x 100
| DRtg | Interpretation |
|---|---|
| < 108 | Elite defense |
| 108-111 | Very good defense |
| 111-114 | Good defense |
| 114-117 | Average defense |
| 117-120 | Below-average defense |
| > 120 | Struggling defense |
Note: For defense, a lower number is better (fewer points allowed).
Combined Net Rating
| Net Rating | Interpretation |
|---|---|
| > +8 | Elite team, contender |
| +5 to +8 | Very good team |
| +2 to +5 | Above-average team |
| -2 to +2 | Average team |
| -5 to -2 | Below-average team |
| < -5 | Weak team |
Join ProbWin for free
Access our AI predictions for NBA, NHL, Soccer & Tennis. Sign up in 30 seconds.
Why Use Net Rating?
The Problem with Win-Loss Records
Win-loss records are influenced by factors that don't predict future performance:
| Factor | Impact on Record |
|---|---|
| Close games (clutch) | High variance, low predictive value |
| Strength of schedule | Inflates or deflates the record |
| Sporadic injuries | Record doesn't reflect full roster talent |
| Pure luck | Bounces, contested shots, officiating |
Net Rating Cuts Through the Noise
Net Rating measures average performance, not outcomes:
Record: "How many games did you win?"
Net Rating: "By how much do you dominate on average?"
| Team | Record | Net Rating | Diagnosis |
|---|---|---|---|
| A | 38-22 | +2.5 | Inflated record (clutch, schedule) |
| B | 32-28 | +5.8 | Deflated record (bad luck, schedule) |
| C | 35-25 | +4.2 | Aligned with talent |
Team B is probably the best despite the worse record.
Correlation with Future Results
Studies show Net Rating predicts the future better than win-loss records:
| Metric | Correlation with future record |
|---|---|
| Current record | 0.65 |
| Net Rating | 0.75 |
| Point differential | 0.72 |
Net Rating is 15% more predictive than win-loss record.
Analyzing Net Rating in Detail
Breaking Down Offensive Rating
ORtg depends on several factors:
| Factor | Impact on ORtg |
|---|---|
| eFG% (Effective FG%) | Shooting efficiency |
| TOV% (Turnover Rate) | Ball security |
| ORB% (Offensive Rebound Rate) | Second-chance points |
| FT Rate | Free throws generated |
These are Dean Oliver's "Four Factors," which explain roughly 90% of ORtg.
Breaking Down Defensive Rating
DRtg depends on the same factors on the opponent's side:
| Factor | Impact on DRtg |
|---|---|
| Opponent eFG% | Preventing efficient shots |
| Opponent TOV% | Forcing turnovers |
| DRB% (Defensive Rebound Rate) | Limiting second chances |
| Opponent FT Rate | Avoiding fouls |
The ORtg/DRtg Balance
| ORtg | DRtg | Net Rating | Profile |
|---|---|---|---|
| 118 | 110 | +8 | Elite on both ends |
| 120 | 116 | +4 | Offense-dominant |
| 112 | 106 | +6 | Defense-dominant |
| 115 | 115 | 0 | Balanced but average |
Net Rating and Pace
What Is Pace?
Pace measures the number of possessions per game for a team.
Pace = Possessions per 48 minutes
| Pace | Style |
|---|---|
| > 102 | Fast-paced (run & gun) |
| 98-102 | Average pace |
| < 98 | Slow-paced (half-court) |
Why Net Rating Is "Pace-Adjusted"
Net Rating is calculated per 100 possessions, which neutralizes pace:
Team A: 110 points in 100 possessions = ORtg 110
Team B: 110 points in 95 possessions = ORtg 115.8
Team B is more efficient despite scoring the same number of points.
Why Pace Matters for Totals
We cover Pace in greater depth in a dedicated article, as it is crucial for totals betting.
Net Rating Variations
"Clutch" Net Rating
Net Rating in close situations (within 5 points in the last 5 minutes):
| Clutch Net Rating | Interpretation |
|---|---|
| > +5 | Excellent in crunch time |
| -5 to +5 | Inconsistent |
| < -5 | Struggles in crunch time |
Caution: Clutch Net Rating relies on a small sample size and is less predictive.
Net Rating by Lineup
Every 5-man combination has its own Net Rating:
| Lineup | Minutes | Net Rating | Notes |
|---|---|---|---|
| Starting five | 200 | +12.5 | Highly effective |
| Bench unit | 150 | -3.2 | Struggling |
| "Death lineup" | 80 | +18.4 | Dominant |
Net Rating Home vs. Away
| Situation | Typical Net Rating |
|---|---|
| Home | +3 to +4 vs. average |
| Away | -3 to -4 vs. average |
The home/away gap is approximately 6-8 Net Rating points in the NBA. For a thorough analysis of this factor, check out our guide on Home Court Advantage.
Using Net Rating for Betting
Strategy #1: Comparing Net Ratings
To estimate a spread:
Estimated Spread ~ (Net Rating A - Net Rating B) / 2.5 + Home Court Advantage
| Net Rating A | Net Rating B | HCA | Estimated Spread |
|---|---|---|---|
| +8 | +2 | A at home | A -5 to -6 |
| +5 | -3 | B at home | A -1 to -2 |
| +10 | +10 | A at home | A -3 |
Strategy #2: Identifying Undervalued Teams
Compare Net Rating with the record:
| Net Rating | Record | Win % | Diagnosis |
|---|---|---|---|
| +6 | 28-32 | 47% | Undervalued (unlucky) |
| +1 | 38-22 | 63% | Overvalued (lucky) |
| +4 | 35-25 | 58% | Aligned |
Bet on undervalued teams — their Net Rating suggests they should be winning more. This is the core principle of value betting.
Strategy #3: Recent vs. Season-Long Net Rating
Recent form matters:
| 10-Game Net Rating | Season Net Rating | Signal |
|---|---|---|
| +10 | +5 | Hot streak, overperforming |
| +2 | +7 | Struggling |
| +6 | +6 | Stable |
Strategy #4: Cross-Referencing ORtg vs. Opponent DRtg
To estimate a team's scoring output:
Team A ORtg vs. Team B DRtg = Expected efficiency for A
Points A ~ (ORtg A + DRtg B) / 2 x (Combined Pace / 100)
| ORtg A | DRtg B | Game Pace | Estimated Points A |
|---|---|---|---|
| 118 | 115 | 100 | ~116-117 |
| 112 | 108 | 98 | ~108-109 |
| 115 | 120 | 102 | ~119-120 |
Join ProbWin for free
Access our AI predictions for NBA, NHL, Soccer & Tennis. Sign up in 30 seconds.
The "Pythagorean Win %"
The Pythagorean Formula in Basketball
The Pythagorean Win % estimates the winning percentage a team "should" have based on points scored and allowed:
Pythagorean Win % = Points For^14 / (Points For^14 + Points Against^14)
(The exponent varies across studies, typically between 13.91 and 16.5)
Comparing with Actual Record
| Team | Actual Record | Pyth Win % | Difference |
|---|---|---|---|
| A | 65% (39-21) | 58% | +7% (lucky) |
| B | 55% (33-27) | 62% | -7% (unlucky) |
| C | 60% (36-24) | 59% | Aligned |
Team B will likely improve, while Team A will likely regress.
Link to Net Rating
Pythagorean Win % and Net Rating are highly correlated:
| Net Rating | Approx. Pyth Win % |
|---|---|
| +10 | ~73% |
| +5 | ~62% |
| 0 | ~50% |
| -5 | ~38% |
| -10 | ~27% |
Net Rating in the Playoffs
Regular Season vs. Playoffs
In the playoffs, Net Rating dynamics change:
| Factor | Impact |
|---|---|
| Fewer games | Higher variance |
| Adjustments | Teams adapt during a series |
| Stars in playoff mode | Elevated individual performance |
| Officiating | Different (fewer fouls called) |
Does Net Rating Predict Playoff Success?
| Season Net Rating | Playoff Win % |
|---|---|
| > +8 | 68% |
| +5 to +8 | 55% |
| +2 to +5 | 48% |
| < +2 | 35% |
Net Rating remains predictive but with higher variance in the playoffs.
Full Example: Net Rating Analysis
The Game
Celtics @ Heat — Tuesday
The Data
| Team | Record | Net Rating | ORtg | DRtg | Pace |
|---|---|---|---|---|---|
| Boston | 42-18 | +9.2 | 120.5 | 111.3 | 99.8 |
| Miami | 35-25 | +3.8 | 113.2 | 109.4 | 97.5 |
The Analysis
Boston: - Elite Net Rating (+9.2) - Dominant ORtg (120.5) — best offense in the NBA - Solid DRtg (111.3) - Record aligned with Net Rating - True elite team
Miami: - Good Net Rating (+3.8) - Defense (109.4) stronger than offense (113.2) - Slow Pace (97.5) — controls the tempo - Defensive team, capable of slowing the game down
Calculating the Spread
Net Rating differential: +9.2 - (+3.8) = +5.4
Spread conversion: 5.4 / 2.5 ~ 2.2 points
Miami HCA: +3 points
Estimated spread: Boston -2.2 + 3 = Miami +0.8
Or roughly: Boston -1 / Pick'em
The Bookmaker's Line
Spread: Boston -4.5
Total: 215.5
The Verdict
Spread analysis: - Our estimate: Boston -1 or pick'em - Line: Boston -4.5 - Gap: 3.5 points in Miami's favor - Miami +4.5 is potentially good value
Totals analysis: - Miami plays slowly (Pace 97.5) - Boston's elite offense meets Miami's stout defense - Estimated combined pace: ~98.5 - Under worth considering (Miami imposes its tempo)
Recommended bets: 1. Miami +4.5 (value on the spread) 2. Under 215.5 (Miami imposes its pace)
Where to Find Net Rating
Free Sources
| Source | URL | Data |
|---|---|---|
| NBA.com/stats | nba.com/stats | ORtg, DRtg, official Net Rating |
| Basketball Reference | basketball-reference.com | Net Rating, Four Factors |
| Cleaning the Glass | cleaningtheglass.com | Net Rating filtered (no garbage time) |
| ESPN | espn.com/nba/stats | Advanced stats |
Key Columns to Look For
| Column | Meaning |
|---|---|
| ORtg | Offensive Rating |
| DRtg | Defensive Rating |
| NRtg | Net Rating |
| Pace | Possessions per game |
| eFG% | Effective Field Goal % |
Cleaning the Glass: The Gold Standard
Cleaning the Glass filters out garbage time and non-competitive situations, providing a "purer" Net Rating:
Standard Net Rating: Includes everything
CTG Net Rating: Excludes garbage time (games already decided)
CTG is often more predictive.
Common Mistakes
Mistake #1: Relying Solely on Win-Loss Record
Wrong: "They have 40 wins, they must be better"
Right: "They have 40 wins but a Net Rating of +2 — they've been lucky"
Mistake #2: Ignoring Context
Wrong: "Net Rating +8, elite team"
Right: "Net Rating +8 but based on 30 games — limited sample size"
Mistake #3: Forgetting Home Court Advantage
Wrong: "Boston +9 NRtg vs Miami +4 NRtg = Boston -2"
Right: "Boston +9 vs Miami +4 BUT Miami at home = tighter game"
Mistake #4: Not Looking at ORtg and DRtg Separately
Wrong: "Both have +5 Net Rating, same level"
Right: "A is +5 (offense 118, defense 113), B is +5 (offense 110, defense 105) — different styles"
Mistake #5: Ignoring Injuries
Wrong: "Their Net Rating is +10"
Right: "Their Net Rating is +10 WITH their star, -2 WITHOUT her"
How ProbWin Uses Net Rating
Our NBA model integrates Net Rating as a core metric:
1. Season-long Net Rating (garbage time filtered)
2. Recent Net Rating (10-15 games)
3. ORtg and DRtg separately
4. ORtg vs. opponent DRtg cross-reference
5. Pace adjustment for the matchup
6. Net Rating by lineup (when available)
7. Comparison with the market line
Net Rating is more predictive than win-loss record — we prioritize it. Discover how our AI uses these metrics to generate profitable picks.
Check out our NBA picks to see this analysis in action.
Summary: Net Rating in 7 Key Points
| # | Takeaway |
|---|---|
| 1 | Net Rating = ORtg - DRtg (points per 100 possessions) |
| 2 | Net Rating > +5 = good team, > +8 = elite |
| 3 | More predictive than win-loss record (~15% better) |
| 4 | Compare with record to find mispriced teams |
| 5 | Look at ORtg and DRtg separately to understand a team's style |
| 6 | Recent Net Rating reveals current form |
| 7 | Don't forget Home Court Advantage (+3 pts) |
Next Steps
You now understand Net Rating. But to estimate NBA totals, you're missing a crucial piece: Pace.
Discover how Pace influences the number of possessions — and therefore the number of points — and how to use it to find value on Over/Under markets.