Introduction

For these two exercises, I pulled the 2021-22 NBA Advanced Stats table from Basketball Reference into Excel, saved it as a CSV file, and loaded it into RStudio.

I used that data to look at team performance through the four factors: shooting, turnovers, rebounding, and free throw rate.

Part 1: Four Factors Team Comparison

OKC’s Rebuild-Season Profile

This section summarizes each team’s offensive and defensive four factors, then specifically takes a closer look at the Oklahoma City Thunder that season.

I thought OKC was a good team to use because the 2021-22 season came during the middle of its rebuild. The team was still struggling at that point, but now it looks like one of the strongest teams in the league. That makes the comparison more interesting because it shows what the team looked like before the rebuild fully turned around.

The offensive and defensive numbers were compared by subtracting the defensive value from the offensive value for each factor. This gives a cleaner look at where OKC had advantages, where it had problems, and what kind of team profile it had during that season.

Column Explanation
Column Meaning
eFG% Effective field goal percentage. This measures shooting efficiency and gives extra value to three point shots.
TOV% Turnover percentage. This estimates how often a team turns the ball over.
ORB% Offensive rebounding percentage. This measures how often a team gets an offensive rebound after its own missed shot.
DRB% Defensive rebounding percentage. This measures how often a team gets the defensive rebound after an opponent miss.
FT/FGA Free throws per field goal attempt. This measures how often a team gets to the free throw line compared to its shot attempts.
ORtg Offensive rating. This estimates points scored per 100 possessions.
DRtg Defensive rating. This estimates points allowed per 100 possessions.
R Squared R squared shows how much variation in the response variable is explained by the regression model.

Offensive Four Factors by Team

This table summarizes each team’s offensive four factors. On offense, the main idea is pretty simple: teams want to shoot efficiently, limit turnovers, create extra chances through offensive rebounds, and get to the free throw line. A higher eFG%, ORB%, and FT/FGA are usually good, while a lower TOV% is better because it means the team is taking care of the ball.

NBA Offensive Four Factors by Team
Team offensive performance by shooting, turnovers, rebounding, and free throw rate
Team Offensive eFG% Offensive TOV% Offensive ORB% Offensive FT/FGA
Atlanta Hawks 0.543 10.8 23.0 0.205
Boston Celtics 0.542 12.4 24.0 0.195
Brooklyn Nets 0.540 12.5 23.9 0.198
Charlotte Hornets 0.544 11.6 23.3 0.173
Chicago Bulls 0.541 11.8 20.4 0.201
Cleveland Cavaliers 0.538 13.2 24.0 0.198
Dallas Mavericks 0.538 11.7 21.3 0.192
Denver Nuggets 0.556 13.2 21.9 0.194
Detroit Pistons 0.494 12.6 23.4 0.194
Golden State Warriors 0.552 13.5 22.8 0.181
Houston Rockets 0.534 14.5 21.7 0.202
Indiana Pacers 0.531 12.7 25.5 0.184
Los Angeles Clippers 0.531 12.5 20.6 0.178
Los Angeles Lakers 0.537 12.8 21.1 0.190
Memphis Grizzlies 0.522 11.2 30.0 0.180
Miami Heat 0.547 13.4 23.5 0.204
Milwaukee Bucks 0.546 11.9 23.0 0.199
Minnesota Timberwolves 0.539 12.4 24.4 0.198
New Orleans Pelicans 0.517 12.5 26.9 0.208
New York Knicks 0.513 12.0 25.1 0.208
Oklahoma City Thunder 0.497 12.5 21.8 0.169
Orlando Magic 0.503 13.0 19.8 0.175
Philadelphia 76ers 0.534 11.6 20.1 0.232
Phoenix Suns 0.549 11.6 22.3 0.176
Portland Trail Blazers 0.515 13.0 22.9 0.188
Sacramento Kings 0.525 12.5 21.3 0.203
San Antonio Spurs 0.527 11.1 23.7 0.166
Toronto Raptors 0.510 11.0 28.4 0.177
Utah Jazz 0.555 12.7 25.4 0.208
Washington Wizards 0.532 12.1 20.9 0.197

Defensive Four Factors by Team

This table summarizes each team’s defensive four factors. On defense, lower opponent shooting efficiency is better. A higher opponent turnover percentage is better because it means the defense forced more mistakes. A higher defensive rebounding percentage is better because it means the team finished more defensive possessions. A lower opponent free throw rate is better because it means the defense avoided sending teams to the line.

NBA Defensive Four Factors by Team
Team defensive performance by opponent shooting, turnovers forced, rebounding, and free throw rate allowed
Team Defensive eFG% Defensive TOV% Defensive DRB% Defensive FT/FGA
Atlanta Hawks 0.543 11.5 76.9 0.177
Boston Celtics 0.502 12.5 77.3 0.183
Brooklyn Nets 0.521 11.7 75.1 0.201
Charlotte Hornets 0.544 13.1 74.8 0.187
Chicago Bulls 0.541 11.9 78.3 0.199
Cleveland Cavaliers 0.520 12.3 76.5 0.172
Dallas Mavericks 0.521 12.2 78.0 0.185
Denver Nuggets 0.537 11.7 78.3 0.188
Detroit Pistons 0.541 13.1 75.6 0.226
Golden State Warriors 0.509 13.0 78.7 0.201
Houston Rockets 0.554 12.3 74.4 0.206
Indiana Pacers 0.552 11.9 76.2 0.204
Los Angeles Clippers 0.514 12.0 74.4 0.167
Los Angeles Lakers 0.540 12.2 75.8 0.192
Memphis Grizzlies 0.523 13.3 77.8 0.195
Miami Heat 0.524 13.8 78.0 0.209
Milwaukee Bucks 0.536 11.6 78.6 0.165
Minnesota Timberwolves 0.535 14.2 74.9 0.227
New Orleans Pelicans 0.547 13.0 78.2 0.196
New York Knicks 0.521 11.7 78.8 0.198
Oklahoma City Thunder 0.533 11.8 76.1 0.169
Orlando Magic 0.532 11.7 77.2 0.196
Philadelphia 76ers 0.524 12.1 76.8 0.192
Phoenix Suns 0.510 13.0 77.1 0.195
Portland Trail Blazers 0.559 12.7 76.9 0.222
Sacramento Kings 0.551 11.9 76.0 0.176
San Antonio Spurs 0.532 12.1 75.1 0.176
Toronto Raptors 0.535 14.4 75.6 0.199
Utah Jazz 0.521 10.9 78.3 0.164
Washington Wizards 0.529 10.7 76.9 0.202

League Average Summary

This table shows the league mean for each offensive and defensive four factor, as well as how OKC stacks up against league average.

League Mean four factors Summary
Comparison team: Oklahoma City Thunder
Metric League Mean OKC Value OKC Rank Percent Difference
Offensive eFG% 0.532 0.497 29 −6.5%
Offensive TOV% 12.343 12.500 15 1.3%
Offensive ORB% 23.213 21.800 21 −6.1%
Offensive FT/FGA 0.192 0.169 29 −12.2%
Defensive eFG% 0.532 0.533 16 0.2%
Defensive TOV% 12.343 11.800 22 −4.4%
Defensive DRB% 76.753 76.100 20 −0.9%
Defensive FT/FGA 0.192 0.169 4 −12.1%

OKC Thunder Offensive vs Defensive Comparison

The table below compares the Thunder’s offensive and defensive Four Factors by subtracting the defensive value from the offensive value. Because each Four Factor has a different direction, the differences have to be interpreted carefully. For example, a positive shooting difference is usually good because it means the team shot better than its opponents. A positive turnover difference can be bad because teams want fewer turnovers on offense and more forced turnovers on defense.

Oklahoma City Thunder Four Factors Comparison
Offensive factor minus defensive factor
Team Offensive eFG% Defensive eFG% eFG% Difference Offensive TOV% Defensive TOV% TOV% Difference Offensive ORB% Defensive DRB% Rebounding Difference Offensive FT/FGA Defensive FT/FGA
Oklahoma City Thunder 0.497 0.533 −0.036 12.5 11.8 0.7 21.8 76.1 −54.3 0.169 0.169

Oklahoma City Thunder Four Factors League Rank

This chart shows where OKC ranked in each four factors category during the 2021-22 season.

Part 1 Summary

The four factors show why OKC struggled during the 2021-22 season. The team was still in the middle of its rebuild, and the numbers reflect that. The Thunder ranked near the bottom of the league in key offensive areas, especially shooting efficiency and free throw rate, and were only above average in defensive FT rate and offensive turnovers…barely


Part 2: Simple Linear Regression Comparison

Rating Explained One Factor at a Time

This section uses simple linear regression to see which four factors were most connected to team rating.

For offense, each model used ORtg as the outcome and tested one offensive factor at a time. For defense, each model used DRtg as the outcome and tested one defensive factor at a time.

Regression Setup

Each regression tests one four factors category at a time. The R squared value shows how much of the difference in team rating is explained by that one factor.

Simple Linear Regression Setup
Each model tests one Four Factor at a time
Side Response Predictor Main Question
Offense ORtg Offensive eFG% How much does shooting efficiency explain offensive rating?
Offense ORtg Offensive TOV% How much do turnovers explain offensive rating?
Offense ORtg Offensive ORB% How much does offensive rebounding explain offensive rating?
Offense ORtg Offensive FT/FGA How much does free throw rate explain offensive rating?
Defense DRtg Defensive eFG% How much does opponent shooting efficiency explain defensive rating?
Defense DRtg Defensive TOV% How much do forced turnovers explain defensive rating?
Defense DRtg Defensive DRB% How much does defensive rebounding explain defensive rating?
Defense DRtg Defensive FT/FGA How much does opponent free throw rate explain defensive rating?

Offensive Regression Results

Offensive Rating Regression Results
Response variable: ORtg
Factor Model R Squared
Offensive eFG% ORtg on Offensive eFG% 0.622
Offensive TOV% ORtg on Offensive TOV% 0.164
Offensive ORB% ORtg on Offensive ORB% 0.123
Offensive FT/FGA ORtg on Offensive FT/FGA 0.080

Defensive Regression Results

Defensive Rating Regression Results
Response variable: DRtg
Factor Model R Squared
Defensive eFG% DRtg on Defensive eFG% 0.829
Defensive DRB% DRtg on Defensive DRB% 0.148
Defensive TOV% DRtg on Defensive TOV% 0.088
Defensive FT/FGA DRtg on Defensive FT/FGA 0.050

Regression Results Summary

The regression results were pretty clear on both sides of the ball. Offensive eFG% had the highest R squared for offensive rating, while defensive eFG% had the highest R squared for defensive rating.

The difference was that offensive eFG% stood out by a much larger margin. Shooting efficiency explained far more of the variation in ORtg than the other offensive factors. Defensive eFG% was also the strongest defensive factor, but the gap between it and the other defensive factors was not as extreme.

Regression Visual

This chart compares the R squared values from the offensive and defensive regressions.

Part 2 Summary

The regression results showed that shooting efficiency was the strongest factor on both sides of the ball.

Offensive eFG% had the strongest relationship with ORtg. That means shooting efficiency explained more of the difference in offensive rating than turnovers, offensive rebounding, or free throw rate.

Defensive eFG% had the strongest relationship with DRtg. That means opponent shooting efficiency explained more of the difference in defensive rating than forced turnovers, defensive rebounding, or opponent free throw rate.

The biggest takeaway is that all four factors matter, but they did not explain team rating equally. In this dataset for this season, I would have to conclude eFG% stood out as the most important individual factor on both sides of the ball.