Basketball discourse in the country is dominated by two things: the eye test and traditional statistics.
The eye test relies on the human mind and is limited by it. We can’t watch everything, we don’t see everything, and we don’t remember everything. On the other hand, traditional statistics, while handy, are often flawed and fall short in describing important elements in basketball.
That is why we at Dribble Media are sharing our advanced statistics. Fair warning: these numbers are NOT the be-all and end-all of basketball discussion. They are just an additional set of tools to aid the eye test and the traditional statistics. These are meant to capture what might be missed by the other tools. At best, these numbers are guides that lead to proper conclusions, and not the conclusions themselves. Please exercise caution in using these numbers and try to frame them within the proper context.
Note: Minimum games and minutes needed to qualify for league leaderboards are six games and 15 minutes per game.
Table of Contents
1. Stat Presentations
1.1. Per 36 Minutes, Pace Adjusted
What it is: It adjusts statistics to what they would be if a player played 36 minutes at league average pace.
Why we use it: Simply adjusting statistics to per 36 minutes does not make it a fair comparison. Teams play at a different pace and we have to equalize the pace.
Example: In the 2020 PBA Philippine Cup, Christian Standhardinger finished fifth in points per game after averaging 19.9 points in 37.4 minutes. His team, NorthPort, was sixth in possessions per game. Stanley Pringle finished 10th on the scoring list with 18.5 points per game in 36.1 minutes per game. Ginebra finished with the league’s third-lowest possessions per game mark.
Simply equalizing their minutes played would still show that Standhardinger (19.2 points per 36 minutes) outscored Pringle (18.4 points per 36 minutes). But controlling both minutes and pace, we can see the true rate at which players score. After adjusting their numbers to what they would be after playing 36 minutes at a league average pace, Standhardinger averaged 18.8 points per 36 minutes while Pringle averaged 19.3 points per 36 minutes. After we account for and equalize pace, we can see Pringle scored at a higher rate than Standhardinger.
Limitations: The only limitation is that there’s no guarantee that the rate in which points (or any other stat) are acquired will remain the same in more or less minutes played.
1.2. Per 100 Possessions
What it is: It adjusts stats to what they would look like after playing 100 possessions.
Why we use it: It is an important variable in many of the more complicated statistics, and big numbers are fun to look at. Additionally, it gives us an equal baseline since the league average for possessions differs per conference. In the 2020 Philippine Cup, teams averaged roughly 89.8 possessions per game; in the 2019 edition of the same conference, teams averaged 93.1.
Limitations: The only limitation is that there’s no guarantee that the rate in which points (or any other stat) are acquired will remain the same in more or less minutes played.
2. Player Stats
2.1. True Shooting Percentage (TS%)
What it is: A measure of total shooting efficiency of a player taking into account the cumulative value of field goals, three-pointers, and free throws.
Why you use it: It is more accurate than field goal percentage (FG%) in terms of looking at scoring efficiency.
Example: Barangay Ginebra’s LA Tenorio shot 42.3% from the field in the 2020 Philippine Cup, while Meralco’s Raymond Almazan shot 52.8%. If basing only on FG%, the conclusion is Almazan was more efficient. But if we add free throws and three-pointers, Almazan had a TS% of 54.3% while Tenorio had a TS% of 56.6%, which means LA was more efficient.
Limitations: TS% does not account for role or quality/difficulty of shots. For example, Alaska’s Rodney Brondial has a 67.8 TS%, but the shots he took were easier than Kiefer Ravena’s, who had a 60.6 TS%.
2.2. Effective Field Goal Percentage (eFG%)
What it is: An improved version of FG% that takes into account the additional value of three-point shots, but does not consider FTs.
Why you use it: eFG% may be more useful than TS% in some situations,l as it strictly talks about field goals. It is a leap from FG% because it acknowledges that three-pointers are 50% more valuable than two-pointers.
Example: TNT’s Poy Erram shot 51.3% from the field, while Magnolia’s Paul Lee shot 44.7%. But Erram made 0.5 threes per game, while Lee made 3.0 threes per game. After taking account of the additional value from three-pointers, we can see that Lee was more efficient than Erram from the field; Lee had a 55.9 eFG% while Erram had a 54.1 eFG%.
Limitations: It does not consider free-throw proficiency, which means it doesn’t capture the complete picture in terms of efficiency.
2.3. Relative TS% and Relative eFG% (rTS% and rEFG%)
What it is: A stat that compares TS% and eFG% of a player to the league average.
Why you use it: A player is only efficient relative to other players. If a player shoots at 55 TS% while the league shoots at an average rate of 60 TS%, he is relatively inefficient. A player’s TS% and eFG% should be looked at relative to the players he competes against.
Example: Dallas’ Luka Doncic had a 58.7 TS% in the 2020-21 NBA season. In the 2017-18 PBA season, GlobalPort’s Stanley Pringle had a 54.7 TS%. This doesn’t exactly mean that Pringle was more inefficient because he has to be judged by the standards of the context he plays in (and vice versa). The average TS% in the NBA in 2021 was 57.2%. For the PBA, it was 51.3% in 2018. This means that Luka had an rTS% of +1.5, while Pringle had an rTS% of +3.4. This means that relative to the competition, Pringle was “more efficient”.
The same concept is applied for Relative eFG%.
2.4. Offensive, Defensive, and Total Rebounding Percentage (OREB%, DREB%, and TREB%)
What it is: An estimate of the amount of rebounds a player grabs out of all possible rebounding opportunities when he’s on the floor.
Why you use it: Per game rebounding stats can be misleading due to the amount of minutes played and the pace that a team plays at.
Example: In the 2019 Philippine Cup, both Rain or Shine’s Beau Belga and Ginebra’s Japeth Aguilar averaged 5.3 defensive rebounds per game. But Ginebra averaged roughly three more possessions per game than Rain or Shine, and Belga played two minutes less than Aguilar. These factors are taken into consideration in DREB%. Using that stat, we can see that Belga grabbed 19.2% of all defensive rebounds when he was playing while Aguilar only grabbed 17.4% of the same.
2.5. Assist Percentage (AST%)
What it is: An estimate of the percentage of made shots a player has assisted on for his teammates while he was on the floor.
Why you use it: Similar to TREB%, AST% is unaffected by minutes and pace. It is a more accurate representation of assist generation.
Example: In the 2019 PBA Commissioner’s Cup, Columbian’s Rashawn McCarthy averaged 6.2 assists per game in 37.5 minutes. Alaska’s Chris Banchero averaged 5.5 assists per game in 33.4 minutes. But McCarthy’s Columbian played at the league’s highest pace, while Banchero’s Alaska played at the second-slowest pace. As a result, Banchero had a 29.8 AST% and McCarthy had a 24.3 AST%. Using AST%, we see that Banchero is actually assisting his teammates more than McCarthy when he’s on the floor.
Limitations: It has all the limitations of the assist. This statistic cannot distinguish a player’s role, the types of passes he makes, and the ability of his teammates.
2.6. Steal and Block Percentage (STL% and BLK%)
What it is: Steal percentage is an estimate of the percentage of an opponent’s possessions that a player steals when he’s on the court, while block percentage is an estimate of two-point shots that a player blocks when he’s on the court.
Why you use it: Similar to the prior stats, it is unaffected by minutes and pace.
Example: In the 2019 PBA Governor’s Cup, NLEX played at the league’s fastest pace, while Rain or Shine played at the second-slowest pace. Beau Belga averaged a steal per game for ROS while Kiefer Ravena averaged 1.6 steals. Accounting for the difference in pace and minutes, we find that Belga had a 2.6 STL% while Ravena had a 2.5 STL%. Even if the per game stat shows that Kiefer has more steals, Beau actually steals a higher percentage when he’s on the floor.
2.7. Turnover Percentage (TO%)
What it is: An estimate of the percentage of turnovers committed by a player during possessions where he is directly involved in the end result — in made or missed shots, free throw attempts, and turnovers.
Why you use it: It factors in the amount of shots and possessions a player uses in determining what percentage results into turnovers. A player may have more turnovers, but it can be because he is asked to do more on offense.
Example: In the 2019 PBA Philippine Cup, Phoenix’s Matthew Wright averaged 2.8 turnovers per game and Meralco’s Chris Newsome averaged only 2.1 turnovers. But by using TO%, we can see that Wright had a 13.1 TO% compared to Newsome’s 13.9 TO%. The disparity here is because Wright used more possessions owing to a much bigger scoring role with Phoenix than Newsome had with Meralco. On a per possession basis, Wright seems to have handled the ball much safer than Chris.
Limitations: It doesn’t take into account a player’s role as a passer, and it presents an incomplete picture of the amount of possessions a player is involved in, as it only looks at turnovers, shots, and free throws. It undersells the ability of playmakers to keep the ball safe.
2.8. Usage Rate (USG%)
What it is: An estimate of the percentage of a team’s total possessions that ends up in a turnover, a shot, or free throws by a player.
Why you use it: It reflects the scoring load that a player takes on his team.
Example: In the 2019 PBA Philippine Cup, Rain or Shine’s James Yap averaged 13.5 points per game and 2.1 turnovers per game in 26.2 minutes. In the same conference, Magnolia’s Paul Lee averaged 13.4 points per game and 2.6 turnovers per game in 27.5 minutes. At first blush, it looks like Paul Lee carried a greater offensive burden. But after taking a team’s pace and individual minutes into consideration by computing their usage rates, we find out that James Yap took a slightly bigger scoring role based on the metric. Lee’s usage rate was 25.3% while Yap’s was 27.0%. This means that 27% of all of ROS’ possessions with Yap on the floor ended in a turnover, a field goal attempt, or a free throw attempt by Yap.
Limitations: It doesn’t take into account a player’s role as a passer, and it presents an incomplete picture of the amount of possessions a player is involved in as it only looks at turnovers, shots, and free throws. It doesn’t paint an accurate picture of the role of a playmaker.
2.9. Offensive Load (OffLoad)
What it is: It is a metric developed by Ben Taylor of Backpicks.com and Thinking Basketball which aims to measure how much a player is directly involved in offense. It attempts to fix the limitations of usage rate by filling in the missing variables, like passing and creation. You can read more about it here:
Why you use it: It provides a more complete picture of a player’s role in an offense, as it incorporates a playmaking component that is missing in usage rate.
Example: In the 2019 PBA Philippine Cup, Magnolia’s Paul Lee had a 25.3% usage rate and Jio Jalalon had a 24.7% usage rate. This could lead us to the conclusion that Lee had a bigger role in the Hotshots offense. But taking their playmaking into consideration, we find that Lee had an Offensive Load of 39.0% while Jalalon had an Offensive Load of 41.2% which shows that usage rate undersells his role on offense by not incorporating his playmaking.
2.10. Adjusted Turnover Percentage (aTO%)
What it is: It is a metric developed by Ben Taylor of Backpicks.com and Thinking Basketball aiming to “fix” the limitation of turnover percentage (TO%) by incorporating a player’s playmaking into the calculation. Read more about it on: https://backpicks.com/2017/10/16/offensive-load-and-adjusted-tov/
Why you use it: Playmakers who don’t score a ton are underrated by TO% because it only looks at turnovers and scoring possessions. It doesn’t accurately reflect passing and assisting. This stat aims to correct it.
Example: In the 2020 PBA Philippine Cup, Chris Ross had a 19.0% turnover percentage while his teammate Alex Cabagnot had a 13.1% turnover percentage. The six percentage point difference is massive and somewhat paints Ross as a careless ball handler. Using adjusted turnover percentage, the difference between the two is reduced. Ross has a 13.2 aTO% while Cabagnot has a 10.1 aTO%. This reduces the gap between the two and more accurately represents Chris Ross.
Limitations: The adjustment that benefits playmakers might still be too small, as some common scoring actions are less likely to result in turnovers compared to playmaking and orchestrating offenses.
2.11. Steal-to-Foul and Block-to-Foul Ratio (STL/FL and BLK/FL)
What it is: A figure of steals or blocks a player records per foul. It’s a measurement of how well a player can get defensive counting stats without fouling.
Why you use it: It provides an additional level of context to defensive counting stats. Being able to block shots or steal possession without fouling is pretty valuable. A player can block a ton of shots, but if he’s constantly in foul trouble, his rim protection doesn’t help the team as much as raw counting stats would suggest.
Example: In the 2020 PBA Philippine Cup, Phoenix’s Justin Chua averaged 1.6 blocks per game while Ginbera’s Japeth Aguilar averaged 1.4 blocks per game. But Aguilar had a higher block to foul ratio at 0.75 compared to Chua’s 0.44.
Limitations: Right now, it is impossible to determine which fouls were committed while attempting to steal the ball or block a shot, which makes the statistic less accurate.
2.12. Individual Offensive Rating (ORTG)
What it is: It represents the amount of points a player would generate over 100 possessions (league average hovered at around 89.8 in 2020, for context). It takes into consideration scoring, playmaking, and offensive rebounds.
Why you use it: It is not recommended to use for an individual player, but it’s an input for better metrics. It’s best used in comparing players with similar roles because it is easier to maintain a higher offensive rating when a player has a smaller role on offense.
Example: You can use it to compare Ginebra’s LA Tenorio (118.8 ORTG) and NLEX’s Kiefer Ravena (119.8 ORTG), but not with Raul Soyud (147.3 ORTG) who is a big that’s reliant on an easier diet of shots.
Limitations: It rewards highly efficient and low usage players more than other types of players. It is not uncommon to see bigs who live off of easy layups and putbacks at the top of the list.
2.13. Individual Defensive Rating (DRTG)
What it is: It is an estimate of how many points a player allowed per 100 possessions he individually faced while on the court.
Why you use it: It is not recommended to use for an individual player, but it’s a good variable used for better metrics.
Limitations: It’s closely anchored to a team’s defensive rating, meaning a good defensive player can look bad because his team is bad. It also relies a lot on counting stats, which aren’t always representative of defensive impact. A player with low steals might be seen as worse than he actually is. This is an estimate and does not actually reflect how a team did when he was on the floor.
2.14. Win Shares (Offensive and Defensive Win Shares) and Win Shares Per 48 Minutes (WS, OWS/DWS, WS/48)
What it is: It is an estimate of a player’s contribution to his team’s total wins. Offensive Win Shares is the offensive component and Defensive Win Shares is the defensive component. Win Shares Per 48 Minutes is an estimate of how much a player contributes to winning in a span of 48 minutes or one game.
Why you use it: It uses a ton of math derived by Justin Kubatko to provide an attempt to answer the question on individual contribution. It is generally a good statistic for offensive production, but like all statistics, it is pretty limited defensively due to the difficulty to quantify defense.
Example: In the 2020 PBA Philippine Cup, TNT’s Ray Parks led the PBA in Win Shares and Offensive Win Shares. There is a good chance that he was the best performer during that specific conference.
MPORTANT: THESE METRICS SHOULD NEVER BE USED AS A SUBSTITUTE FOR WATCHING THE GAME. FILM AND STATS GO HAND IN HAND. CLAIMING A PLAYER IS THE BEST BASED ON STATS MAY LEAD TO THE RIGHT CONCLUSION, BUT USES THE WRONG PROCESS TO ACHIEVE THE CONCLUSION.
Limitations: It utilizes individual Offensive and Defensive Ratings as components, which means the limitations of those metrics similarly apply. Although it has done a good job of limiting the weaknesses of individual offensive rating in its computation, defensive metrics should always be taken with a grain of salt due to the inherent difficulty of measuring defense.
Additionally, all these stats lean towards players on good teams. Though it is true that good teams have good players, some players may experience bumps to their ratings when on better teams even if they aren’t necessarily better in reality. Also, the PBA’s conference model doesn’t provide a great sample size that would lead to more reliable results.
2.15. Box Plus/Minus (Offensive Box Plus/Minus and Defensive Box Plus/Minus) and Value Over Replacement Player (BPM, OBPM/DBPM, and VORP)
What it is: Box Plus/Minus is a purely box-score based metric that attempts to measure a player’s contribution while on the floor. Offensive Box Plus/Minus is the component of BPM that measures offense, while Defensive Box Plus/Minus measures defense.
VORP takes BPM and converts it into a scale wherein a replacement-level player (a non-rotation player) is equivalent to zero. A value above zero means a player is above replacement level and a value below zero means that a player is below replacement level.
Why you use it: Among the statistics here, this is probably the most accurate representation of offensive contribution. It is pretty good at measuring offense, but it is a bit limited since it is a box score metric and doesn’t incorporate actual plus/minus stats or play-by-play stats. The defensive component is a useful guide because it balances a player’s contribution to how well his team plays. For example, it attempts to see if a player who blocks shots well is actually defensively impactful.
Limitations: IT IS NOT GOOD TO ARGUE THAT A PLAYER IS BETTER THAN ANY OTHER PLAYER SOLELY BECAUSE OF THE METRIC. It is highly recommended to watch the film and use the numbers as a guide. Role, team context, and shooting streaks play a huge role in the computation of the metric and these must be considered in interpreting the data. DBPM should be taken as a guide and not as a holy grail for defense. If a player is commonly known as a good defender or if the film doesn’t match the metric, feel free to ignore DBPM.
It’s also important to note that the coefficients used in the metric are designed for the NBA and may not be 100% accurate to the situation in the PBA.
Feel free to read more about it here, https://www.basketball-reference.com/about/bpm2.html, and here, https://hackastat.eu/en/learn-a-stat-box-plus-minus-and-vorp/.
2.16. Game Score
What it is: A measure of a player’s single game productivity created by John Hollinger.
Why you use it: It’s a pretty straight forward metric that is interpreted the same way we do with points. A 40 in Game Score is outstanding, 10 is decent, and so on and so forth.
Limitations: It isn’t really the most complicated or advanced metric, but it works in a pinch.
2.17. Linear Player Efficiency Rating (PER)
What it is: It is a linear version of John Hollinger’s Player Efficiency Rating created by Zach Fein. PER is Hollinger’s attempt at an all-in-one metric that measures a player’s contributions in one number.
Why you use it: Linear PER is much easier to compute than PER and provides results that are very close to the real deal.
Limitations: One number metrics should be taken with a grain of salt because it’s difficult and borderline impossible to condense basketball into one number. Some of PER’s flaws include failing to account for defense properly (it is hard to account for defense in general, but being a one-number metric, its incorporation of defense leaves a lot to be desired), it favors high volume shooters, and over-emphasizes rebounding.
2.18. Box PIPM (PIBOX + OPIBOX and DPIBOX)
What it is: It is the box score component of Player Impact Plus-Minus (PIPM) created by Jacob Goldstein. It is an incomplete measurement because there are steps to manipulate this data to come up with PIPM, but I thought it would be fun to include it just to have a foil or a second opinion in comparison to BPM.
Limitations: Do not use this as anything more than a guide. It is an incomplete data point because we do not have the necessary information to compute for PIPM in its complete form. It’s just a fun little number that I wanted to compute for.
2.20. Box Creation (BoxCr)
What it is: It is a metric created by Ben Taylor of Backpicks.com and Thinking Basketball that estimates how many opportunities a player creates for his teammates based on his passing, playmaking, scoring, and shooting skills.
Why you use it: It attempts to provide a more complete view of playmaking and tries to eliminate the effect of idle passes that lead to assists. It tries to see how a player’s scoring and playmaking ability comes together to help their teammates.
Example: In the 2020 PBA Philippine Cup, NLEX’s Kiefer Ravena averaged 7.8 assists per 100 possessions. Ginebra’s Scottie Thompson was second in the league with 8.9 assists per 100 possessions. By simply using assists, one would think that Thompson creates more opportunities for his teammates than Kiefer. Computing for their Box Creation would result in a total of 9.4 opportunities created per 100 for Kiefer and just 5.2 opportunities created per 100 for Scottie.
Even if Scottie’s assists are higher, once you factor in Kiefer’s superior ability to shoot and score, we can see that he opens the game better for his teammates.
Limitations: Box creation is an estimate of opportunities created and not the actual count.
2.19. Passer Rating (PasRate)
What it is: it is a measure of passing efficiency graded on a 1-10 scale.
Why you use it: Assists and assist percentage don’t really tell of the quality of passing, and passing itself is not purely isolated in these numbers. The Assist-to-turnover ratio isn’t really a fair comparison of passing efficiency as well. This stat, inspired by the Passer Rating metric by Ben Taylor of Thinking Basketball and Backpicks.com and the one created by CraftedNBA.com, attempts to grade players’ passing ability and isolates it from other factors like scoring and shooting ability.
Limitations: This metric only tells you about how good a player is at making the type of passes that he attempts. A player with a high passer rating does not mean he can make every single type of pass in the book or that he can pass as well in a different role. Also, this is not the exact formula Ben Taylor uses, because layup assist percentage is not available here.
For example, NLEX’s JR Quinahan had a 6.6 Passer Rating in 2020, while LA Tenorio had a 6.7 Passer Rating. This does not mean that Quinahan should be running the pick and roll as a ball handler.
You can read more about Passer Rating here: https://backpicks.com/2018/07/15/nba-passer-ratings-since-1978/
2.21. Passing Production (PasPro)
What it is: it is a measure to use alongside Passer Rating. Passer Rating measures passing efficiency, while Passing Production measures passing volume. This metric is taken from Ben Taylor of Thinking Basketball and Backpicks.com.
Why you use it: It attempts to measure the productivity of a player’s passes and rates it on a roughly 1-10 scale.
Example: NLEX’s Jericho Cruz had a Passer Rating of 5.7 in 2020, while Rain or Shine’s Gabe Norwood had a Passer Rating of 6.1. This does not mean that Cruz is a worse passer because he makes up for the efficiency deficit with volume. Cruz has a Passing Production of 6.3 compared to Norwood’s 2.5 which shows that he is a more productive passer while only being a bit less efficient.
Limitations: It makes use of estimates as parts of its formula, which means that it is not an exact measurement.
2.22. Gunner Rating (Gunner)
What it is: It is the ratio of True Shot Attempts (FGA and FTA) to the amount of assists per game. It looks to see how many times a player tries to shoot compared to how many times they get an assist.
Why you use it: It provides a somewhat clear picture of whether or not a player is pass-first or not.
Limitations: It doesn’t really take into consideration all passes a player makes because it only looks at assists. It also doesn’t distinguish between a player’s role.
For example, a player whose role is strictly to finish plays (spot-up shooter or roll-and-cut bigs) will look like a ball hog next to players who handle the ball more. Japeth Aguilar had a rating of 10.2 in 2020, while LA Tenorio had a rating of 1.9. This doesn’t mean that Japeth hogs the ball, but rather, his role requires minimal playmaking which inflates this number for him.
What it is: It is an attempt to capture a player’s effect on his team’s spacing. It is inspired by CraftedNBA.com and NBA Underground’s version of the metric. The farther the number is from zero, the better the rating is.
Why you use it: If nothing else, this is a good stat to look at to see the intersection between a player’s shooting volume and efficiency.
Example: A good example would be Ginebra’s Prince Caperal and Blackwater’s KG Canaleta. In the 2020 PBA Philippine Cup, Canaleta shot 34.7% from deep on 9.5 attempts per 36 minutes (pace adjusted). Caperal shot 42.0% from deep on 5.6 attempts per 36 minutes (pace adjusted). Canaleta had a Spacing rating of 6.1, while Caperal’s was 5.2.
Although Caperal shot better, Canaleta shot more at an above-average clip. This additional volume makes his shot more dangerous for defenses. Defenses won’t need to guard a player or won’t be bent if the player chooses not to shoot more often than not.
Limitations: The exact amount by which volume and efficiency should be weighed is not as precise as I would want it to be. The metric is still subject to future change.
3. Team Stats
3.1. Possessions (POSS)
What it is: It estimates the total amount of possessions used by a team per game.
Why you use it: This accounts for the varying playstyles of various teams and allows us to adjust for them.
Limitations: Play-by-play data is needed to come up with the true amount of possessions in a game. This stat is just an estimate.
3.2. Points Per Shot (PPS)
What it is: It is the amount of points scored by a team per shot on average.
Why you use it: It is a measure of efficiency, and this is a baseline for what types of shots would be valuable.
Limitations: It does not account for turnovers.
3.3. Points Per Possession (PPP)
What it is: It is the amount of points a team scores in one possession on average and accounts for both scoring attempts and turnovers.
Why you use it: It is a measure of efficiency and this is a baseline for what types of shots would be valuable and forms the basis of Team Offensive Rating.
Limitations: It is based on an estimate of possession and not the total amount.
3.4. Team Offensive Rating (ORTG)
What it is: It is the amount of points a team scores per 100 possessions.
Why you use it: It is the best measurement of how efficient a team’s offense is. Raw point totals fail to account for pace of play and may lead to wrong conclusions.
3.5. Team Defensive Rating (DRTG)
What it is: It is the amount of points a team allows per 100 possessions.
Why you use it: It is the best measurement of how efficient a team’s defense is. Raw point totals fail to account for pace of play and may lead to wrong conclusions.
3.6. Net Rating (NetRtg)
What it is: It is the difference between a team’s Offensive Rating and Defensive Rating.
Why you use it: It shows the average margin of victory of a team and it is a good measure for a team’s overall ability to execute on both ends.