Welcome back to Stats for Dummies! Let’s continue our journey to discover the differences between traditional and advanced statistics: today we switch to the individual ones, the Ratings.
When you want to compare the offensive or defensive abilities of two or more players, what do you look at? Limiting only to the points scored does not guarantee a clear view of the player’s offensive contribution to the team: a player can be useful for the assists he distributes, for the offensive rebounds he grabs and he can also make negative contributions by losing balls or making a lot of missed shots.
A similar argument can be done for the defensive phase: looking only at steals does not provide a complete picture, we have to take into account other factors such as rebounds, blocks, fouls, contested shots and forced lost balls. Therefore, comparing the offensive and defensive efficiency of the players is complicated: there are so many variables that must necessarily be considered together to understand how a player behaves on the court.
But there is a solution! Obviously, the Ratings: as the team ones, they are divided into Offensive Rating for the attack and Defensive Rating for the defense. Anyway, their calculation is much more complicated but allows us to obtain two simple numbers that contain all the contributions just mentioned.
So do we have a complete view of the player’s efficiency with the ratings? Well, no. There are the so-called intangible contributions (or simply intangibles): in this group are included all those contributions that are not normally computed in the classic box-score. Some examples can be the screens made for the teammates or the boxouts that guarantee a rebound to the teammates; in other words, they are the efforts made by a player who guarantees an advantage to their team, but that remain invisible or, in fact, intangible to the boxscore. We will be back soon on the intangibles because in the NBA they have become for the most part tangibles because there are statistics related to these. It’s another world compared to Europe.
Back to the Ratings: find a number that can include all the contributions is virtually impossible, especially when the calculation starts from the box score. But, having said that, the individual ratings, born from the mind of Dean Oliver, are nevertheless a valuable tool for the analysis of players’efficiency. Knowing their limits it is possible to carry out exhaustive analysis, without running into some mistakes. Let’s take a close look at the Offensive Rating first.
With the Offensive Rating, we can know the points generated by one player using 100 possessions: the first advantage of this statistic is that it is calculated on a common basis, that is the famous 100 possessions. Comparing a group of players with different minutes played involves imprecision: a player who averages 30 minutes per game will have more opportunity to make shots, grabs rebounds, etc. By reshaping everything on 100 possessions, all players are compared on the same unit of measure, which makes the analysis more accurate. These possessions are individual: they can be calculated through some formulas from the box score, starting from some data:
- Field goals attempted;
- Offensive rebounds;
- Free throws attempted;
And what about the points generated? They are not simply the points made, but a sum of some contributions:
- Field goals made;
- Offensive rebounds;
- Free throws made;
Do not think it is a simple sum: all these contributions are joined together not before having compared them with the team performance: for example, the offensive contributions of Shved, known as a selfish player, will have a different weight than those of Jenkins, who is his teammate with other offensive duties.
The points generated are divided by the possessions used and then multiplied by 100, thus obtaining the Offensive Rating. The weaknesses behind this statistic are therefore some intangibles that can not be taken from the box score, such as screens. For example, in the last season for Utah Jazz, Mitchell was the best in Offensive Rating with 107.6, while Gobert was the second with 106.9.
However, this value does not take into account a crucial contribution of the French big man: Gobert averages 6 screens per game that allow the teammates to score. This contribution is invisible to the box score and, therefore, will not be included in the big man’s Offensive Rating. Probably then, if you count this thing too, Gobert would have a better offensive rating.
Through the Defensive Rating we take into account different players’defensive contributions: this statistic consists of two distinct parts. The first one takes into consideration the tangible contributions such as:
- Defensive rebounds;
While the second considers almost-intangibles. I write “almost” because through some approximations it is possible to get them from the normal box score:
- Opponent missed shots, but not blocked: the blocks are subtracted from the opponent’s missed shots and redistributed to each team member in proportion to the minutes played;
- Turnovers but not steal: the steals are subtracted from the opponent’s turnovers and redistributed to each team member in proportion to the minutes played;
- Fouls that bring the opponent to attempts a free throw;
All these contributions (contested shots, forced turnovers, fouls) are not computed in the classic box score, but, as you can see, you can find them anyway. The imperfection is that each player will receive a slice of those contributions based on playing time and not based on real efforts. It remains a good approximation, but certainly not perfect.
By combining the two parts through a final formula, the Defensive Rating is obtained, that is the points allowed every 100 possessions by a player. The lower is the value, the better will be the defensive contribution.
Even here, there are some gaps: for example, a deflection that becomes a teammate’s steal will never be counted. Or, again, a drawn foul with a charge. Remaining in Utah, Rubio averages 3 deflections per game, but his defensive rating is ranked third. Probably if these aspects were taken into consideration, there would be more differences between the players’defensive rating.
Defensive Rating also remains a reliable tool for analyzing a player’s defensive phase, even if it takes into account these approximations. When you read these values, especially in the European context, it is good practice to remember these limits and also rely on some eye-test to improve the analysis.