DPR – Dynamic Player Rating
DPR is a proprietary metric on a scale of 20 to 100 which measures 12 different statistics to assess a player’s production on a game-to-game basis. There are three different adaptations of the metric for point guards, wings, and big men.
Our DPR grading scale is the following:
95 or higher = ⭐
94-90 = A+
89-85 = A
84-80 = A-
79-75 = B+
74-70 = B
69-65 = B-
64-60 = C+
59-55 = C
54-50 = C-
49-45 = D+
44-40 = D
39-35 = D-
34 or below = F
Anything above 90 is in historic territory. It’s important to understand that it would be close to impossible to average anything above a 90 through a whole season—for example, Russell Westbrook’s 2016-17 season in which he averaged a triple-double came out to an 81.43—but in a single game, it has happened. Anything above 94 is a ⭐ for us. The grading scale is most relevant on a game to game basis, to see the league leaders in DPR this season click here.
There’s also no way to break or stump the metric, as it allows for games above 100, which has happened in truly historically great performances. The most notable being Kobe Bryant’s 81-point game that produced a 121.2 DPR.
This season we’ve had a few such games, including DeMarcus Cousins’ 43-point triple-double, with 23 rebounds, 10 assists, and 4 steals—for a DPR of 109.2. James Harden had one such game as well, going off for 60 points, 11 assists, and 10 rebounds—for a 111.3 DPR.
The rating is divided up into three different variations, one for point guards, wings, and big men. The stat takes into account Points, Game Score, Offensive Rating, Defensive Rating, Usage Percentage, and Effective Field Goal Percentage.
Additionally, each position has a slightly different weight for +/-, Free Throws Made, Assists, Rebounds, Blocks, Steals, Turnovers, 3-pointers made, and a bonus for a lack of fouls.
Every statistic has a specific weight with added minimum and maximum thresholds. Thus, a player is rewarded for everything good and bad that he does in a game. The metric rewards usage but also efficiency.
Note: our game logs have N/A and DNP. N/A indicates a player being inactive due to injury. DNP indicates that the player was active but was not put in the game by the coach, thus, he did not play.
BSN’s other in-house stats
A statistic used by the Euroleague, we wanted to bring it to the NBA. Blocks against is a per-game stat.
Blocks Against %
Blocks against % is the percentage of field goal attempts the player has had blocked.
Is a simple stat that can’t be found other places, this too is self-explanatory.
Other advanced stats we use
Player Efficency Rating (or PER)
Created by former ESPN columnist and current Memphis Grizzlies Vice President of Basketball Operations John Hollinger. It is explained here.
True Shooting %
True shooting percentage is a measure of shooting efficiency that accounts for field goals, 3-point field goals, and free throws. It is PTS/(2*True Shooting Attempts).
Win Shares (or WS)
The godfather of advanced statistics Bill James created this and it’s since been modified. In basic terms, it is similar to WAR in baseball.
Usage % (or USG %)
Estimates the percentage of plays run through a player accounting for field goal attempts, free throw attempts, minutes played, and turnovers while on the floor.
Effective Field Goal %
Is similar to True Shooting Percentage, however, it adjusts for 3-point field goals being worth an added point.
Is also a team stat but we’re using it for individual players. It is based on points allowed per 100 possessions.
Like defensive rating, it is based on points per 100 possessions, only its based off of points that a player contributes to producing on offense as opposed those he’s allowed.
Game Score is another Hollinger stat based on a per-game basis—PER is only overall and can’t be used on a game-to-game basis which is why DPR doesn’t take it into account. Like PER it is a composite stat, the formula is explained here.