I will soon be posting the rankings of players based upon what is called box score metrics. This is an attempt by some experts to assign weights to various components of the game as reflected in a box score. Many of you remember this chart from the other message board. I will offer some definitions for those who will be viewing this chart for the first time.
John Hollinger’s Game Score (GS) formula: https://www.nbastuffer.com/analytics101/game-score/ In 2023, I chose to modify the pure Hollinger formula. With my modifications, a player’s score will be reduced if his two point shooting percentage is less than 50% or if his three point shooting percentage is less than 35.3%. Those numbers are close to the Division 1 averages. Along with the 2023 modification, I recalculated the results for the years prior to 2023. These numbers go back to the 2006-07 season.
David Berri’s original Win Score (WS) formula: https://www.nbastuffer.com/analytics101/win-score/ In 2011, in response to criticism that he was overvaluing rebounds, Berri reduced the weight of a defensive rebound to 0.5. The weight of an offensive rebound remained at 1.0.
The main differences between the GS and the WS are that the GS more heavily weighs points scored while the WS more heavily weighs rebounds and punishes inefficient shooters regardless of their scoring. The weights assigned to assists and turnovers are slightly different. Both agree that a turnover has more of a negative impact than an assist has a positive impact.
The Win Score and Game Score are both cumulative numbers. The more possessions you play, the higher your score. Each year I adjust (Adj) both raw numbers to reduce the scores of players on teams with higher than the league average numbers of possessions and I increase the scores of players on teams with lower than the league average numbers of possessions.
I also offer the WS and GS number on a per 40 minute basis. This may help identify players who were underutilized or may show the contribution of players who missed games as a result of injuries and could not accumulate the total minutes of those at the top of the list.
Shooting efficiency (eff.) is the portion of the GS formula which is based solely upon shooting and scoring. The Berri efficiency is the same for the WS. A players total GS or WS starts with the respective shooting efficiency score to which is added the results of the formula as calculated based upon rebounds, personal fouls, assists, turnovers, blocked shots and steals.
Effective field goal percentage formula: https://www.nbastuffer.com/analytics101/effective-field-goal-percentage-efg/ A three point field goal counts for 1.5 times the value of a two point field goal.
True shooting percentage formula: https://www.nbastuffer.com/analytics101/true-shooting-percentage/ This adds foul shooting to the calculation.
The Pure Point Rating (an improvement over the assist to turnover ratio): https://ilevy.wordpress.com/stats-glossary/ppr/
These numbers do not claim to identify who the “best” player in the league is or who should be first team all-conference. Rather, it is a way to identify those whose contributions to their team may have been underappreciated.
A REMINDER: Box score metrics can measure only those numbers which appear in a box score. There is no positive (or negative) score for (1) boxing out so that a teammate pulls down a rebound (or failing to do so), (2) going to the right spot on the floor and setting a good screen (or failing to do so), (3) spotting and executing a good entry pass to initiate the offense (or failing to do so) or for (4) the “hockey assist” (the pass before the pass that gets credit for the assist). Leadership qualities on the court or in the locker room do not appear in a box score. Most importantly, there is no box score number for playing good on-the-ball defense (or for failing to do so) beyond those defensive skills that are reflected in defensive rebounds, blocked shots or in steals.