This statistic was invented by John Hollinger to provide a rough measure of a player's performance in a given game. The scale upon which the player's game score is based is the same as points scored. If a player has a game score of 40, that is amazing, while a game score of 10 is average.
The formula for calculating game score is as follows: (Points x 1.0) + (FGM x 0.4) + (FGA x -0.7) + ((FTA-FTM) x -0.4) + (OREB x 0.7) + (DREB x 0.3) + (STL x 1.0) + (AST x 0.7) + (BLK x 0.7) + (PF x -0.4) + (TO x -1.0).
As can be plainly seen, this statistic takes into account almost everything a player does during a game that can be quantified. These statistics are then weighted and added together to get a game score. This method is similar to Dr. Dean Oliver's Four Factors, but attempts to add value by considering all the stats, rather than just the four most important ones.