by Andrew on Sun Aug 01, 2021 8:25 pm
Per 36 tends to be a good way to normalise player ratings, and not underrate players' abilities just because they're playing fewer minutes, or overrate them because they're playing more minutes. For example, if someone is averaging close to a steal per game in less than 24 minutes, it's an indication that they're a pretty good pickpocket, and would likely be up there among the leaders if there were getting more court time. Conversely, if someone is averaging around the same amount of steals in over 30 minutes per game, they'd have a lower rating because their (for example) 0.8 steals isn't the same as the 0.8 steals averaged by the player playing less than half the game.
Of course, sample size needs to be taken into account, as a couple of games of unusually productive garbage team shouldn't equal superstar ratings due to Per 36 numbers. So context matters.