Points per game is an outdated statistic. We used to use points per game to determine which NBA teams had the best offense in the league. We know better than that now. Now, Offensive Rating is the universally used stat to measure how good a team is at offense. Offensive Rating measures how many points a team scores per 100 possessions. This makes sense because some teams play at a faster pace than others. Scoring more points on more opportunities doesn't mean my team had a better offense. It simply means my team had more chances to score the ball. Since there are about 100 possessions in an average NBA game, it makes perfect sense to normalize Offensive Rating to per 100 possessions. We know that points per game isn't what we should be using to measure how good a team's offense. So why do we use points per game to measure how good an individual player is at offense?
I don't look at per game statistics when evaluating players. Instead, I look at per possession numbers. The same logic applies that we use with teams, except with players it makes even more since. Every team plays the same number of minutes in a regular game. The same can't be said for every player. Different players play a different number of minutes on a per game basis. When you add that to the fact that individual teams play faster or slower, per game statistics don't make nearly as much sense as per possession statistics. The chart above shows the top 10 scorers from last season in terms of per 75 possessions. I use 75 possessions because that is about how many possessions a normal star plays in a game. The other stats are rebounds per 75, assists per 75, and real True Shooting percentage. True Shooting is the best measure of efficiency, and real True Shooting percentage measures your efficiency against league average (just know that the higher number means more efficient). Do any of the numbers above jump out at you? Try to look at the per possession numbers and guess which player fits in which row.
This is the same table with the names now included. Yep, there's Wilt Chamberlain. And that's not just any Wilt Chamberlain season, that's his famous 50 PPG and 25 RPG season. Not only is it not number one in terms of per possession scoring from last season, but he would've been seventh. He finds himself right behind second year Trae Young. His per possession rebounding is lower than Giannis and barely higher than Embiid. His assist numbers are embarrassing. Does this change your opinion on Wilt Chamberlain? Chamberlain’s 50 point per game season is universally seen as the best offensive season ever. I would argue it isn't in the top 50 of all time. I would argue the season's Giannis and Harden had last season were far better.
So why are Wilt's numbers so inflated? The answer lies in the number of possessions. The 1961-1962 Philadelphia Warriors averaged 131.1 possession per game. For comparison, the league leading Milwaukee Bucks averaged 105.1 possessions per game. Not only did Wilt have way more opportunities because of the possessions, but he also played in every minute of every game that season. For comparison, Giannis only played 30.4 minutes per game last season. Don't get me wrong, it's very impressive for someone to play that many minutes. One of the flaws of per possession numbers is that it doesn't account for volume or fatigue. Giannis and Harden probably wouldn't be able to sustain their level of scoring for a full 131 possessions. However, the game was also different back then. There was far less running on a possession-to-possession basis. There were also so many possessions that you didn't have to hustle up and down the floor every play. The amount of minutes Wilt played is impressive, but you also have to consider the style of play back then. Just for fun, here are all of the current day players adjusted for Wilt's era:
The point of this article is not to bash Wilt Chamberlain. While I do think Wilt is arguably the most overrated player of all time, that is an article for another day (which will most certainly have a link to this article in it). The point was to bash per game statistics. We know that we shouldn't use them when evaluating teams. It makes even less sense to do it when evaluating players. I almost never look at per game stats for a reason. Even when we just look at the per 75 possession scoring rankings, they are very different than the per game ones. Giannis actually leaps Harden for the number one spot. I am not saying Harden is an overrated scorer, but he definitely isn't by far and away the best scorer in the league. Names like Giannis, Luka, and Kawhi shoot up the rankings when you look at scoring per possession (rightfully so). I like to adjust my numbers to per 75 possessions but adjusting to any number of possessions should do the trick. Hopefully, next time you see that player X averaged Y number of points per game, you'll think twice about what that actually means.
Comments