In 1966 Wilt Chamberlain attempted 25.2 field goals per game. He led the 76ers in True Shooting% by a comfortable margin. In other words, Wilt was the team’s most efficient scorer and he shot it the most. But did he shoot too much? This idea might seem counterintuitive; Wouldn’t a team want its most efficient player to shoot more, not less?
Surprisingly, less can be better.
Based largely on Braess’s Paradox, there is an excellent paper by Brian Skinner called “The Price of Anarchy in Basketball” that explains how this is possible. (Here’s an intersting blog post which might be easier to ingest.) In Chamberlain’s case, his scoring attempts represented one possible “path” for his team to score. In 1966, he converted those attempts at 54.7 TS%.
But there are other “paths” the team followed to score. Sometimes Chamberlain twisted, turned and passed. Sometimes, he didn’t even touch the ball before a teammate shot. For all the “paths” that didn’t end with a Wilt attempt, the 76ers team scored at a rate of 47.1 TS%.
Instinctively, one might ask why doesn’t Chamberlain shoot more since he converts so much more frequently than his teammates? Perhaps he should shoot every time and his team’s TS% would gravitate toward 54.7%? I hope it’s intuitively obvious why that’s not a good idea and why we’ve never seen something like that work at basketball levels above Lisa Leslie’s high school games. (Efficiently declines as usage increases, and practically speaking, the other team might adjust by double or triple-teaming if it were working.)
Indeed, it’s possible for Chamberlain to actually shoot less and have his team’s overall efficiency improve. If he looks to pass the ball more and aid his teammates in scoring (or even be used as a distraction off the ball), he can shoot a lot less — even shooting at the same or lower (!) efficiency — allowing his teammates to be more efficient, leading to an overall increase in team efficiency.
After all, the goal is to optimize team performance, not an individual’s, and it turns out at high levels of basketball this can rarely be done with an individual shooting too much. In the case of Wilt’s 76ers, the optimal offensive balance was one in which Wilt shot less. Significantly less.
In 1967,* Chamberlain shot the ball eleven fewer times per game. This, after several record-setting seasons in which he shot at least 25 times per game in each season. The result? The league’s best offense by a landslide — using the simple method of pace estimation, 6.7 points better than average — and the highest rated offense to that point in league history.
*The 66 to 67 Phily team is a great example of this phenomenon due to little team turnover and retaining the same core group of players, one through six.
His teammates scored at 50.7% efficiency. Wilt’s TS% went up as well. But, here’s the most interesting wrinkle: Even if Chamberlain’s TS% had remained the same, the overall team efficiency would have gone from 49.0% to 51.4%, and at well over 100 “attempts” per game, that results in a colossal shift in scoring efficiency. (As it were, the team’s TS% increased to 52.8% because Chamberlain’s TS% also increased.)
In 30 NBA player seasons has someone attempted more than 25 FGA’s per game. Only one of them, point guard Tiny Archibald in 1973 (who also led the league in assists), played on a league-best offense. That doesn’t mean offensive machines like McAdoo, Gervin, Jordan, Bryant and even the oft-maligned Allen Iverson don’t significantly help a team’s offense in high-volume roles. It just means the best offenses in NBA history are never too focused on an individual shooting the bulk of the attempts. And sometimes, as in the case with Wilt’s Sixers, a team’s optimal approach can be to increase everyone else’s role and decrease the superstar’s.
Most of those players carried offenses with a good deal of success. Some were even elite. But as impressive as 25 points per 75 possessions on positive TS% is, it can also be less optimal for the team if it would perform better with that player taking fewer shots. The more evenly distributed pathway is often the better one.
Practically speaking, it’s easier to defend an individual than a balanced attack of five players.