In recent years, the game of basketball has changed a lot. The game is more heavy on three-pointers than it’s ever been and fans are concerned about the decline of shot types and locations that used to be much more prominent than they are today. The topic of how the game has changed in the past fifty to seventy years has always been very interesting to me. Therefore, I decided to dive into a little tracking exercise. I wanted to get a rough overview of how ball movement, player movement and shot locations changed over the course of NBA history, so I asked some people to give me random games to watch from each of the last four decades prior to shot data becoming available, and while watching these games, I tracked the following things:
1) Scoring frequency and efficiency by shot type and distance to get an idea where teams shot from and what kind of shots they took, with a special focus on assisted vs. self-created shots (isolation frequency).
2) A shot chart as an illustration of shot locations. These will obviously not be perfectly accurate and decrease in accuracy the further they are back in history due to worse footage and camerawork.
3) Passes per possession as a rough estimate of ball movement. It’s far from perfect, but I added my own notes to add more context to the numbers.
4) A subjective (!) scale from 1–5 to describe player movement within a possession. Context will also be added in the form of notes.
Disclaimers:
- None of these methods are perfectly accurate. They are not supposed to be more than rough estimates. More importantly, these one-game sample sizes do not represent a whole decade/era, which is why this exercise is pretty meaningless, but it’s interesting and (somewhat) fun, and sometimes that’s more important.
- I defined assisted shots as those where the shooter does not create any further advantage after catching the ball (e.g. if a fastbreaking player gets a long outlet passes, takes four dribbles and finishes an uncontested layup, I credited the passer with an assist because the scorer didn’t do anything after catching the ball to increase the likelihood of scoring. On the other hand, if a player catches a pass in the post, puts his shoulder into the opponent to create space and then scores without a dribble, he has helped his own scoring probability and the passer was not credited with an assist).
Los Angeles Lakers-Boston Celtics, April 24, 1963 (Game 6 of the Finals)
https://www.youtube.com/watch?v=aA8l1Jr7jwc
The discrepancy between assisted and self-created shots sticks out from the shot type data. Very similarly to the Raptors and Bulls in the ’90s game, the Celtics have a roughly even splits between assisted and self-created shots, while almost 80% of the Lakers’ shots were self-created. The offense of the Lakers revolved around Elgin Baylor and Jerry West, who were not only the teams’ best scorers but also its best playmakers. Therefore, they had to create a lot of their offense for themselves. Additionally, both were score-first players, and it wasn’t their primary focus to create open shots for others. For the Celtics, on the other hand, the offense ran through Bob Cousy, who made sure that his score-first teammates wouldn’t have to create for themselves as much.
While the Celtics’ offense ran through Cousy, it was designed to get everyone involved and it relied quite heavily on the fastbreak. In the halfcourt, there was quite a bit of off-ball movement and passing, but occasionally it also involved some station-to-station passing (moving the ball from one stationary player to another). The offense of the Lakers was slower and involved less ball/player movement due to the team being less balanced. This didn’t hinder its effectiveness, however, as the Lakers posted a superior offensive rating to the Celtics in eight of the nine seasons Jerry West and Bill Russell were both in the league.
If you look closely enough at the shot chart, you will also discover an important difference there. The Celtics managed to get way more shots directly at the rim. In his brilliant analytical breakdown of this game, Justin Jacobs called this the “Bill Russell effect”. Since the most efficient shots in basketball lie at the basket, having a rim deterrent like Bill Russell is/was incredibly valuable. On quite a few of the Lakers’ layup attempts, the offensive player actually jumped from 10+ feet and tried to finger roll it in because if he drove all the way to the basket, Bill Russell was more likely to impact the shot. I will attach my hand-tracked boxscore below because it shows quite nicely what happened when the Lakers challenged Russell.
Lastly, the midrange shooting numbers are interesting to look at even though it’s a meaningless one-game sample size. In total, the Celtics shot 22–54 (40.7%) on these jump shots and the Lakers made 23–57 (40.4%). Obviously, the league average on these shots could have been significantly lower (or even higher), but it’s still interesting that these are roughly the same numbers modern players shoot on midrange jump shots.
San Diego Clippers-Los Angeles Lakers, October 12, 1979
https://www.youtube.com/watch?v=FEexi0iSm2U
This was the first game of the 79/80 season, Magic Johnson’s NBA debut and for this exercise particularly interesting because it was the first NBA regular season game either of the teams played with the three-point line. Neither team actually attempted a three-pointer, but one shot was very close. A Cleveland guard dribbled the ball up the court directly into a pull-up from the top of the key with both feet on the three-point line … that’s something you won’t see in the modern NBA. The story of the game was definitely World B. Free, who had an unbelievable performance in terms of shot-making. He finished the game with 46 points on 33 shooting possessions and I am quite confident that he was single-handedly responsible for more than half of the 22 field goals the Clippers scored from midrange.
The offensive strategy of both teams differed greatly. The offense of the Lakers had some similarities to that of Cousy’s Celtics because they ran the fastbreak a lot, relied on their guard(s) for creation and also initiated a lot of actions in the post. In this game, the midrange shooting wasn’t working at all (8–31), but they managed to win the game by one point by taking advantage of the terrible inside and transition defense of the Clippers, who allowed the Lakers to convert 19 layups/dunks. The Clippers, on the other hand, played a slightly more modern game as they entered the ball into the post way less, shot a lot of jump shots, especially off screens at the elbows, and frequently let their guards create in isolation. They also used their ability to hit jump shots to open up driving lanes for their guards, which happened much less frequently in the ’60s and early ’70s due to the lack of spacing.
Philadelphia 76ers-Boston Celtics, April 5, 1987
The Celtics and Sixers also played two very different offensive styles, which is reflected in the data. The Sixers moved the ball quite a bit, looking for the open man, but resorted to more and more isolations by Sir Charles and Andrew Toney as the game entered its final stages. Both of them were keen on getting shots in the paint, and the team in general lacked a consistent shooting threat, which is one of the reason so little of their shots were taken outside of the paint. Another reason is Barkley’s offensive rebounding, which created about half a dozen open layups.
The Celtics also moved the ball a lot and stuck to that strategy until the end. They scored in a variety of ways including the post, where they frequently drew double teams and created open shots that way, with drive-and-kicks and even some down-screen-actions. They fielded more capable shooters than Sixers, hence the higher shot frequency from midrange. The shot chart might seem a bit right-dominant, and that’s the case because Bird preferred to operate in the right post, even though he didn’t have much success on shots from there in this game. Nevertheless, both teams’ styles deserve to be called attractive and that’s only enhanced by their star players. In this game, Barkley stood out. Just a fantastic athlete and a pure joy to watch (and very good defensively in this game).
Finally, the shot data on post shots stands out. Both teams attempted an unusual amount of fadeaways and didn’t have much luck on those, but the Celtics did really well on hook shots, which is largely down to Kevin McHale, who received a lot of praise from the commentary team (more post moves than anyone ever, should be in the MVP conversation, improved every year etc.).
Toronto Raptors-Chicago Bulls, March 24, 1996
https://www.youtube.com/watch?v=7nr1HtRNZVI
This game was also characterized by a clash between two entirely different offensive styles, which were eventually overshadowed by the brilliance of two individual players. The Raptors’ offense was, to put it nicely, uninspiring. It mostly consisted of Damon Stoudamire dribbling around, looking to find an open shot for himself or waiting for one of his teammates to get open. Seeing how little the other players were trusted to operate with the ball in their hands, helps you appreciate the plays that 8th men on modern teams, like Dorian Finney-Smith, can make on the ball.
The Raptors team was still fun to watch in this game, however, and that is fully down to the shot-making of Damon Stoudamire. He’s almost single-handedly responsible for the fact that the Raptors made 6 of their 8 unassisted three-point-attempts. Over and over again, he went around screens and drilled high-difficulty off-dribble shots, which is especially impressive if you consider that he was a rookie during this game. Other than that, their offense lacked ball movement, player movement or anything else that could have made it fun to watch.
Of course, the Bulls relied on the the “triangle offense”. The style is slow and methodical, involves much more ball movement than the offensive strategy of the Raptors, but also a lot of moving the ball from station to station. I personally don’t find it overly compelling to watch, but I’m sure there’s a lot of people that feel otherwise (For the Raptors, on the other hand, there’s no way anybody would find it attractive (if Stoudamire doesn’t go berserk)). Some Bulls didn’t have their greatest game so it remained close until the end. Then, Michael Jordan took over. When his team was in desperate need of a basket, he started hitting one tough shot after another. Stoudamire was awesome, but it was a MJ-fadeaway over two defenders that made my jaw drop. Jordan was such a fascinating spectacle that you almost feel thankful for being able to see it.
In terms of shot distribution, it’s important to add that the game had moved further outside than in past decades. The distribution between paint shots, midrangers and three-pointers wasn’t nearly as balanced in any of the other games.
Conclusions:
- The recent surge in three-point shooting might seem to have occurred suddenly, but it’s actually a development that has been going on for multiple decades. At first, the league greeted the new line with suspicion and a certain stubbornness because it was used to playing one way and didn’t want to change (on the 1987 broadcast, Tommy Heinsohn said, “He was too concerned with the three-point shot instead of just getting the good shot” when a Celtics player stepped out of bounds attempting to step behind the three-point line. Heinsohn also said that he appreciated Sir Charles’ ability to handle the ball, but preferred the guards to do it since that’s what they were getting paid for.) There was a preconceived notion of how the game should be played and it took another decade for the three-pointer to become more prevalent. Damon Stoudamire’s Lillard-esque off-dribble shot making was an indicator in which direction the league was headed.
- Another interesting element that would’ve been worth tracking is the defensive pressure on all of those shots. Unfortunately, that thought came to me too late. Judging by my eyes test, however, I would guess that there is a relatively high correlation between spacing and “openness” on shots. The incredible spacing of the modern game makes it much more likely (and easier) to create open shots than it was 60 years ago. Even in the ’90s game I watched, there were significantly more open layups and dunks than in Game 6 of the 1963 Finals.
- Passes per possession is not a great indicator of ball movement and it’s probably a measurement I wouldn’t use again. It works quite well when comparing two teams that are playing each other, but it’s not very valuable when comparing eras. In the ’60s, there were free throws for every single foul. This leads to a lot of zero-pass-possessions, which skews the “passes per possession” stat quite a bit.
This exercise was inspired by a conversation between Danny Leroux and Ben Taylor on episode 47 of the “Thinking Basketball” podcast.
