Drew Fossen – Evolution of the NBA Three Point Shot

The Three Point Shot was introduced into the NBA in 1979. It took teams a while to get used to this new aspect of the game, and for the first eight years, the three pointer was almost irrelevant in the NBA. Average teams shot less than five Three Point attempts from 1979-1987, but as players began to figure out how to use the Three Point shot as a new weapon to stretch the floor, it began to take off.

The Three Point shot has become much more relevant in the NBA, as the number of attempts and made Three Pointers has increased nearly every year. We have seen an even steeper increase in Three Point attempts and makes during the last five years. A large part of this increase can be attributed to Steph Curry, Klay Thompson, and the Golden State Warriors, as they have been winning many games by shooting massive amounts of three pointers at high percentages, causing other teams to try and copy this winning playstyle.

After examining my data, I kept wondering “What went on in the 90’s that made shooting threes rise so suddenly?” My original thought was that the defense had fallen off from the physical defense of teams such as the “Bad Boys” Pistons in the early 90s and players were now less physical, resulting in an increase in general scoring. Actually this was not the case, instead, in the 1994 season, the NBA attempted to address decreased scoring by shortening the distance of the Three Point line from 23 feet, 9 inches with 22 feet at the corners to a uniform 22 feet around the basket. However, this change in the length of the Three Point line only lasted a mere three years. From the 1997 season on, the NBA reverted the line to its original distance of 23 feet, 9 inches with 22 feet at the corners (https://en.wikipedia.org/wiki/Three-point_field_goal).

The increase in Three Point attempts and makes has led to a decrease in Two Point attempts and makes. Both the Two Point shot and Three Point shot follow a fairly linear trend line, which I thought would be interesting to extend. Then we can ask the question: when will NBA players be shooting as many Three Point shots as Two Point shots?

If the trend continues as it has in the last 35 years, NBA teams will be attempting the same amount of Two and Three Point shots per game by the year 2048. They are predicted to be attempting approximately 43 of each shot type, making 30 Two Point shots per game and 16 Three Point shots per game. Analyzing these stats, we can see that in 2048 teams are expected to score 60 points off of Two Point shots, while only scoring 48 points off of Three Point shots, while taking the same amount of both. This data tells us that we cannot expect the Two Point shot to decrease or the Three Point shot to increase this rapidly over the next thirty years, because the trend suggests that as teams take more Three Point shots, they become less efficient, and the Two Point shots that they choose to take become more efficient. However, this is untrue because the NBA average for Three Point shooting and Two Point shooting consistently hovers around 35% and 48% year after year. A better way to analyze the trend of increased Three Point attempts and decreased Two Point attempts is to examine the expected value of each type of shot. You can do this simply by taking the number of shots, divided by the number of attempts, and multiply by the value of the shot. I found that in the year 2019, the expected value of the Two Point Shot ([33makes/60attempts]x2points=1.1) will be equivalent to the expected value of the Three Point Shot ([9.5makes/26attempts]x3points=1.1). This means that the increase in Three Pointers will likely come to a halt within the next ten years unless players become much more efficient shooters from behind the Three Point line. Now that we’ve taken a look at the future of the Three Point shot, let’s see how much of an impact the Three has on winning in the NBA today.

If you take a look at the last 15 years in the NBA for the Warriors, Spurs, 76ers and Knicks, you will see a close correlation with increases in wins and increases in Three Point shots made per game. The “Wins” and “Three Pointers Made per Game” lines for the Warriors basically mirror each other. The Warriors wins went from 47 in 2012 to 51 wins, 67 wins, and 73 wins, while their made Three Pointers went from 8 to 9.4 to 10.8 to 13.1 respectively. You could take these numbers as just coincidences of being a good team that scores points, but if you look at the other three teams, when there is a spike in wins one year, there is usually a spike in Three Pointers Made per Game. The exception to this is the 76ers, whose consistent increase in losses lately can be attributed to poor management of their team and roster. Sometimes teams shoot the Three Pointer collectively, spreading the ball around and not relying on just a few players to make Three Point shots. Other teams have players that consistently lead the NBA in Three Point shooting each year. You might wonder: Does having a player who leads the league in Three Point shooting lead to wins for your team?

In the last 12 years, there have been 20 teams to reach 60 wins in the regular season. Only 7 of these teams have not had a player in the top ten in Three Pointers made for that season. This means 65% of teams that have achieved 60 regular season wins have done so with a top ten Three Point shooter on their roster. Three of these teams have had multiple players in the top ten in Three Pointers made that season. The odds of having a top ten Three Point shooter on any given team is 33%, since each NBA team has 1/30th of the players in the league, and ten players are on this exclusive list ([1/30]x10=0.33). This further shows that making Three Pointers in the NBA is becoming part of the new winning formula.

The Golden State Warriors are the perfect example of how the Three Pointer is being used heavily in the game plan to lead to more victories. The Warriors’ Steph Curry and Klay Thompson have been first and second respectively in Three Pointers made each of the last three NBA seasons. The year before that, Curry was first and Thompson was third. The skillsets of these two players are so unique that it makes it almost impossible to replicate their stats or style of play. Curry is already considered the best NBA shooter of all time and he’s only in his 7th NBA season. He has won back to back MVP awards, showing that he’s not only limited to Three Point shooting. He hits a number of ridiculous layups, floaters, and midrange shots. To top it all off, he led the league in steals this year as well. His teammate in the backcourt, Klay Thompson, isn’t too bad either. Thompson is only in his 5th NBA season and is already a two-time all-star. He is also not limited to Three Point shots but is known around the NBA as a great defender. You just don’t see these types of skillsets in players very often, and the Warriors have two on the same team. Having top tier Three Point shooters certainly increases your chances of winning, but we cannot expect to see another team replicate the ridiculous sharpshooting combination that the Warriors have in Steph Curry and Klay Thompson.


As the data shows, Three Point makes and attempts are currently on the rise in the NBA. We can’t expect them to continue increasing at the current rate, but we can expect them to reach an equilibrium point where they will no longer increase so rapidly. The data suggests that this equilibrium point will occur sometime within the next ten years. Of course, any change in the game could alter the amount of Three Point shots like we saw in the 1994 season when the NBA decided to change the length of the Three Point line. If players become too efficient at shooting the Three Pointer, the NBA might decide to move the line back even further than the current 23ft 9in Three Point line. Although the data isn’t incredibly strong, it does support that there is some correlation between a team that increases their Three Point makes and a team that increases their wins. You cannot ignore the fact that teams who have achieved 60 wins in a season over the last 12 years have been twice more likely to roster a top ten Three Point shooter than any other team during this period. Some teams, including the Warriors, have even had multiple top ten Three Point shooters contribute to their 60 win seasons. The Three Pointer has become an increasingly large part of the NBA basketball game, and mastering how to use it and how often to use it is a challenge that leads to wins when understood correctly.


Average Shot Makes and Attempts Per Game

This visualization shows the number of Two Point shot attempts, Two Point shot makes, Three Point shot attempts, and Three Point shot makes. The data is displayed on a per game basis from each year. The yearly per game averages were calculated by compiling these yearly statistics from every team in the league, dividing by the number of teams in the league at the time, and dividing again by the number of games played. I collected the data from basketball-reference.com, where basketball stats are archived from every year of professional basketball in the United States.

I chose to include four lines on this visualization at the same time because I felt that I cannot just show how many times a shot type was attempted or made, rather I must show both how many were attempted and made. If I did not show both of these, and for example, only showed two lines, one displaying Two Point shot makes and one showing Three Point shot makes, I would not be telling the full story. I also think it still looks clean with four lines. I choose to color each line differently, but in two different sets of colors. I gave the Two Point shots cool colors (green and blue), while I gave the Three Point shots warm colors (red and orange). I did this because I wanted to differentiate each type of shot, but I also further wanted to differentiate between a make and an attempt of that shot type. The cool colors contrast the warm colors and allow the viewer to easily follow the visualization after looking at the legend. I chose to make the “shot attempted” categories compliments of each other because it makes sense logically. Green (Two Pointers attempted) compliments red (Three Point shots attempted), while blue (Two Point shots made) compliments orange (Three Point shots made). The colors are also pleasing to the eye and not overwhelming or confusing. This visualization served as background data to my next visualization, which follows the trends of the lines presented in this graph to predict of future of these shot types.

Predicted Average Shot Makes and Attempts Per Game

This visualization shows the predicted number of Two Point shot attempts, Two Point shot makes, Three Point shot attempts, and Three Point shot makes over the next 37 years. The data is based off of the same data that I used in the previous visualization. I collected the data from basketball-reference.com, where basketball stats are archived from every year of professional basketball in the United States.

In this visualization, I chose to use the same four lines and colors as in the previous visualization. I felt this offered consistency and allowed viewers to make a connection between the two visualizations more easily. I chose to add the trend lines to this visualization because I think it’s obvious that each shot type is has been following its trend line fairly closely since the introduction of the three pointer in the NBA. I included an annotation at the point where the trend lines of Two Point attempts and Three Point attempts meet in 2048. I also included an annotation where the expected value of the Two Point Shot and the expected value of the Three Point Shot are equal in 2019. This adds an interesting aspect to the visualization and gives the viewer something fun to think about. By creating a visualization that attempts to use data to predict the future, I am attempting to get the viewers thinking about this flip flop in shot types and hopefully pay attention to it.

Wins vs Three Pointers Made Per Game

This visualization shows how wins correlate to Three Pointers made per game by four different NBA teams. The data goes back to the beginning of NBA history. I collected the data from basketball-reference.com, where basketball stats are archived from every year of professional basketball in the United States. I used data from the website and calculated my own column of “Three Pointers Made Per Game” by dividing total Three Pointers made by the amount of games played that season. This allows for consistent data throughout each year and doesn’t make the strike shortened seasons become outliers in the dataset. I didn’t do this for the previous visualization because I think it’s important that the viewers know about the shortened seasons, so I needed to show their data in at least one visualization.

I chose to place year on the y-axis like all of my other visualizations, to keep it consistent and to show trends over time. I chose to have both Wins and Three Pointers made per game on the x-axis, so that it is easy to see the relationship they have together. I fixed the scales on each graph’s x-axis so that the data wouldn’t be skewed and present stories that aren’t actually there. On the y-axis (Year), I included a slider where you can choose the date range that you would like to focus in on. This gives viewers the opportunity to more closely examine the increases and decreases of Wins/Threes Made year to year, along with letting viewers see what this looks like over a long period of time. This feature adds a lot of versatility to the visualization. I chose to include only four NBA teams in this visualization because it would be too messy and difficult to include all NBA teams, and four side by side team visualizations still looks clean and accomplishes what needs to be presented. In terms of colors, I chose to color each team’s graph with their team colors. I felt this was an obvious choice, since I was dealing with multiple teams, each team needed two colors, and each team has two main colors. By coloring the visualizations with team colors, it allows viewers to easily identify which team they are looking at data for without actually reading any words. However, I couldn’t use these colors as replacements for words because only people who are familiar with the NBA would recognize these teams by color.

Top Ten Shooters

This visualization shows the NBA’s top three point shooters each year since the three point shot was first introduced in the NBA in 1979. I collected the data from basketball-reference.com, where basketball stats are archived from every year of professional basketball in the United States. I came across a list of the ten players in terms of three pointers made for each year in the NBA, along with a separate list of top ten players in three pointers attempted. I combined these two lists in excel and added a column calculating each player’s three point percentage for the season.

I chose to place the year on the x-axis and three point field goals made on the y-axis to let the audience see the trend of how many threes are being made by the top shooters over time more easily. I used two colors in this visualization, blue and grey. These colors have good contrast and it is easy to see the blue dots among the grey, which is why I chose to make the blue dots the most important dots in the visualization. The blue dots represent the player with the most three pointers made that season in the NBA, while the grey dots represent the rest of the top 10 shooters from that year. I decided to leave the names and stats of all of the players off of the fixed visualization, but rather give the viewer the option to mouse over any dot and view the player name, year, three pointers made, three pointers attempted, and three point percentage. This keeps the visualization much cleaner than if I would have fixed any of these details on the visualization. There were just too many dots to include that data cleanly. I included annotations on the years of 1998 and 2011 to show that these seasons were shortened by a strike, causing outlier years in the visualization. I chose to make each player’s dot a certain radius, depending on their three point percentage. As their percentage increases, the dot gets larger, showing they were more effective and more significant than the players who shot a lesser percentage. I chose to do this to account for the players who made the top ten list by shooting a bunch of threes at a low percentage. I felt that the effectiveness of the player’s three point shooting should be highlighted somehow and that this was the best way to do it.