3-Pointers: The Game Changer

Author

Ethan Leamer

Introduction

Since the inception of basketball, players have continuously sought ways to elevate their performance. These improvements have taken many forms—more advanced training routines, refined game strategies, and even innovations in equipment. Typically, such changes have been gradual, taking years to fully integrate into the game. However, in recent years, basketball at all levels has undergone one of the most dramatic transformations in its history.

Thanks to the influence of players like Stephen Curry and James Harden, basketball players around the world are now attempting three-point shots at an unprecedented rate. What was once a situational tactic has become a core offensive strategy. I want to explore just how rapidly this transformation has taken place and how it has reshaped the way basketball is played—particularly in the NBA and NCAA.

To better understand this revolution, I will examine several key factors. First, I’ll look at how the number of three-pointers attempted per game has changed over time. Then, I’ll compare how many players are now scoring at a high level compared to the early years of the three-point line. I’ll also assess how scoring efficiency has evolved as teams have placed a greater emphasis on the three-point shot. Finally, I’ll explore changes in player mentality and habits by analyzing how the frequency of free throws has shifted over the years.

Data Set Overview

This dataset contains player statistics from the 1981 NBA season through the 2025 season. Each row represents a single player’s performance for a specific season, with columns detailing their per-game statistics. It’s important to note that these rows do not represent a player’s entire career—rather, there is a separate entry for each season they played. This structure makes it much easier to analyze trends and observe how the game of basketball has evolved over the years.

Another key point is that all statistics are presented on a per-game basis. This means each stat reflects the average amount recorded per game, calculated by dividing the total stat by the number of games played. This approach ensures fair comparisons across different seasons and player workloads.

The data was obtained by web-scraping basketball-reference.com, a comprehensive resource that provides per-game statistics for every NBA season. I scraped the data year by year and then combined it into a single unified dataset for analysis. Lastly, it’s worth noting that the dataset begins in 1981, which marks the introduction of the three-point line in the NBA—a pivotal moment that makes this starting point particularly significant for studying the 3-point revolution.

Key Variables

Player Info

Player: Name of the player
Age: How old the player was at the start of the season
Team: What team that player was playing for
Pos: What position that player was listed as by their team

Player Statistics

X3P

The average number of 3-pointers a player made per game

X3PA

The average number of 3-pointers a player attempted per game

X3P.

The percentage of a players 3-point attempts that were made

Note: These statistics are the same for X2P, X2PA, and X2P. but for 2-pointers instead

MP

The average number of minutes a player played per game

Pts

The average number of points a player scored per game

Other Characteristics

Awards: The awards that a player won that season
Season: Which season the player recorded these statistics in

Summary Statistics

Summary Statistics for NBA Player Performance (1981–2025)
	Per-Game Averages
Metric	Mean	Standard Deviation
Points (PTS)	8.09	6.02
Rebounds (TRB)	3.45	2.49
Assists (AST)	1.85	1.83
Steals (STL)	0.65	0.46
Blocks (BLK)	0.40	0.48
Field Goal % (FG%)	43.96	10.24

These summary statistics offer an interesting glimpse into trends within the NBA. One of the first things that stands out is that the standard deviations for both assists and blocks are either greater than—or very close to—their respective means. This suggests a wide spread in the data: players tend to record either a high number of these stats or very few.

This variation aligns with positional roles on the court. Guards, particularly point guards, are typically responsible for facilitating the offense, so they naturally accumulate a high number of assists. In contrast, forwards and centers—who are usually positioned closer to the basket—tend to record far fewer assists. The opposite is true for blocks: big men like power forwards and centers often rack up block numbers due to their size and presence in the paint, whereas guards generally contribute very little in this category.

Visualizing the Evolution of Scoring

3-Pointers Attempted

This graph shows a steady upward trend in three-point attempts from 1981 to 2014. Then, around 2014, there’s a sharp increase. In just ten years, the average number of three-pointers taken per player per game jumped from 1.6 to 2.6—a change that previously took about 25 years to occur.

This sudden rise is largely due to players like Stephen Curry, who made the three-pointer a central part of his offensive game and proved how effective it could be.

There’s also a noticeable spike from 1995 to 1997, which happened when the NBA temporarily shortened the three-point line. The line was moved back to its original distance in 1998, and attempts dropped as a result.

Number of Players Scoring at a High Level

We can see that more players than ever are now averaging at least 20 points per game—in fact, the number has nearly doubled compared to any season before 2015. This indicates that players are not only scoring more, but doing so more efficiently. The rise in high-scoring players closely correlates with the increase in three-point attempts per game. This trend suggests that NBA coaches and players have recognized that, in most cases, the three-point shot is the most effective way to score in today’s game.

3-Point Percentage

This graph shows that players are not only attempting more three-pointers—they’re also making them at a rate never seen before. This improvement can largely be attributed to the growing emphasis on the three-point shot in modern basketball. As its value has become more apparent, players have dedicated significantly more time to practicing these shots, which has naturally led to better in-game performance.

I’d also like to once again highlight the noticeable spike between 1995 and 1997, which mirrors the trend seen in the earlier graph. This was due to the NBA temporarily shortening the three-point line during those seasons.

2-Point Percentage

This graph shows that as players attempt more three-pointers, they are also taking fewer long two-point shots. These mid-range shots are nearly as difficult as three-pointers but are only worth two-thirds of the points, making them a less efficient choice. As a result, players today tend to either shoot from beyond the arc or take shots much closer to the basket. These closer-range attempts are significantly easier and are made at a much higher percentage compared to the mid-range shots that were popular in the 1990s and early 2000s.

Free Throws per Game

The most common reason a player shoots free throws is that they drove to the basket and were fouled while attempting a layup. This graph illustrates how the league’s shift toward more three-point shooting has led to a decline in the number of players driving to the rim and drawing fouls. As perimeter shots become more central to offensive strategy, opportunities for drawing contact in the paint—and therefore earning free throws—have decreased.

NCAA Data

Summary

This dataset contains data from the past 11 years of NCAA Men’s Basketball, covering the 2013 to 2023 seasons. Unlike the NBA dataset, which tracks individual player statistics, this one focuses on team-level performance by season. Each entry represents the collective stats for a given team, rather than breaking down performance by individual players.

Additionally, this dataset only spans from 2013 to 2023, which is a bit more limited in scope. While this shorter time frame doesn’t allow for as long-term a perspective as the NBA dataset, it still offers valuable insights. We can use this data to make interesting comparisons to the NBA and see how similar trends have emerged at the collegiate level over the past decade.

Three Point Percentage

From this graph, we can observe that during the mid to late 2010s, three-point percentage steadily increased. However, in recent years, it has dropped back down to levels similar to a decade ago. This trend contrasts with what we saw in the NBA data. One possible explanation is that college players, who tend to be less experienced and perhaps more nervous, are trying to emulate what they see in the professional game. They might be attempting more three-pointers than their skill level supports, leading to lower shooting percentages. While NBA players have had years of specialized practice to master the three-point shot, many college athletes simply haven’t had the same level of preparation or consistency with shooting from distance.

That said, I anticipate that this trend could reverse in the near future. As younger players who have grown up practicing three-point shooting throughout their youth basketball careers enter college, we may see an improvement in these shooting percentages as these players bring more developed skills to the collegiate level.

Two Point Percentage

Unlike the three-point percentage graph, this one shows a trend more similar to what we observed in the NBA data. There has been a steep increase in two-point percentage, rising by over 2.5 points in just a decade. This can likely be attributed to many of the same factors I discussed in relation to the NBA. Most of the two-pointers being taken are close-range shots near the basket, which are higher percentage attempts. Additionally, players are increasingly choosing not to take the lower-percentage mid-range jumpers, instead opting for three-pointers or driving to the hoop for higher-percentage shots.

Conclusion

The three-point revolution has profoundly transformed basketball, impacting nearly every aspect of the offensive game. Beginning with the influence of players from the 2014-2016 era, such as Stephen Curry, the number of three-pointers attempted per game skyrocketed. This dramatic increase has been accompanied by a surge in scoring, with more players than ever reaching high-point averages.

At the same time, both three-point and two-point percentages have risen across the league. This can be attributed to a combination of factors: a greater emphasis on practicing three-pointers, improved shot selection, and more efficient offense that prioritizes high-percentage shots. Players have shifted away from low-percentage mid-range jumpers, focusing more on shots from beyond the arc or closer to the basket.

Lastly, we see that the rise in perimeter play has contributed to a sharp decline in free throws per game, as fewer players are driving to the basket and drawing fouls. This marks a shift in how the game is being played at both the professional and collegiate levels.

Speaking of the college game, while some of these trends have begun to take hold in NCAA basketball, the data suggests that players may not yet be fully equipped for this transition. Many college athletes are still adjusting to the increased reliance on the three-point shot, and as a result, shooting percentages are not as high as in the NBA. However, as younger generations of players, who have grown up with the three-pointer as a key part of their development, enter the college ranks, we may see these trends continue to evolve.

--- title: "3-Pointers: The Game Changer" # Name of your HTML output author: "Ethan Leamer" # Author name toc: true # Generates an automatic table of contents. format: # Options related to formatting. html: # Options related to HTML output. code-tools: TRUE # Allow the code tools option showing in the output. embed-resources: TRUE # Embeds all components into a single HTML file. execute: # Options related to the execution of code chunks. warning: FALSE # FALSE: Code chunk sarnings are hidden by default. message: FALSE # FALSE: Code chunk messages are hidden by default. echo: TRUE # TRUE: Show all code in the output. --- ```{r} #| echo: FALSE library(tidyverse) library(lubridate) library(knitr) library(kableExtra) nba_data <- read_csv("https://myxavier-my.sharepoint.com/:x:/g/personal/leamere_xavier_edu/EWg2S_Nv6AtAosGiGXzD90oBzn83C2bzeCF9tDT4TLTIUw?download=1") ``` ## Introduction Since the inception of basketball, players have continuously sought ways to elevate their performance. These improvements have taken many forms—more advanced training routines, refined game strategies, and even innovations in equipment. Typically, such changes have been gradual, taking years to fully integrate into the game. However, in recent years, basketball at all levels has undergone one of the most dramatic transformations in its history. Thanks to the influence of players like Stephen Curry and James Harden, basketball players around the world are now attempting three-point shots at an unprecedented rate. What was once a situational tactic has become a core offensive strategy. I want to explore just how rapidly this transformation has taken place and how it has reshaped the way basketball is played—particularly in the NBA and NCAA. To better understand this revolution, I will examine several key factors. First, I’ll look at how the number of three-pointers attempted per game has changed over time. Then, I’ll compare how many players are now scoring at a high level compared to the early years of the three-point line. I’ll also assess how scoring efficiency has evolved as teams have placed a greater emphasis on the three-point shot. Finally, I’ll explore changes in player mentality and habits by analyzing how the frequency of free throws has shifted over the years. ## Data Set Overview This [dataset](https://myxavier-my.sharepoint.com/:x:/g/personal/leamere_xavier_edu/EWg2S_Nv6AtAosGiGXzD90oBzn83C2bzeCF9tDT4TLTIUw?download=1) contains player statistics from the 1981 NBA season through the 2025 season. Each row represents a single player’s performance for a specific season, with columns detailing their per-game statistics. It’s important to note that these rows do not represent a player’s entire career—rather, there is a separate entry for each season they played. This structure makes it much easier to analyze trends and observe how the game of basketball has evolved over the years. Another key point is that all statistics are presented on a per-game basis. This means each stat reflects the average amount recorded per game, calculated by dividing the total stat by the number of games played. This approach ensures fair comparisons across different seasons and player workloads. The data was obtained by web-scraping [basketball-reference.com](https://www.basketball-reference.com/), a comprehensive resource that provides per-game statistics for every NBA season. I scraped the data year by year and then combined it into a single unified dataset for analysis. Lastly, it's worth noting that the dataset begins in 1981, which marks the introduction of the three-point line in the NBA—a pivotal moment that makes this starting point particularly significant for studying the 3-point revolution. ## Key Variables ### Player Info `Player` : Name of the player `Age` : How old the player was at the start of the season `Team` : What team that player was playing for `Pos` : What position that player was listed as by their team ### Player Statistics `X3P` : The average number of 3-pointers a player made per game `X3PA` : The average number of 3-pointers a player attempted per game `X3P.` : The percentage of a players 3-point attempts that were made Note: These statistics are the same for X2P, X2PA, and X2P. but for 2-pointers instead `MP` : The average number of minutes a player played per game `Pts` : The average number of points a player scored per game ### Other Characteristics `Awards` : The awards that a player won that season `Season` : Which season the player recorded these statistics in ## Summary Statistics ```{r} #| echo: FALSE summary_stats <- nba_data %>% summarise( Metric = c("Points (PTS)", "Rebounds (TRB)", "Assists (AST)", "Steals (STL)", "Blocks (BLK)", "Field Goal % (FG%)"), Mean = c( mean(PTS, na.rm = TRUE), mean(TRB, na.rm = TRUE), mean(AST, na.rm = TRUE), mean(STL, na.rm = TRUE), mean(BLK, na.rm = TRUE), mean(FG., na.rm = TRUE) * 100 # Convert to percentage ), SD = c( sd(PTS, na.rm = TRUE), sd(TRB, na.rm = TRUE), sd(AST, na.rm = TRUE), sd(STL, na.rm = TRUE), sd(BLK, na.rm = TRUE), sd(FG., na.rm = TRUE) * 100 # Convert to percentage ) ) %>% mutate( Mean = round(Mean, 2), SD = round(SD, 2) ) summary_stats %>% kable( caption = "Summary Statistics for NBA Player Performance (1981–2025)", col.names = c("Metric", "Mean", "Standard Deviation"), align = c("l", "c", "c"), format = "html" ) %>% kable_styling( bootstrap_options = c("striped", "hover", "condensed"), full_width = FALSE, position = "center" ) %>% column_spec(1, bold = TRUE) %>% add_header_above(c(" " = 1, "Per-Game Averages" = 2)) ``` These summary statistics offer an interesting glimpse into trends within the NBA. One of the first things that stands out is that the standard deviations for both assists and blocks are either greater than—or very close to—their respective means. This suggests a wide spread in the data: players tend to record either a high number of these stats or very few. This variation aligns with positional roles on the court. Guards, particularly point guards, are typically responsible for facilitating the offense, so they naturally accumulate a high number of assists. In contrast, forwards and centers—who are usually positioned closer to the basket—tend to record far fewer assists. The opposite is true for blocks: big men like power forwards and centers often rack up block numbers due to their size and presence in the paint, whereas guards generally contribute very little in this category. ## Visualizing the Evolution of Scoring ### 3-Pointers Attempted ```{r} #| echo: FALSE nba_data %>% group_by(season) %>% summarise(avg_3PA = mean(X3PA, na.rm = TRUE)) %>% arrange(season) %>% ggplot(aes(x = season, y = avg_3PA)) + geom_line(color = "blue", size = 1.2) + geom_point(color = "darkred", size = 2) + labs(title = "Average 3-Point Attempts per Game by Season", x = "Season", y = "Average 3PA per Game") + theme_minimal() ``` This graph shows a steady upward trend in three-point attempts from 1981 to 2014. Then, around 2014, there’s a sharp increase. In just ten years, the average number of three-pointers taken per player per game jumped from 1.6 to 2.6—a change that previously took about 25 years to occur. This sudden rise is largely due to players like Stephen Curry, who made the three-pointer a central part of his offensive game and proved how effective it could be. There’s also a noticeable spike from 1995 to 1997, which happened when the NBA temporarily shortened the three-point line. The line was moved back to its original distance in 1998, and attempts dropped as a result. ### Number of Players Scoring at a High Level ```{r} #| echo: FALSE nba_data %>% filter(PTS > 20) %>% group_by(season) %>% summarise(count = n()) %>% ungroup() %>% ggplot(aes(x = season, y = count)) + geom_line(color = "blue") + geom_point(color = "blue") + labs(title = "Number of NBA Players Scoring > 20 PPG by Season", x = "Season", y = "Number of Players") + theme_minimal() + scale_x_continuous(breaks = seq(min(nba_data$season), max(nba_data$season), by = 5)) ``` We can see that more players than ever are now averaging at least 20 points per game—in fact, the number has nearly doubled compared to any season before 2015. This indicates that players are not only scoring more, but doing so more efficiently. The rise in high-scoring players closely correlates with the increase in three-point attempts per game. This trend suggests that NBA coaches and players have recognized that, in most cases, the three-point shot is the most effective way to score in today’s game. ### 3-Point Percentage ```{r} #| echo: FALSE nba_data %>% filter(!is.na(X3P.)) %>% group_by(season) %>% summarise(avg_3pt = mean(X3P., na.rm = TRUE)) %>% ungroup() %>% ggplot(aes(x = season, y = avg_3pt)) + geom_line(color = "red") + geom_point(color = "red") + labs(title = "Average 3-Point Percentage by Season", x = "Season", y = "Average 3-Point Percentage") + theme_minimal() + scale_x_continuous(breaks = seq(min(nba_data$season), max(nba_data$season), by = 5)) + scale_y_continuous(labels = scales::percent_format(scale = 100)) ``` This graph shows that players are not only attempting more three-pointers—they're also making them at a rate never seen before. This improvement can largely be attributed to the growing emphasis on the three-point shot in modern basketball. As its value has become more apparent, players have dedicated significantly more time to practicing these shots, which has naturally led to better in-game performance. I’d also like to once again highlight the noticeable spike between 1995 and 1997, which mirrors the trend seen in the earlier graph. This was due to the NBA temporarily shortening the three-point line during those seasons. ### 2-Point Percentage ```{r} #| echo: FALSE nba_data %>% filter(!is.na(X2P.)) %>% group_by(season) %>% summarise(avg_2pt = mean(X2P., na.rm = TRUE)) %>% ungroup() %>% ggplot(aes(x = season, y = avg_2pt)) + geom_line(color = "green") + geom_point(color = "green") + labs(title = "Average 2-Point Percentage by Season", x = "Season", y = "Average 2-Point Percentage") + theme_minimal() + scale_x_continuous(breaks = seq(min(nba_data$season), max(nba_data$season), by = 5)) + scale_y_continuous(labels = scales::percent_format(scale = 100)) ``` This graph shows that as players attempt more three-pointers, they are also taking fewer long two-point shots. These mid-range shots are nearly as difficult as three-pointers but are only worth two-thirds of the points, making them a less efficient choice. As a result, players today tend to either shoot from beyond the arc or take shots much closer to the basket. These closer-range attempts are significantly easier and are made at a much higher percentage compared to the mid-range shots that were popular in the 1990s and early 2000s. ### Free Throws per Game ```{r} #| echo: FALSE nba_data %>% filter(!is.na(FTA)) %>% group_by(season) %>% summarise(avg_fta = mean(FTA, na.rm = TRUE)) %>% ungroup() %>% ggplot(aes(x = season, y = avg_fta)) + geom_line(color = "purple") + geom_point(color = "purple") + labs(title = "Average Free Throw Attempts per Game by Season", x = "Season", y = "Average Free Throw Attempts") + theme_minimal() + scale_x_continuous(breaks = seq(min(nba_data$season), max(nba_data$season), by = 5)) ``` The most common reason a player shoots free throws is that they drove to the basket and were fouled while attempting a layup. This graph illustrates how the league's shift toward more three-point shooting has led to a decline in the number of players driving to the rim and drawing fouls. As perimeter shots become more central to offensive strategy, opportunities for drawing contact in the paint—and therefore earning free throws—have decreased. ## NCAA Data ### Summary This [dataset](https://myxavier-my.sharepoint.com/:x:/g/personal/leamere_xavier_edu/EQr_psC0_75CmpKlgv7r0n0B5hywBX6qIsJ8pK_DJiQTXQ?download=1) contains data from the past 11 years of NCAA Men's Basketball, covering the 2013 to 2023 seasons. Unlike the NBA dataset, which tracks individual player statistics, this one focuses on team-level performance by season. Each entry represents the collective stats for a given team, rather than breaking down performance by individual players. Additionally, this dataset only spans from 2013 to 2023, which is a bit more limited in scope. While this shorter time frame doesn’t allow for as long-term a perspective as the NBA dataset, it still offers valuable insights. We can use this data to make interesting comparisons to the NBA and see how similar trends have emerged at the collegiate level over the past decade. ### Three Point Percentage ```{r} #| echo: FALSE cbb_data <- read_csv("https://myxavier-my.sharepoint.com/:x:/g/personal/leamere_xavier_edu/EQr_psC0_75CmpKlgv7r0n0B5hywBX6qIsJ8pK_DJiQTXQ?download=1") cbb_data %>% group_by(YEAR) %>% summarize(avg_3p = mean(`3P_O`, na.rm = TRUE)) %>% ggplot(aes(x = YEAR, y = avg_3p)) + geom_line(color = "blue", size = 1) + geom_point(color = "red") + scale_x_continuous( breaks = seq(min(cbb_data$YEAR, na.rm = TRUE), max(cbb_data$YEAR, na.rm = TRUE), by = 2) # Labels every 2 years ) + labs( title = "Average 3-Point Percentage by Year", x = "Year", y = "Average 3P%" ) + theme_minimal() ``` From this graph, we can observe that during the mid to late 2010s, three-point percentage steadily increased. However, in recent years, it has dropped back down to levels similar to a decade ago. This trend contrasts with what we saw in the NBA data. One possible explanation is that college players, who tend to be less experienced and perhaps more nervous, are trying to emulate what they see in the professional game. They might be attempting more three-pointers than their skill level supports, leading to lower shooting percentages. While NBA players have had years of specialized practice to master the three-point shot, many college athletes simply haven’t had the same level of preparation or consistency with shooting from distance. That said, I anticipate that this trend could reverse in the near future. As younger players who have grown up practicing three-point shooting throughout their youth basketball careers enter college, we may see an improvement in these shooting percentages as these players bring more developed skills to the collegiate level. ### Two Point Percentage ```{r} #| echo: FALSE cbb_data %>% group_by(YEAR) %>% summarize(avg_2p = mean(`2P_O`, na.rm = TRUE)) %>% ggplot(aes(x = YEAR, y = avg_2p)) + geom_line(color = "yellow", size = 1) + geom_point(color = "blue") + scale_x_continuous( breaks = seq(min(cbb_data$YEAR, na.rm = TRUE), max(cbb_data$YEAR, na.rm = TRUE), by = 2) # Labels every 2 years ) + labs( title = "Average 2-Point Percentage by Year", x = "Year", y = "Average 2P%" ) + theme_minimal() ``` Unlike the three-point percentage graph, this one shows a trend more similar to what we observed in the NBA data. There has been a steep increase in two-point percentage, rising by over 2.5 points in just a decade. This can likely be attributed to many of the same factors I discussed in relation to the NBA. Most of the two-pointers being taken are close-range shots near the basket, which are higher percentage attempts. Additionally, players are increasingly choosing not to take the lower-percentage mid-range jumpers, instead opting for three-pointers or driving to the hoop for higher-percentage shots. ## Conclusion The three-point revolution has profoundly transformed basketball, impacting nearly every aspect of the offensive game. Beginning with the influence of players from the 2014-2016 era, such as Stephen Curry, the number of three-pointers attempted per game skyrocketed. This dramatic increase has been accompanied by a surge in scoring, with more players than ever reaching high-point averages. At the same time, both three-point and two-point percentages have risen across the league. This can be attributed to a combination of factors: a greater emphasis on practicing three-pointers, improved shot selection, and more efficient offense that prioritizes high-percentage shots. Players have shifted away from low-percentage mid-range jumpers, focusing more on shots from beyond the arc or closer to the basket. Lastly, we see that the rise in perimeter play has contributed to a sharp decline in free throws per game, as fewer players are driving to the basket and drawing fouls. This marks a shift in how the game is being played at both the professional and collegiate levels. Speaking of the college game, while some of these trends have begun to take hold in NCAA basketball, the data suggests that players may not yet be fully equipped for this transition. Many college athletes are still adjusting to the increased reliance on the three-point shot, and as a result, shooting percentages are not as high as in the NBA. However, as younger generations of players, who have grown up with the three-pointer as a key part of their development, enter the college ranks, we may see these trends continue to evolve.