Bruin Sports Analytics
How Can We Accurately Compare NBA Players Across Different Eras?
By: Daniel Wang
Jordan or LeBron: Who is the “GOAT”? Cross-era comparisons are often at the center of NBA discussions, but the style of play in the NBA has changed drastically over time, making it nearly impossible to directly compare players from different eras. In this article, I will analyze how the game of basketball has evolved over time and use this information to discuss two approaches to more accurately compare players from different eras.
How Has the Style of Basketball Evolved Over Time?
The style of play in the NBA has undergone drastic changes throughout history due to significant rule changes, technological advancements, and new players and teams that have revolutionized the game. One major rule change was the introduction of the three-point line in the 1979-80 season, which has dramatically changed the game of basketball. While the game didn’t see drastic changes right away with everyone having to adapt to the new rule, the three-point shot now dominates today’s game. The following graph shows the steady increase in three-point volume over time, with the increase becoming more significant in recent years.
Today’s analytically-driven world suggests that three-point shots are far more efficient than mid-range shots, so many teams are hunting for threes by prioritizing spacing and freedom of movement within the offense, which naturally speeds up the pace of the game. Teams like the Houston Rockets and Golden State Warriors have generated a lot of success within the past decade by running small-ball lineups and shooting a high volume of three-point shots, which has inspired other teams to follow suit.
By playing a small-ball lineup and replacing the traditional center with another guard or wing, the game becomes much faster because shots are taken quicker during each possession and there is less offensive rebounding, both of which create more possession changes. Mike D’Antoni’s revolutionary “7 seconds or less” offense encapsulates the modern-day style of play, defined by pushing the ball and spacing the floor. Because of the increased volume and success in three-point shots over time, the NBA has also seen an increase in offensive efficiency and pace.
The pace in the 1970s was extremely high mainly due to sloppy play and an overreaction to the introduction of the shot clock in 1954, as many teams would attempt shots very early in the 24-second clock. As the game became more controlled and pace naturally declined, it eventually reached a time in the late 1990s and early 2000s that was known as a “dead-ball era” characterized by physical defense and grind-it-out offense. There was a lot of isolation basketball where teams sought out one-on-one opportunities. One reason for this was because of illegal defense rules that made it such that defenders had to be near the offensive player they were guarding, making zone defense impossible and restricting help defense. Because of the slow pace and physical brand of basketball, games during this time were extremely low scoring. In 2001, the illegal defense rules were eliminated and the defensive three-second rule was implemented instead. In addition, the time offenses had to get the ball past half-court was reduced from ten seconds to eight. While seemingly minor, this change enticed defenses to press up and pressure the offense in the backcourt which promoted open court action and increased pace. In 2005, banning hand checking was strictly enforced, which decreased the level of physicality allowed by defenders. This played a role in increasing offensive output by giving offensive players more space to operate. Additionally, there was a noticeable spike in pace in the 2018-19 season. This can be attributed to the introduction of the rule that the shot clock resets to 14 seconds on offensive rebounds rather than 24, making teams get back into action quicker following an offensive rebound rather than completely resetting the possession.
As pace increases and there are more possessions in each game, the games become higher scoring, as depicted by the following graphs.
Due to increases in pace over time, modern-day NBA stats are naturally inflated compared to prior eras like the slow-paced late-1990s, making it unfair to compare individual player statistics from different eras. Many rule changes dating back to the introduction of the three-point line itself have been implemented to create these higher-scoring, faster-paced games in order to attract a larger audience by making the game more entertaining, so the evolution of the game that we are observing is no surprise. While players are undoubtedly becoming more and more skilled as the game evolves, pace is a major factor for why historic NBA records are being broken left and right in today’s modern game.
How Can We Compare Player’s Across Different Eras?
Method 1: Adjusting Stats
Because we cannot directly compare the stats of players who played in different eras due to changes in the style of play, we have to find a way to adjust each player’s stats based on the season they played in such that they are all generalized. To do so, we can first calculate each player’s “per 75 possessions” stats to balance out pace differences across different seasons. “Per 75 possessions” is a good measurement to use because the average star NBA player today plays around 75 possessions per game. However, beyond balancing for pace differences, we must also account for changes in the style of play – in other words – how each possession is used. We can use league average generalized statistics like offensive rating (estimate of points per 100 possessions), as well as assists per 100 possessions and rebounds per 100 possessions for each of the different individual seasons as a whole, then find what percent change each of the statistics has undergone between seasons. We are essentially finding the “inflation factor” between different seasons, then we can use this to adjust a player’s “per 75 possessions” stats accordingly. Aside from the most general statistics like points, rebounds, and assists, we can also account for shooting efficiency by finding the player’s true shooting percentage and comparing it to the league average true shooting percentage for the season they played in to find the player’s relative true shooting percentage. Due to differences in style of play, some eras were more or less efficient than others, so we cannot directly compare the shooting percentages of players from different eras. True shooting percentage effectively assesses a player’s efficiency shooting the ball by adjusting for three-pointers and free throws.
Let’s compare one of the top players from every 5 years between 1980 and 2020 using this method to generalize their stats to 2020 values.
While the adjustment does not significantly alter a player’s stats, it makes the stats comparable with other players from different eras to account for changes in pace and style of play. From the first graph, it looks like the player stats all become relatively close despite how different some of the eras were in terms of play style. Two things that stand out even after adjusting the stats are Giannis’ all-around output in 2020-21 and Steph Curry’s scoring efficiency in his unanimous MVP season. While we do need to take modern-day player stats with a grain of salt due to the extremely fast-paced, high-scoring nature of today’s game, it’s clear that we can’t discount the unprecedented play we are seeing from generational talents like Steph Curry and Giannis Antetokounmpo.
Reggie Miller vs Klay Thompson
We can also use this method to look at a more specific comparison between two players like Klay Thompson and Reggie Miller who are widely considered to be two of the greatest shooters of all time.
Klay Thompson was a notable snub from the NBA’s 75th Anniversary Team and many people argued that Klay deserved Miller’s spot on the list because Klay boasts a higher career scoring average and higher career shooting percentage to go along with his four championships. However, Miller played much of his career during the dead-ball era, so his real performance is much more impressive than what his general statistics may suggest, especially compared to the numbers that modern-day players are putting up day-in and day-out.
After adjusting their stats, Reggie Miller’s adjusted points per game becomes higher than Klay’s in their respective seasons, even though he technically averaged less points per game. In addition, Miller’s relative shooting efficiency is much higher compared to Klay for both three-pointers specifically and true shooting percentage as a whole, setting them apart.
Steve Nash vs Tyrese Haliburton
Another interesting player comparison to make is Steve Nash from his 2005-06 MVP season with Tyrese Haliburton thus far in the 2022-23 NBA season. Nash averaged 18.8 points and 10.5 assists per game in his 2005-06 MVP season, while Haliburton has averaged 19.1 points and 10.9 assists in 22 games so far this season. Haliburton’s numbers stand out as they are seemingly better than a former MVP’s stats, but we have to first adjust the stats to account for changes in the style of play between the different eras before jumping to any conclusions.
After adjustment, we see that Nash’s point and assist averages become higher than Haliburton’s, although they are still close. Digging further, we see that Nash’s shooting efficiency sets him apart from Haliburton. Nash had a 50/40/90 season that year (over 50% from the field, over 40% from three, and over 90% from the free throw line) and posted a true shooting percentage 10.3% better than league average, whereas Haliburton is just about league average in that department. Additionally, Nash orchestrated a Suns offense that averaged a league best 108.6 points per game during that season and won 54 games, which definitely added to his MVP case. While Haliburton has been tremendous so far this season, this comparison helps us understand what sets Steve Nash’s 05-06 MVP season apart from what Haliburton has done thus far this season, something that is hard to tell by just comparing their general stats.
Method 2: Comparing Relative Dominance
Another way we can compare players from different eras is by analyzing how much a player dominated their respective era and comparing different players’ level of dominance. The way we can do this is by finding the league average player stats for a given season and finding the standard deviation for each of the stats, then finding how many standard deviations away from the mean the player of interest was for each of the stats. The more standard deviations away from the mean a player is, the more dominant they were.
Michael Jordan vs LeBron James vs Kobe Bryant
Using this method, let’s compare the age-27 seasons of Michael Jordan, LeBron James, and Kobe Bryant. Between Jordan, LeBron, and Kobe, there’s a lot of debate about who is the “GOAT” but it’s hard to reach a consensus decision because they played in different eras.
This graph suggests that Kobe Bryant scored the most, LeBron James was the most well-rounded, and Michael Jordan was the most efficient scorer.
Magic Johnson vs Stephen Curry vs Oscar Robertson
Let’s also compare the relative dominance of Magic Johnson, Oscar Robertson, and Stephen Curry, three of the best point guards of all time.
Since they all played during drastically different eras, it’s difficult to compare them and reach a consensus decision about who was actually the best. Let’s compare each of their best MVP seasons: Curry in 2015-2016, Magic in 1986-1987, and Oscar in 1963-1964.
This graph suggests that Oscar and Magic dominated their eras with their passing, recording unprecedented assist numbers for their time. Curry still posted assist numbers nearly 3 standard deviations higher than the mean, but he dominated with his scoring and efficiency – which makes sense since he’s widely regarded as the greatest shooter the NBA has ever seen. Interestingly, while Oscar and Magic are more well-known for their rebounding ability compared to Curry, their rebounding stats are not significantly greater than Curry’s when we account for the context of their different eras.
While the comparison between players across eras will never be perfect, we can attempt to make them as accurate as possible by adjusting their stats to account for changes in the NBA’s style of play or comparing their relative dominance by analyzing how much a player dominated their respective era. However, much more should be taken into consideration than just points, rebounds, assists, and shooting percentages. Ideally, we would dig into more advanced stats and consider other factors like defense and team contributions to make these comparisons even more in-depth. Still, the two analysis methods we performed can provide a lot of additional insight about how players from different eras stack up. The next time a new record is broken or a cross-era comparison emerges, it’s important to first consider the evolution of basketball before jumping to conclusions.