In Defense of the Mid-Range Jumper
In 1979, the three-point line was painted onto NBA courts, and this sparked an ongoing debate over the shot’s value compared to mid-range jumpers (defined as shots outside of the paint but inside the three-point line). The basic argument in favor of taking more three-point shots is simple. A three-pointer is worth 50% more than a 2 pointer, so, all else equal, it is a more valuable shot. However, we know all else is not equal; three-pointers are farther away from the basket and are harder to make than mid-range shots or layups.
A shot’s value is most practically defined as the expected points per shot. In 2019, NBA players had an average three-point field goal percentage of 35.5%. If we multiply this percentage by 3 for the number of points per shot, we get the average value of a three-point shot: 1.07 points. NBA players made mid-range shots at a 40.5% clip in 2019, valuing them at .81 points per shot.
From this surface-level analysis, it seems clear teams should take more three-pointers and less mid-range jump shots. In fact, since 1979, this is the trend we have seen in the NBA:
Teams have been shooting more and more three-pointers each year. The anomalous bump in the 90s is due to the NBA’s decision to move the three-point line from 23 feet, 9 inches and 22 feet in the corners to a uniform 22 feet around the basket in the 1995 season. This change was reversed after the 1997 season, leading to the subsequent drop in 3PA per game. Interestingly, people are now calling for the line to be moved back farther from the basket to reduce the frequency of three-point shots. One argument is that this shot is getting too easy for NBA players and that the game is broken by just shooting as many threes as possible. But is this change really necessary? First, let’s look at how three-point percentages have changed with the boom in three-pointers attempted:
While players appear to have improved their three-point accuracy until about 2000 (again, the 1995–1997 seasons had a shorter three-point line), 3P% has since leveled out. This suggests teams have decided it’s a good idea to shoot more threes despite not making them at higher rates. This is more clear looking at the combined charts since 2000:
The cost of this rise in three-point shooting has been a decline in mid-range jumpers. According to more and more analytics experts, the midrange shot is dead. NBA teams appear to, on average, agree:
Leading the charge against mid-range jumpers is Houston Rockets GM Daryl Morey. He has turned the Rockets into an analytics-driven team of players who are told exactly where they should and should not shoot. This strategy is colloquially called “Moreyball” (though mostly by Rockets fans) in reference to Billy Beane’s “Moneyball” advanced analytics and sabermetrics-based approach to baseball. Based on this article’s first graph (albeit more segmented data by three-point location, mid-range shot area, etc.), Morey tells his players to take almost exclusively three-pointers and layups and to avoid mid-range jumpers at all costs. It shows in their shot selection data.
With only 5% of their shots coming from the midrange, the NBA averages 300% more mid-range shots than the Rockets. The Rockets also take far more threes than the league average. What is incredible about this graph is that this snapshot is in the context of a league that already shoots far less mid-range jumpers than previous seasons (as shown in a previous chart). For reference, if we compared the shot selection of these 2019 Rockets to the league average shot selection in 1998 (the last year Michael Jordan won a championship), the results are staggering.
But has Morey earned the comparison to Billy Beane? The Rocket’s and NBA’s trend toward analytics-driven basketball is certainly not without controversy. Kevin Durant — a former NBA MVP, two-time NBA champion, and a prolific mid-range shooter — weighed in about analytics:
Charles Barkley — another NBA MVP, Hall of Famer, and current NBA analyst — agrees, saying “First of all, I’ve always believed analytics was crap. You know I never mention the Rockets as legitimate contenders ’cause they’re not.”
Morey shot back at Barkley.
Despite the passion on both sides of the mid-range/three-point debate, both seem to have reasonable arguments. Barkley, Durant, and other mid-range jumper advocates argue basketball is an art, not a science — that it’s a dynamic game requiring constant adjustments and intuitive and unquantifiable feel. Morey uses data showing three-pointers are on average more valuable and advocates for taking no mid-range shots. Of course, the best test of this argument is measuring which approach has been more successful. The definition of success is something on which both sides agree: championship rings. Lakers legend Magic Johnson summarizes what most NBA players and executives feel.
Statistics and regular season wins are means to an end. The short answer, then, is that despite sustained regular season success, Morey has not earned his title of “Moreyball.” The Rockets have not been successful with their analytics-driven strategy. Five seasons of no championships might feel like an insignificant sample size, but through this time the Rockets had an elite three-point shooter in Eric Gordon, an MVP in James Harden, a surefire Hall of Famer in elite point guard Chris Paul, and a dominant big man in Clint Capela. They had all the pieces to win a championship, but their analytics-based strategies didn’t pay off.
Some might argue the Rockets didn’t have a bad strategy, but they were simply outshot by another team with a similar strategy: the “Splash Brothers”- led Golden State Warriors. Actually, despite the highlight reels of Stephen Curry and Klay Thompson swishing threes, Golden State takes a lot of mid-range shots. Most championship teams do. The following graphic compares 2018 regular season offensive efficiency (team points scored per 100 possessions) to proportion of mid-range shots taken:
The Golden State Warriors, first in offensive efficiency and 2018 championship-winners, shot a lot of mid-range jumpers, and they actually made the most mid-range jumpers in the league. I ran a correlation on the above two variables and found a correlation of -.15, suggesting there is a weak negative correlation between shooting more mid-range shots and offensive efficiency. A simple linear regression shows proportion of mid-range shots taken is not a statistically significant predictor of offensive efficiency alone.
However, not every team is competing for the championship each year. While not all non-contenders require reminders from NBA commissioner Adam Silver not to tank so obviously, there are clearly some teams that are not pursuing winning strategies. For this reason, I’ll restrict the above scatterplot to contenders — defined here as the top ten teams in offensive efficiency.
Now the correlation drops to (-.07) — practically 0 — and it is clear that proportion of mid-range shots does not significantly predict offensive efficiency (p-value of only .85). This seems contradictory to the initial analysis, because aren’t three-point shots more valuable in terms of expected value? However, this scatter plot also does not indicate that increased proportion of mid-range shooting actually helps offensive efficiency, either. Yes, Golden State had the highest offensive efficiency and was among the highest in mid-range shooting proportion, but Houston is #2 in offensive efficiency using their “mid-range shots aren’t allowed Moreyball strategy.” But the regular season doesn’t determine who wins championships. Let’s look at how these stats compare for top teams in the playoffs. For the following graphic, I compare the final 8 teams of the 2018 playoffs (the playoffs for the season we just looked at). These 8 teams represents the ones who made it out of the first round of the playoffs.
This graph appears to contradict the “analytics” even more. There is now a medium-strength positive correlation (.48) between shooting more mid-range shots and offensive efficiency. Interestingly, we see that Houston’s offensive efficiency dropped tremendously from about 112.5 points per hundred possessions to 107, falling far behind the Raptors and Warriors. The Warriors led this group by a significant margin in both mid-range shooting and offensive efficiency. While there appear to be too few data points to confirm that mid-range shooting is beneficial in the playoffs, each of the data points above represent at least 9 (New Orleans Pelicans) and as many as 22 games played (Cleveland Cavaliers), so this scatterplot is relatively stable.
This model shows that an increase in proportion of shots taken from mid-range is correlated with higher offensive efficiency. While this regression does not show statistical significance, the model does not account for the fact that each data point is the aggregation of many games played. If all playoff games were mapped, it would strengthen this regression further.
I contend that a refusal to adapt to in-game dynamics and take mid-range shots has hurt the Rocket’s playoff performance. Mid-range shots are undervalued, especially in the playoffs, for two major reasons. First, the concept of “gravity” (usually applied to individual players). Committing to taking mid-range shots opens up the floor for easier three-pointers, because defenses are forced to guard the mid-range. Second, playoff defenses are more intense and better planned, making static strategies like never taking mid-range jumpers less effective.
Gravity
Even if mid-range shots taken by a team are less “valuable” than three-pointers — when directly compared — they open up the floor for better looks at three-pointers. Committing to consistently taking mid-range jumpers forces defenses to respect and defend the mid-range area; defenses can’t simply crowd the paint and three-point line. In this way, mid-range shots can be thought of as analogous to loss-leaders; even if a team’s mid-range shots derive slightly less value than three-pointers, they can increase three-point shooting percentages.
Let’s look at the San Antonio Spurs. Their coach, Greg Popovich, said he “hates” the three-point line and said of it, “I don’t think it’s basketball. I think it’s kind of like a circus sort of thing.” That alone shows one expert opinion on the value of the mid-range jumper. Popovich boasts 5 NBA championships (1999, 2003, 2005, 2007, 2014), 3 NBA Coach of the Year awards (2003, 2012, 2014), and 4 NBA All-Star Game head coach appearances (2005, 2011, 2013, 2016). His Spurs have consistently committed to mid-range shooting throughout Popovich’s 24-year tenure as head coach, and this has helped them claim their 5 titles. But let’s take a closer look at a more recent Spurs example.
Between the 2018 and 2019 seasons, the Spurs kept all major players and coaches intact with one change: they added DeMar DeRozan. DeMar DeRozan takes a lot of mid-range shots and almost no threes. The following tables summarize his shooting in 2019 (first year with the Spurs):
Ignoring the concept of gravity, we’d expect the Spurs’ three-point shooting to get worse with the addition of DeRozan and his abysmal 3P%, based on the first graphic of this article showing three-pointers have more expected value than mid-range shots. But that’s not what happened. In 2018 — before acquiring DeRozan — they were ranked 17th in team offensive efficiency. In 2019 — Derozan’s first year — they jumped to 5th in offensive efficiency. Let’s take a closer look at their team three-point stats in 2019 and how they changed from the previous season.
The addition of DeRozan greatly increased not only the Spurs’ offensive efficiency but also their three-point shooting from the previous season. It is also striking that the Spurs have the two ranked players in mid-range frequency (proportion of one’s shots from the mid-range) and three in the top 10. DeRozan and Aldridge also ranked first and second in total number of mid-range shots attempted in 2019. Defenses now have to worry about not only threes and layups, but also mid-range shots, causing defenders to gravitate toward the mid-range area. The three-point line subsequently opened up, explaining these players’ significant gains in 3P%. This huge volume of mid-range shooting had an interesting effect on the team as a whole. While fourth from the bottom in three-pointers attempted in 2019, the Spurs led the league in team 3PT%. Their mark of 5th in the NBA in offensive efficiency is particularly impressive, because the NBA is a league dominated by teams with two or more superstars (the Spurs have none). The Spurs show the mid-range can ignite an offense by gravitating defenses away from the three-point line.
“Playoff Defense”
The casual NBA fan knows the playoffs are a different game than the regular season. For one thing, strategies that work well against the basement dwellers of the NBA standings won’t inflate team’s stats. Every team in the playoffs has, well, made the playoffs and is over a .500 record. Defenses exert more energy, and defensive gameplans are more intricate. This hurts teams with static strategies, like the Rockets taking all three-pointers and no mid-range shots. Defenses latch onto trends and eliminate them.
This helps explain why Houston performs so much worse in the playoffs. Another metric we can look at is effective field goal percentage (eFG%). eFG% adjusts FG% for the fact three-pointers are worth more than two-pointers. In 2018, the Rockets were second in the NBA in effective field goal percentage in the regular season with an eFG% of 55.1%. In the playoffs, when teams played harder defense and pressed the three-point line, they faltered. They had only the 8th best eFG% in the playoffs with 51.4%. This happened partially because they refused to adjust when defenses overplayed the three-point line. They have seen similar drops in each of the past 4 postseasons. One specific example is Game 7 of the 2019 Western Conference Finals against the Golden State Warriors. When Houston was up by 15 points, Golden State adjusted and shut down the three-point line, daring them to take mid-range shots. Morey’s Rockets refused — listening to the analytics and not the natural dynamics of the game — and continued to launch contested threes. The result? They missed 27 straight threes and lost the game. Some might argue this was a fluke with bad timing, that Golden State didn’t lock down the three, but the Rockets just happened to get unlucky. However, a binomial distribution shows this is extremely unlikely. The following binomial distribution is based on the Rockets’ 2019 season 3P% average of 34.8%.
As you can see in the distribution, the probability of them missing 27 straight threes without GSW playing tighter defense is low — 1 in 166,000 to be exact. This was no fluke. Golden State shut down the three-point line, and Houston could have remedied this by adjusting their offense to the new defense — taking mid-range shots. But they didn’t, because they were focused on over-simplified statistics saying mid-range shots are always bad.
Here’s an example of when a team did use mid-range jumpers as a strategy in response to playoff defense locking down the three-point line. In the 2019 playoffs, the Philadelphia 76ers covered Kawhi Leonard tightly on the three-point line but sagged to the paint the moment he stepped inside the line. Leonard and the Raptors adapted, and he wrecked the 76ers by shooting 65% from mid-range and averaging 38 points per game in the first four games of the series before the 76ers were forced to abandon their defensive strategy. This 65% shooting represents an expected value of 1.3 points per shot, far and away better than average three-pointers. Here is this value compared to the original chart:
Within the natural flow of the game, the Raptors found an unquantifiable weakness in the 76ers defenses. While analytics descriptively say the shot was valuable, the prescriptive use of aggregate statistics would not have said to take mid-range shots. That is why Morey’s Rockets crashed out of the playoffs to the Warriors and the Raptors defeated the 76ers. The Raptors-76ers series end with Kawhi hitting a buzzer-beater winner in the fourth quarter of game seven, from the mid-range. Using mid-range shots can be an effective strategy that can win teams playoff series.
Three-pointers are an important tool, but so are mid-range shots. Unlike Billy Beane’s sport of baseball, a mostly static game with short, quantifiable bursts of activity, basketball is more of an art. It’s a dynamic game with more unquantifiable variables. Kevin Durant describes it well.
Latching onto descriptive stats — like three-pointers being the most valuable shot — do not extrapolate well in the real world where every play has a slightly different feel and different openings. It also leaves these teams without practiced options when elite defenses counter their strategies in the playoffs. It doesn’t allow for intuition to creep into the game. Analytics certainly play an important role in basketball, but let this article not only encourage NBA teams to take more mid-range shots but also warn them to be wary of extreme analytical recommendations.
In summary, let’s look at the three mid-range shooting teams I examined above in terms of championship success. The Raptors won the championship this year. The Spurs, as mentioned above, won championships in 1999, 2003, 2005, 2007, and 2014. The Warriors won championships in 2015, 2017, and 2018. All in all, 9 titles in 20 years to the Rockets 0. Interestingly, all of these championship seasons have ties to the soft-spoken professor of the mid-range game, Greg Popovich. Warriors head coach Steve Kerr worked as his assistant, and Kawhi Leonard learned the NBA game as his player. If other NBA teams listen to the actions and achievements of the quiet Popovich and his followers more than the outspoken Morey, perhaps they will find better success competing for championships.
Data Sources
stats.NBA.com
Basketball-Reference.com
pbpstats.com
ESPN.com
twitter.com
Note: unless otherwise stated, facts come from ESPN.com
Created by Tanner Probst for Professor Prasanna Tambe’s Analytics & the Digital Economy Course using RStudio
