Nylon Calculus: Understanding offense-defense tradeoffs at each position
I had a discussion last week about building around point guards and where the best point guards derived the most value – on offense, defense, or going both ways. It inspired me to dig a little deeper into the side of the ball that contributes the most to each position and the implications that positional value holds for team building.
In order to explore this phenomena, I documented the offensive and defensive Real Plus-Minus (courtesy of ESPN) for the top 30 players at each position by total RPM. I then charted the distribution for the ORPM and DRPM in each position with a non-parametrized kernel distribution. Considering the relatively low skewness of the resulting distributions, I converted them for visualization purposes to idealized and smoother normal probability distribution functions, displayed for each position below.
The point guard RPM distribution functions corroborated my discussions last week. Only two of the point guards in the top 30 in total RPM this season have a DRPM greater than 1.0 — Chris Paul, clubhouse leader, mayor, and president, with a DRPM of 4, and Patrick Beverly, resident menace but barely making the cut at a DRPM of 1.01. Paul also has an ORPM greater than 1.0, Beverley’s is just 0.15. With point guards having the highest variance and average ORPM of any position, this indicates that there’s a lot of good-to-great players manning the point, a sign that teams don’t need to splurge to get adequate floor generals to run their offense. Being a two-way player is least important for the point guard spot. Of course, that makes Paul all the more special; in a position dominated by players who are offensive maestros but defensive sieves, he has no such weaknesses in his game.
Read More: The conundrum of the Miami Heat core
As we move up in position, offense matters less and defense matters more. This makes perfect sense, seeing as there’s been ample research to show that big men and interior defenders have the most influence on a team’s overall defense. The one position that has a true claim to being a two-way spot is the small forward. Think about how every team clamors for 3-and-D wings. In this era of pace and space, two-way wings are incredibly valuable, since they are also some of the most athletically versatile players on the court and can cover a lot of space on both ends. Players who’ve developed into the league’s brightest stars, like Kawhi Leonard, Jimmy Butler, and Kevin Durant, all fit this archetype.
That then brings us to the current renaissance of big men and the diverse array of skills that they bring. Power forwards and centers have the lowest variance in their distributions, an indication of the rarity of players who can lead an offense while defending capably, hence the “unicorn” moniker for the wave of skilled young big men.
As small ball becomes more and more popular, an elite big is rare (and enviable); rather most teams have a crop of smart, mistake-free centers who can rebound and anchor a defense at a simply sufficient level, like a Cody Zeller or Andrew Bogut. That’s why players like Jahlil Okafor, big men who are not transcendent offensive talents but are also utterly incapable of holding their own on defense, are simply not worthy of a top pick. Someone like Okafor is difficult to build a team around, and the NBA is stacked with a middle class of power forwards and centers that demonstrate at least a decent proficiency on the side of the court where they can offer the most value (for much less opportunity cost).
This understanding of positional value between offense and defense is instructive for team building. Teams like the Golden State Warriors and Cleveland Cavaliers, the prohibitive favorites to again challenge for a championship, have followed this recipe to great success. They both have explosive offensive talents in the backcourt (Kyrie Irving, J.R. Smith if healthy, Steph Curry, Klay Thompson), two-way wings that bring the entire lineup together (Kevin Durant, LeBron James, Andre Iguodala in the Death Lineup), and opportunistic big men who can capably anchor their defensive interior (Draymond Green, Zaza Pachulia, Tristan Thompson). Of course, no one will confuse Kevin Love for a Defensive Player of the Year candidate, but he’s also playing the best ball of his career due in large part to a much improved defensive aptitude, as he currently boasts a DRPM of 1.6.
Teams like Portland, on the other hand, have some soul searching to do. The Trail Blazers’ entrenched two best players are stud offensive guards. However, in their starting lineup, only Al-Farouq Aminu can claim to be a great defender, with a DRPM of 3.05. And he plays up at power forward when sharing the court with Mo Harkless, putting him at a size disadvantage at that end. With a lack of defensive support around Aminu, especially from the center spot (starter Mason Plumlee has a DRPM of 0.83), there’s no way for Portland to cover for their back court’s leaky defense. And what’s worse is that they’ve virtually locked themselves into this current setup for the near future.
Next: The secret to Kyle Lowry's success around the rim
Ultimately, team building in the NBA is an optimization problem with a multitude of parameters and constraints, and understanding something as simple as the tradeoff between offense and defense at each position can help towards navigating that resource efficiency puzzle.