Stat Guides: Julius Randle's Breakout Season

What do the numbers tell us about Julius Randle's leap? The second in a series of Cleaning the Glass Stats guides, from contributor Rohit Naimpally.

By Rohit Naimpally

This is one in a series of guides that will hopefully be helpful in demonstrating how to use Cleaning the Glass data to analyze players and teams. Whenever a specific stat is referenced, a video will demonstrate how to locate and parse out that number on cleaningtheglass.com/stats. Unless specified otherwise, all data is from Cleaning the Glass, and all stats were current through the end of the 2020-21 regular season.

Very few teams in the 2020-21 season outperformed their pre-season expectations as much as the New York Knicks. First-time All Star Julius Randle won 98 out of a possible 100 first place votes in running away with the award for Most Improved Player. What changed in Randle’s game and how sustainable is his big leap?

Randle has always displayed the ability to create baskets for his teammates, assisting on nearly 16% of his teammates’ made shots over the prior three seasons. In the 2020-21 season, Randle elevated his playmaking to assist on over a quarter of his teammates’ baskets.

That assist percentage places Randle in the 97th percentile for all bigs, up there with elite playmaking bigs like Bam Adebayo and Domantas Sabonis.

Further buttressing Randle’s case for being a bigger creator in the Knicks offense is that a larger proportion of his made threes in 2020-21 were unassisted. Despite Randle taking more of his shots from beyond the arc than ever before, he was assisted on his made threes at a career low rate. Most importantly, Randle made his threes at a career high 41% clip. Put simply, he created, took, and made more threes.

Worth noting is that this isn’t a case of Randle swapping out long two pointers for threes; he took even more of those long twos in the 2020-21 season. The increase in those long range jumpers came largely from a decrease in attempts at the rim. Under a quarter of Randle’s shot attempts came at the rim, the 11th percentile for all bigs. 

Randle’s move towards the perimeter is especially noticeable when one contrasts his shot chart from the 2019-20 season with the corresponding chart from the 2020-21 season.

Ordinarily, one might question the prudence of a big moving his shot profile away from the rim. But at 6’9” with a below-average wingspan for his height, Randle is not a stereotypical big. He arguably had a bigger margin to improve on as a jump shooter rather than as a rim attacker. Randle has historically been a poor-to-average finisher at the rim, ranking in the bottom quintile for bigs in both the 2019-20 and 2020-21 seasons. 

Even with a head of steam, Randle can struggle to finish over less than imposing big men.

Nevertheless, even though Randle is less efficient at the rim than most bigs, he still shoots a significantly higher percentage on those attempts than he does on long twos. The shift in Randle’s shot profile means that even with his improved three point stroke, his effective shooting percentage remained roughly the same as in the previous season (49.7% eFG% in 2019-20, 51.8% in 2020-21.) 

And therein lies the rub for how sustainable Julius Randle’s offensive leap is: his efficiency on mid-range shots and rim attempts is roughly in line with previous seasons. That 41% mark on threes is well above Randle’s career averages. There is little reason to think that Randle’s playmaking and self-creation will regress, but his ability to be an All Star level offensive force may hinge on whether his 2020-21 three-point shooting is a permanent shift or an outlier. Is his three point shooting in his MIP campaign Marc Gasol’s 2018-19 season (permanent), or Draymond Green’s 2015-16 one (temporary)?

Rohit Naimpally is the Innovation Team Lead at The People Lab, where he works on research to strengthen the public sector and the communities it serves. You can find his basketball writing at From The Logo, and can follow him on Twitter @rohitnaimpally.

Get notified when I publish something new: