Jayson Tatum, #0, Duke Blue Devils, (Photo by Lance King/Getty Images)
What do we really know about the science and development of jump-shooting? How important are free-throws, floaters and touch as indicators of shooting potential?
No fluffy intro this time. For that, go read part one of the shooting synopsis, where I covered volume and versatility as they relate to projecting jump-shooting for NBA Draft prospects.
A quick synopsis of part one: volume is the king of the shooting indicators, the most critical piece of the puzzle that is shooting evaluation. Above all else, great shooters shoot. But volume can fail without the necessary addition of shot versatility, which helps a shooter reach a far higher level of true impact than one that is a spot-up artist.
As a reminder, this series is more analogous to a series of case studies than hard statistical research (though this part will have more of that than the first one). We’re observing shooting development through the lens of one prospect’s shooting profile, Justise Winslow. During his lone season at Duke, Winslow’s shooting profile looked like this:
3PT%: 41.8 percent (46-of-110, .309 3-point attempt rate)
2PT%: 51.6 percent (127-of-246)
FT%: 64.1 percent (100-of-156)
Long 2PT%: 17.6 percent (15-of-85)
Close 2PT%: 68.7 percent (112-of-163)
Catch-and-shoot: 62.2 eFG% (39-of-94), 1.245 PPP, 87th percentile
Off-dribble: 13.2 eFG% (2-of-19), 0.263 PPP, 2nd percentile
Off-screens: 20.8 eFG% (2-of-12), 0.538 PPP, 11th percentile
Spot-up no-dribble jumper: 65.4 eFG% (34-of-78), 1.308 PPP, 87th percentile
Runners: 19 eFG% (4-of-21), 0.318 PPP, 10th percentile
Five seasons into his NBA career, he’s shooting 33.7 percent on 3-pointers (190-of-563) and 33.3 percent on long 2-pointers.
In part one, we outlined Winslow’s lack of shooting versatility and elite volume as points of failure for his shooting projection to the next level. Defenses don’t sell out to Winslow’s heavy diet of catch-and-shoot triples, lacking the pull-up/off-movement accuracy and volume to damage opposing coaches’ sleep patterns with any sort of significance. That means even a reasonable 3-point percentage doesn’t offer much of a floor-spacing effect.
Is free-throw percentage an indicator of shooting upside for NBA Draft prospects?
The other obvious negative data point in Winslow’s college shooting profile is his low free-throw percentage at 64.1 percentage. Free-throw percentage has long been lionized as the skeleton key of shooting translation: prospects who don’t shoot free throws well won’t shoot 3s well at the NBA level. In some circles, there’s pushback against the true value of free-throw percentage, questioning the relevance of an indicator that doesn’t make much intuitive sense in its relation to 3-point shooting. The truth, in my research, is somewhere in the middle of these two parties, as it tends to be.
As a general rule, very good free-throw shooters will be good NBA shooters. Very bad free-throw shooters will not be good NBA shooters. As with all rules, there are exceptions, but this general concept holds true. To illustrate this, a simple query. Since 2008, there are 221 seasons of drafted players (excluding very low volume) who shot equal to or below Winslow’s 64.1 percent from the line. The vast majority of those prospects are either non-shooters or non-impactful shooters.
As with all of these lists, there are some notable exceptions. Joe Harris, Chandler Parsons and Avery Bradley all have good-to-great volume as indicators in their favor. DeMarre Carroll, Marcus Morris, OG Anunoby and Trevor Ariza (who is outside the scope of the barttorvik database but fits these criteria) have broken the trend without a plausible statistical explanation, as outliers do.
Reversing this list predictably proves the opposite: the vast majority of good to great free-throw shooters on volume, picking 84 percent as an arbitrary cutoff point, end up as good to great NBA shooters. Most of the failures here are prospects who just weren’t good enough to crack the NBA — Marcus Paige, Andrew Goudelock, Nigel-Williams Goss — many of whom ended up as very good shooters overseas. There are other outliers — Matisse Thybule lacked volume, Mikal Bridges has his obvious issues — but the combination of high free-throw percentage and 3-point volume is fairly bulletproof as shooting projection goes. Many of the best NBA shooters – Steph Curry, Paul George, Damian Lillard, CJ McCollum, Klay Thompson and more shared these statistical indicators in college.
What happens to the 20 percent with a free-throw percentage between 64 percent and 84 percent? While there isn’t a hard cutoff point for free-throw shooting mattering or not, it does seem to operate on a sliding scale of sorts, with the likelihood of impactful shooting increasing as free-throw percentage does. Less than 55 percent or so is a death sentence, 55-65 percent is tough to overcome, 65-82 percent has increasing levels of variance, above 85% gives you a great chance to shoot well in the NBA. Again these specific numbers are arbitrary and aren’t important. What is of importance is the general range, not specific cutoff points.
Free-throw percentages in the 70s, high 60s and low 80s bring a lot of predictive variance on their own, to the point where other indicators are necessary. In general, high volume and versatility override mediocre free-throw percentage. Looking at prospects who crossed the 125 3-point attempt threshold in a season, there are more shooters than non-shooters among the low to mediocre free-throw percentage group. Some of the failures include DeAndre Bembry, Michael Gbinije, Sam Dekker and Dennis Smith Jr. (and Markelle Fultz, with the largest possible asterisk). Successes include the aforementioned Joe Harris, Zach Lavine, Isaiah Thomas, Kentavious Caldwell Pope, Patrick Beverley and Tim Hardaway Jr. with quite a few more on the list.
Again, the opposite is true. Without the boon of volume, it becomes more difficult to project positive and impactful shooting from prospects sporting middling free-throw percentages. But as always, exceptions arise. Take Pascal Siakam, a career 71.9 percent free-throw shooter in college with 17 3-point attempts over two years. Four years into his pro career, Siakam is bombing 8.1 3s per 100 possessions at 36 percent, blossoming into a legitimate off-dribble shooting threat. What gives?
In college, Pascal Siakam excelled in a critical indicator for pull-up shooting development especially: long 2s. In his sophomore season, Siakam shot 43.3 percent (126-of-291) on long 2-pointers, 52.4 percent of which were assisted. That’s a ton of unassisted long 2s and is likely a primary explanation for Siakam’s pull-up growth. At New Mexico State, he frequented these short mid-range shots:
We touched on Winslow’s lack of pull-up versatility in part one and the long-2 numbers check out here, only making 15 of his 85 long-mid range attempts.
A high volume of long-2s as an indicator for future pull-up shooting development makes intuitive sense. At its core, the pull-up mid is basically the same shot as a pull-up 3; there’s 1-to-1 relation, unlike a free-throw and a 3-pointer. Players who excel at the long-mid range can and often do extend their range to beyond the line.
Maybe the strongest example of this is Kawhi Leonard who, aside from some clear outlier development, shared a similar profile to Siakam in terms of long 2s (35.7 percent, 97-of-272) and free-throws (75.9 percent, 110-145). In fact, most of the league’s best pull-up shooters made long 2s in college (30 makes seems to be a solid benchmark, with the number of unassisted makes being more important than raw efficiency), with the notable non-injury related exceptions being Paul George (elite free-throw shooting) and Zach LaVine (an outlier worker by all accounts).
The most obvious future pull-up stud, and one that many missed on, is Jayson Tatum. At the time of the 2017 draft, many dinged Tatum hard for his poor shot selection and proclivity for difficult long 2s. What they failed to consider was just how strong his shooting profile really was and what those difficult shots meant for his difficult shotmaking upside. Billed as a prospect with limited upside by some, Tatum is the best wing pull-up shooter in the NBA in just year three.
Tatum was a great free-throw shooter on good volume (84.9 percent, 113-of-139), shot 117 3s and made 46 long 2s, a minuscule 6.5 percent of which were assisted on. At the time, scouts focused too much on his decision making and not enough on the difficult shotmaking goodness of a 6-foot-8, 18-year-old wing and its predictive power:
Many of these prospects who improved dramatically as pull-up shooters have one less obvious commonality: height. The concept of height correlating with shooting development may seem confounding initially, but it makes intuitive sense. A tall shooter has more room for error against the monster defenders of the NBA, they can shoot over contests and require less space creation to get shots off. And tall players generally need to be less skilled to make it in the NBA. The bar for small players to carve out NBA roles is astronomical. Small prospects who made tons of long 2s with other awesome indicators won’t be able to improve their shooting if they aren’t in the league. Players like Leonard, Siakam, Tatum, Khris Middleton and Paul George all had a much longer leash than the likes of Goudelock, Paige, Tyshawn Taylor and Kay Felder.
Are touch and runners indicators of shooting upside for NBA Draft prospects?
Another way to describe these prospects’ short to long mid-range success is by ascribing that success to their touch, a much-debated and nebulous concept. Touch is a difficult concept to define. I typically think of touch as a description of a prospect’s ability to make difficult shots, often manifested in long 2s or floaters, another key shooting indicator.
The oft-cited case of touch for shooting development is Luka Doncic, whose 3-point percentage before college was never great (and still isn’t, but that isn’t super pertinent for reasons discussed in part one), but his touch was always outlier elite, with ridiculous circus shots that Luka seems to nail with regularity:
I’d be remiss if I didn’t mention Cole Zwicker’s draft Twitter famous piece on runners and touch, in which he made the case for runners as a predictive shooting indicator. Many draft enthusiasts and scouts took this concept and ran, despite there being inherent issues with using runners as a projection tool, the main one being that almost no prospect will accrue a statistically significant sample of runners over his career. Cole clearly laid this out in his piece, which was far more cautious on the viability of runners as a predictive indicator as many remember.
The two examples Cole brought up were T.J. Warren and De’Aaron Fox, both players with a solid sample of runners and clear touch on tape from floater range. Warren exploded from 3-point range in year five out of seemingly nowhere, and Fox developed into a volume pull-up shooter in year two despite low college a free-throw percentage and 3-point volume (while Fox’s overall 3-point percentage regressed in year three, his pull-up efficiency actually didn’t regress too much, which brings us back to the discussion of impactful shooting and versatility from part one).
Though finding prospects with a genuine sample of runners is difficult, I do think runner success does matter as a piece of the puzzle. A crude scroll through high-volume, high-efficiency runner shooters drafted over the last decade or so brings up Jawun Evans, Elfrid Payton and Dejounte Murray as poor shooters and Warren, Fox, Trae Young, Monte Morris, Aaron Holiday, Josh Richardson and Jordan Clarkson, along with some of the best shooters in the world outside of the NBA in Shane Larkin, Scottie Wilbekin and Nigel Williams Goss.
Conducting any type of in-depth research on runners and shooting projection is nearly impossible due to the lack of data. I’m interested in the concept of runners as an indicator for latent shooting upside, in the case of Warren or someone like Shane Larkin, who wasn’t the excellent off dribble shooter he is now during his NBA tenure. With publicly available runner data only going back about a decade or so, we’ll need more time to see how this theory unfolds.
Now, let’s look at how we can apply all of this information to the 2020 NBA Draft.
The first place to start is Tyrese Maxey, who is often maligned for his poor 3-point percentage (29.2 percent). However, he’s one of the strongest bets to shoot in this class for many of the reasons outlined in this piece. He’s an 83.3 percent free-throw shooter, made 46 long 2s (17.4 percent assisted) and has a relevant sample of runners (39.1 percent, 25-of-64) with obviously elite touch on tape going back to high school:
Moreover, Maxey had to adapt to an alien role off of the ball and, given Kentucky’s tendency to deflate prospects’ abilities, he’s as good a bet to be an impactful shooter as many of the guards at the top of the class.
This class lacks any real wing initiator bets (a prime reason for its general weakness) in the ilk of a Tatum, Kawhi or Siakam. One sneaky candidate for pull-up shooting development is Devin Vassell, typically lauded just for his defense. He’s an underwhelming free-throw shooter (73.8 percent) with average 3-point volume (106 attempts) and garners the same decision making negatives prospects like Tatum did, often missing credit for his difficult shotmaking prowess. Vassell hit 49 long 2s this season (22.4 percent assisted), his high release point negating contests:
Vassell’s rapid development curve is another point in his favor, morphing into a volume pull-up shooter in his sophomore season after shooting exclusively spot-ups as a freshman. Another surprisingly strong wing shooting profile comes is CJ Elleby (82.3 free-throw percentage, 218 3PTA, 50 made long 2s, 14 percent assisted, who is the most similar statistically to a prospect like Tatum). He has a lot of other issues as a prospect, but the size and shooting make him worth tracking.
I can’t write a piece largely about free-throws without mentioning Deni Avdija, a notably terrible free-throw shooter, shooting a frigid 55.6 percent (109-of-196) from the line across all international competitions over the last two seasons and 57.2 percent in FIBA over the last four years. Deni’s volume is solid and his mechanics and 3-point percentage have notably improved this season, up to 33.5 percent from 3 on 179 total attempts. He still struggles as a pull-up shooter without any real off-movement shooting capability.
Avdija could be an outlier hard worker who has already shown improvement and I think developing a respectable 3-point shot is well in the cards for him, despite history being against him. Though I’m skeptical of his shooting ever reaching a truly impactful level given his bottom-barrel free-throw potential and lack of versatility. Aside from Avdija, Precious Achiuwa (59.9 FT%) and Jahmi’us Ramsey (64.1 FT%) are two other shooting black hole prospects in this class to me, despite Ramsey’s efficiency.
I could have included Patrick Williams in part one. He’s one of the most interesting shooting projections in this class, with low volume (50 3PTA), high versatility (especially considering he’s the youngest American prospect in the class at 6-foot-8, jacked and intelligent), free-throw percentage (83.8) and 26 made long 2s (11.5 percent) assisted. Williams is a good reminder of how difficult projecting shooting really is. He’s a prospect with no real historical proxies, as a prospect of his size, age, free-throw goodness, pull-up flashes and low 3-point volume doesn’t exist. The volume is a point of legitimate concern, but I lean positive on Williams’ shot given the bevy of other indicators in his favor, along with his pull-up shotmaking extending to his high-school days:
Josh Green fails the volume and versatility, launching 83 3s, without any pull-up or off movement diversity. He’s always been gun-shy, lacking the chucker gene of most great shooters. On the positive side, Green is a 78 percent free-throw shooter who made a fair amount of runners this season (18-of-46). Despite the touch and free-throw indicators, I don’t buy Green as an impactful shooter. He could be a player who makes spot-ups at a solid rate, but the lack of volume and versatility are too damning given his lack of elite free-throws or long 2s (30 made, 50 percent assisted) to compensate.