In the late 19th century, two men raced to discover as many fossils as possible, and to put them together in recreation of as many dinosaurs as they could. O.C. Marsh and Edward Drinker Cope despised one another, and they went to great lengths in their competition, including destroying fossils and putting bones together in ways that never actually happened. For example: the Brontosaurus — it never existed, but Marsh smushed the bones together and claimed it was a new dinosaur species in order to have two dinosaurs to his tally rather than one. The way we put individual items together to tell a larger story matters. (Much like pointillism!)
There’s another, similar story: The parable of the blind men and the elephant. There’s a group of blind men who each feels a different part of an elephant, and then they describe their findings to one another to recreate the animal. One thinks it is a snake, another a wall, another a tree trunk. So on and so forth. There are many points to the story, but one of them is that limited descriptors are not the same as an overall classification.
Which brings us to Fred VanVleet. I’ve written a huge amount about him this season — well, every season, really. But this season, there are a large number of limited descriptors one could apply to his performance.
- He adapted his role, scaling down his offensive output to let others run the show.
- (His role has since changed, and changed again, and his pick and rolls with Jakob Poeltl are now the offensive engine of the team.)
- By his own admission, VanVleet was particularly poor during an important stretch of the year when Toronto was losing games it needed to win: “Being the leader of the team, you know, that whatever the team is doing is just gonna follow me,” he said. “So I was not playing well, to my standard at a time where we needed to win and we were losing.”
- His 3-point slump continued from midway through the 2021-22 season to January of this year. Even though he’s firmly out of it now, his 3-point percentage on the year is only 35.0, the lowest of his career.
- On that note of the jumper, he has largely been more effective with movement coming into his shot — pulling up or relocating — than as a completely static shooter.
- Coinciding with the return of his jumper in January, he’s been playing like a star for most of 2023.
- His point of attack defense was shoddy through much of the season, right up until March. It has now turned around in a big way, but not enough to change his overall impact for the season on the defensive end to a positive.
- He has been, largely in compensation for his shooting woes, a driving wizard for stretches. (He has been one of the best and most efficient drivers in the NBA in March, shooting 55.6 percent while attacking the hoop.)
- In many ways because of the drop off in his jumper, he has been at his most effective with the ball in his hands. That has created roster-construction issues because the same has been true for all of Toronto’s leaders.
- Broadly, he is not selfish. In fact, he’s been somewhere between very good and excellent as a playmaker. And since the acquisition of Poeltl, he’s been excellent full stop, with the 16th-best assist rate in the league.
That’s a whole lot of body parts to an elephant. But we must be cautious not to build a Brontosaurus out of these components. What image do all these disparate dots create?
Well, if you go by advanced stats, this has probably been the best season of VanVleet’s career. For methodology, I’m using the best five single-number metrics as ranked by NBA staffers and media members in Bryan Kalbrosky’s excellent piece from 2021:
If you limit it to the offensive end, his improvements have been even larger. When I asked Nick Nurse to try to explain this being VanVleet’s best offensive season in the eyes of the catch-all metrics, he pointed first to VanVleet’s good run of health over the last stretch. When I asked VanVleet the same question, he talked about being bad to start the season. So not a lot of unique insight from either of those sources.
Advanced metrics have a host of problems (especially the ones that are measuring something as disparate and undefinable as ‘value,’) and I’m not using these to say that he’s definitely better this year than last. (Our investigation into that hasn’t even begun hehe; this is a real behemoth of a piece.) These catch-all metrics aren’t the end of the argument but the beginning of it. They are picking up something, though. Let’s try to find what it is.
First, a few concessions. Although VanVleet’s defense has picked up recently, it’s probably clear that he was better last season than this. (And in prior seasons as well.) He remains fantastic as a help defender, shooting gaps with his heavy, meat-hook hands and surprising ballhandlers across the court when they turn their backs. He gets strips under the rim that function like blocks — and are sometimes, to his chagrin, counted as them. He’s a great defensive rebounder at the guard spot. He’s sturdy in the post against switches. But you know what? He did all that stuff last year. His point-of-attack defense last year was mostly excellent (until the end of the season, when injury limited him dramatically), and is has been mostly problematic this season (until recently, when it’s seemed like health has improved his performance dramatically). His defense has definitely dropped compared to last year, even if he’s been solid (or better) for a few weeks now. So let’s keep this argument to the offensive end.
It’s also clear that he’s been a lesser shooter this year than last. His 3-point percentage has dropped, of course, but he’s also shot 26.3 percent from deep in the clutch, which is one of the worst marks in the league. The Raptors, of course, are in the midst of a much worse season than they were last year.
What, then, are the catch-all metrics seeing in VanVleet?
For one, his play creation has improved. There’s no perfect (and easy) way to measure passing — which Samson showed by hand-analyzing every pass Scottie Barnes has thrown this season to measure which did and did not create advantages. (I discussed the nuances of VanVleet’s passing with Samson here.) If you look at simple assists per game, or assist rate, VanVleet has certainly improved this year. If catch-all metrics aren’t going to prove a point, then passing metrics won’t either, but it’s worth mentioning that Crafted, for example, has VanVleet as the 35th-“best” passer this season. (I can’t vouch for their reliability here, but their wingspan database is accurate as far as I can tell.)
Teammates don’t shoot well off of VanVleet potential assists, but that could simply be because the team is full of relatively poor shooters. But he also has the 60th-ish-highest shot quality on potential assists, depending how you filter for volume of passes, so he’s not creating completely the most open, high-value shots either. (The leaders there include Trae Young, Steph Curry, Luka Doncic, and Nikola Jokic, so there’s some correlation between elite playmaking and shot quality on potential assists.) But whether he’s in the top 30 or top 60 as a playmaker, his work there is certainly the best its been of his career, especially since Poeltl joined the team.
VanVleet-Poeltl pick and rolls are one of the absolute best high-volume combos across the association (ranking below virtually only Luka Doncic pairings with his multiple bigs). And if you break down what’s actually happening during this action, he’s shooting more frequently in picks with Poeltl than he is in the same actions with players like Pascal Siakam, Barnes, O.G. Anunoby, Chris Boucher, or other partners. On top of that, he’s also averaging a higher frequency of assists on such plays. He’s able to shoot more and assist more with Poeltl because Poeltl creates so much more space with better angles, timing, and simply being bigger, and VanVleet uses that space to get more open shots and at-rim attempts.
Poeltl rolls so well (he’s always below the ballhandler defender!) and presents so well to the ball that VanVleet has much easier passing lanes this roller, as well as a roller who is much better at turning touches into points. Their craft together presents threats to defenses during every stage of the play, and the setup pass and knockout finish can come from all sorts of different angles.
It’s hard to say if VanVleet is better because this is the best pick-and-roll partner he’s ever had, or if he’s better because he tangibly improved. Probably a little bit of both. But either way, it doesn’t matter: The result is the same, in that VanVleet has never been more foundational offensively, and he’s never before been able to be the engine of an entire half-court offense to such an extent.
VanVleet himself attributed the increased performance there simply to opportunity: “Coming into the NBA, I’m gonna do whatever they asked me to do to get on the floor and first and foremost, and that was to play defense and shoot open threes,” he explained. “And so as your role develops and you grow as a player and there’s more opportunity then great. “Once Kyle left and we went kind of more small and there wasn’t as many pick-and-roll opportunities with a traditional five… Then we got Jak whose one of the best bigs in the league as of late…and I feel very comfortable in pick-and-roll situations, and I think it’s looked good for our team.”
That’s he’s running a career high frequency of pick and rolls is significant, too; he’s replaced some isolation possessions with more dynamic actions. He’s never been an efficient isolation player, so it’s good that he’s using fewer possessions there. (He’s actually among the lower-frequency isolation players within the category of high-volume initiators.) And he’s the most efficient he’s ever been in pick and rolls. It’s a good tradeoff.
The VanVleet-Poeltl partnership has also coincided with a return to elite marksmanship from VanVleet, but it’s worth mentioning that his individual scoring wasn’t all bad before Poeltl became a Raptor. Even though VanVleet’s overall 3-point shooting may be down this season, he is still scoring more effectively in a number of other ways. In fact, virtually every area of scoring for VanVleet is at or near his peak, outside of his catch-and-shoot 3-point shooting.
VanVleet’s pullup-shooting might be the most impactful area of scoring improvement for him. On pull-up 2-pointers, he ranks 13th in the league in field-goal percentage among players with at least 100 such attempts, ahead of players known for such shots, including Chris Paul, Steph Curry, DeMar DeRozan, Luka Doncic, and others.
He’s become extraordinary at using hesitations and crossovers to get his hips past his primary defender, as well as altering the actual path the ball takes during his process of shooting a jumper, as well as his release point to get his shot over the help. His jumper is more portable and flexible than it’s ever been in his career, and that’s very apparent in the midrange.
He’s never been this efficient on pull-up 2-pointers. He’s setting a career high in efficiency on such shots, and he’s just below the mark for frequency he set last season. On pull-up 2s out of pick and rolls, he’s shooting over 50 percent — when you factor in that Toronto scores less than a point per play in the half court, that means a VanVleet pull-up 2-pointer out of the pick and roll is lifting Toronto’s offensive efficiency. It’s not a bailout shot so much as a good look defenses might regret giving up. (Considering VanVleet is 12th in total double-teams commanded this season, defenses increasingly are not giving up his pull-up 2s.)
“He’s that efficient,” said Washington Wizards Head Coach Wes Unseld jr. when I asked him about the appeal of double-teaming VanVleet. “At some point, you have to get the ball out of his hands.”
“He shoots the ball off the pick and roll pretty well. Not a big guard, but he plays downhill, and once he gets into that paint, he does a really good job of using his fakes, his pivot game, and just kind of holds and collapses the defense. I think that opens up opportunities for threes, cuts, dump-offs to the bigs.”
VanVleet used that very approach — getting downhill out of the pick and roll, pivoting in the paint, and switching the angle for an and-1 — against Unseld’s Wizards in Toronto’s handy 114-104 win. VanVleet finished with an efficient 28 points.
And VanVleet is even more efficient on pull-up triples out of the pick and roll than he is on pull-up 2s. It’s by a slight margin, and he’s in the top 20-25 of the league in that regard rather as opposed to the top 10-15 for pull-up 2s, but it’s still a real tool that constitutes a good look for Toronto’s offense. It’s not the best the Raptors are going to get out of a possession, but neither is it a bad look by any means. If he played a full game handling in the pick and roll and firing every shot, from 2 and 3, at the rim (without the defense adapting), Toronto would statistically have a better offense than its average fare.
VanVleet’s pull-up triples in transition, however, are just about the best weapon in basketball, with an effective field goal percentage over 70.
This is by far VanVleet’s best season as a pull-up shooter in transition, both in terms of efficiency and frequency. An effective field goal percentage over 70 on any individual shot type is Giannis-at-the-rim type stuff.
It’s important to distinguish, then, between areas of shooting in discussion of a slump for VanVleet. As a catch-and-shoot shooter, he’s in the midst of the worst season of his career. But as a relocation catch-and-shoot shooter, he’s right at the top still compared to his past years. And as a pull-up shooter, when you include 2s as well as 3s, this is the best year of VanVleet’s career. Because standstill catch-and-shoot shooting remains the brunt of VanVleet’s attempts, his overall percentage has dipped. But he has slowly started to target the shots that remain high-value for him and drifted away from shots without any movement into them. His percentages have skyrocketed in the new year.
He’s done much more to benefit the Raptors on the offensive end. Because his pull-up shooting is so dangerous, VanVleet has more and more open lanes to the rim than ever before. This is also his most efficient year driving the basketball.
The Raptors are scoring more points on a possession with him driving the ball than ever before. That is despite him blowing by defenders less frequently and having help defenders in the region more frequently. That’s only possible because he’s improved across the board. He’s shooting a higher percentage on drives than he has since his sophomore year, and he’s turning it over at a career low. He’s getting blocked less often at the rim and shooting 65 percent at the rim when it’s not blocked, a career best.
VanVleet is sporting career-best offensive on/offs despite the down year from deep. His offensive on/offs have leapt to such an extent that they’re now the best on the team and 13th in the league among players with at least 2000 minutes played.
That is partially because he’s improved in so many areas, and partially because Toronto has crippled its initiation depth over the years, with VanVleet the only player on the roster capable of doing a number of things crucial for a half-court offense. That latter point, that VanVleet is unique on Toronto’s roster, is another sneaky component that helps explain why catch-all metrics love him; they rely heavily on on/off stats, which are as much about who is replacing an individual when he sits as what that individual does when he plays.
Still, VanVleet is doing so much that even if he is being helped by context, the metrics are capturing real value. He has shifted his role slightly multiple times throughout the season, and even if his scalability hasn’t been what the Raptors might have wanted because of his cold streak from deep, he’s still found ways to contribute no matter what. His shifting more and more of his jumpers to ones involving movement — relocating, or pull-ups — rather than stationary catch-and-shoots (which he’s hitting now anyway) reflects real awareness of his game and its changes. From those shot changes to increased driving or pick and rolls, more or less off-ball play, and fluctuating usage, VanVleet has been rolling with the punches. And perhaps more than any other Raptor this season, he’s been asked to change his game.
Yet he’s now doing what he does best — running the offense as a dominant on-ball player, a la Kyle Lowry. Coincidentally, it’s also the first time in VanVleet’s career that he’s been at his best in this role. He’s been a little worse at off-ball wizardry and a little better at floor generalmanship; small tweaks in pull-up shooting, pick-and-roll creation, and finishing filter upwards.
Part of the reason general perception of VanVleet’s season is so different from the numerical judgements is because of the change in his role — in how his tweaks have filtered into the overall value that his game offers. He built his name as a low-usage and high-efficiency play finisher. He’d hit his triples, do the little things, and generally impact the game invisibly, doing some good and almost no bad.
He now does lots and lots of good — and some bad, as well. That’s more important! That’s also a normal thing — when players become veterans, they take on more responsibilities, and when players make countless more decisions in a game, of course more of them will be wrong. That’s life when role changes. I wrote about how fans’ experiences with Kyle Lowry factored into judgements about Lowry’s replacement, VanVleet. But it’s also worth mentioning that experiences with VanVleet himself colour current fan-player relations. When he was so successful in a different role — and the Raptors won a championship with VanVleet doing less — that indexed into memories. And VanVleet surely could still do less. But if he did, there’d be no one else to do what the Raptors are currently asking of him. There’d be no one to run pick and rolls, to hit pull-ups, to fuel the offense. And VanVleet’s efficiency might be better for it, but the team would be significantly worse. Nick Nurse said exactly that when I asked about VanVleet’s shifting role this season:
“Right now, [with him in his current role,] the shots we’re creating, the shots we’re getting, all those things, are pretty high level,” he said. “The numbers are showing that. The [expected points per possession] are showing it even more. Expected points per possession are off the charts. We’re trying to maximize what we can do as a group.”
That’s where a season-long evaluation of VanVleet should end. The team is at its best with VanVleet currently doing what he’s doing. That hasn’t always been the case this season, but VanVleet’s improvement and the addition of Poeltl seem to have changed the calculus.
How should one combine those fossil records into one beast? There are so many parts to VanVleet’s season, so many disparate components, that it would be easy to mistake the trunk or the legs for the full beast, VanVleet’s cold shooting streak or his early-season low-usage role for his entire season. But in reality, (on the offensive end) his good play has overwhelmingly outweighed his poor play on the year. For example, he has as many games (nine) scoring over 30 as he does under 10. Whether you better remember the high-scoring explosions or the low-scoring letdowns depends on how you index to him as a player. Preconceived expectations factor here.
And significantly, the Raptors are on the whole better with VanVleet dominating the ball — even if that wasn’t true for a stretch. They are 5-4 in games with him scoring over 30 and 1-8 when he scores under 10. Or, if you want to balance away from the bias of VanVleet’s actual scoring, the Raptors are 6-4 in the games in which VanVleet runs the highest rate of pick and rolls and 4-6 in the games in which he runs the lowest rate.
There have been down stretches of VanVleet’s year. There have been missed shots and technical fouls, rants against referees and subtle complaints about his role. Those are components of his season, yes. But to think they describe the full year is to believe an elephant just a snake because of its trunk. The games in which VanVleet has been dominant, in which he’s driven Toronto’s success, far outweigh his negatives. At this point in the year, he has the most significant role of his career, and he’s playing the best he ever has. Don’t succumb to the politics of archaeology, to the same mistakes that Marsh and Cope made. Don’t build a Brontosaurus in describing VanVleet’s season, misattributing his contributions. If you separate his season from expectations, the fossil records speak with clarity in how they fit together: This has been the best season of VanVleet’s career.
