Cold snap: Digging into Toronto’s shooting woes

The Raptors made a big trade in the off-season. They sent out DeMar DeRozan, their star wing who could not hit a three (31% success rate on almost four attempts per game last season) and Jakob Poeltl, their backup centre who attempted two long range shots all season long. They received Kawhi Leonard, he of the 38% career 3PT% (on nearly four attempts per game), and seemingly as a throw-in also got Danny Green, a career 40% 3 point shooter (on nearly five attempts a game).

Their three point shooting, which ranked 18th in the league (36%), should have been solved.

And yet here we stand, and the Raptors sit 22nd in three point shooting, at only 34.5%. Let’s try to figure out why.

Types of Shots

First, let’s take a big picture look at things. One of the most fundamental ideas with three point shots is you don’t want to just throw as many as possible up — you want open shots, you want catch-and-shoot shots, you don’t want to be tossing up contested step back threes (with few exceptions from some star players).

So, what has changed since last year in that sense? It’s been well-reported that the Raptors are attempting more threes than even last year, when they saw their totals increase pretty significantly. They’ve gone from 24 attempts per game two seasons ago, to 32 last season, and now to 33 this season.

Let’s look specifically at the change over the past two seasons. Where did those extra attempts come from?

Well, in 2017-18, the Raptors attempted 24.5 catch and shoot threes per game, hitting on 37% (8.9) of them. They attempted 7.6 pull-up threes per game, hitting 34% (2.6) of them. 16 of their 32 attempts came as wide open attempts (6+ feet from the nearest defender), on which they hit at a 39% clip. Only 3.2 attempts per game came in the final 4 seconds of the shot clock, a scenario where they only hit 24% of their threes.

This season, they have attempted 24 catch and shoot threes. Note the similarity to last season’s total. They’ve hit on 36% of those threes, an almost identical mark to last season.

They have, however, increased their total threes — which means their pull-up threes have been coming more often. They are up to nearly nine attempts per game, on which they are shooting only 30% — which amounts to 2.6 makes, exactly the same number as last year in spite of taking more than 1 extra attempt per game. So, puzzle piece number one: they are taking more off the dribble threes, and getting zero extra return on them.

Let’s keep going though. In terms of how many of their shots are wide open, there has been a significant change there — last year it was almost exactly half of their attempts. This year, a full 18 of their 33 attempts (55%) have been wide open, and they’ve hit 39% of those. That’s just as good a conversion rate as last season, on a higher volume. So they are getting better looks at the rim, and still hitting well when they do.

But looking closer at the degree of contest on their non-wide-open shots shows an area of concern. Last season, they attempted 5.4 tightly contested threes. This season, only 3.9. That means more of their non-wide-open threes are in that middle ground where a defender is between 4 and 6 feet away. And yet, last season they shot 33% on all non-wide-open threes. This season? 29%. For whatever reason, in spite of taking less well covered shots, the Raptors are hitting fewer of them.

Last big picture thing: those last second shots. Last season, they only took 3.2 of them per game. This year? Nearly four per game, with a conversion rate of 22%. The offence is generating shots later in the clock, and they aren’t hitting well on those at all (which is to be expected).

So, overall, the concerns seem mostly to be the Raptors forcing up more pull up threes compared to last season, while hitting at a worse rate on them; having a harder time hitting on any threes that aren’t wide open; and the offense is too often getting bogged down and forcing them to take threes late in the clock.

Now, let’s take a look at the players and see where these issues are most obviously presenting themselves.

Catch-and-Shoot vs. Pull-Ups

So, if the team as a whole is taking too many pull-up shots, that means some players are taking too many pull-up shots. So, who are the biggest culprits?

Player | Pull-up 3PA/game | 3PT% on pull-ups
Lowry, 2.7 attempts, 29%
Leonard, 2.6 attempts, 34%
VanVleet, 1.7 attempts, 29%
Green, 1.0 attempts, 32%
Wright, 0.7 attempts, 28%
Powell, 0.7 attempts, 30%
Miles, 0.6 attempts, 29%

So. Leonard is hitting at a decent clip on pull-ups, but otherwise everyone else is under a point per possession on these looks. It’s hard to imagine there are benefits to the offence stemming from this aggressive off the dribble shooting that can make up for posting essentially a 90 ORTG on those shots.

The most obvious culprit is Lowry, who we will get to in a second, but the number of attempts coming from players who simply should never be taking these sorts of shots is astonishing. C.J. Miles should not be dribbling the ball, let alone shooting right after doing so. Fred VanVleet taking 40% of his attempts off the dribble is a crime for a player who should be playing off-ball (especially considering he is shooting 42% on catch and shoot threes). Same goes for Danny Green: he is shooting 45% on catch and shoot threes, is not a player who should ever be creating off the dribble, and yet throws up one pull-up attempt a game.

Compare that to the types of threes you want players taking, and you get this list:

Player | Catch-and-shoot 3PA/game | 3PT% on catch-and-shoot 3PA
Green, 4.4 attempts, 45%
Lowry, 3.9 attempts, 34%
Anunoby, 2.9 attempts, 34%
Miles, 2.8 attempts, 31%
VanVleet, 2.7 attempts, 42%
Ibaka, 2.5 attempts, 27%
Leonard, 2.1 attempts, 41%
Siakam, 2.0 attempts, 34%

Even on a team with some lacklustre shooters, that’s six of the top eight shooters with greater than a point per possession on catch and shoot shots, and three shooters with great return on good shots in Green, VanVleet and Leonard. Note in particular the difference for VanVleet, who should always be playing off ball; this shooting disparity is reason enough but far from the only reason.

Contested Threes

Let’s do the same for wide open threes — which players are not getting their share of wide open looks, and more importantly, which players are suddenly shooting poorly on the somewhat contested looks that have bottomed out this year?

Player | Contested 3PA/game | 3PT% on contested 3PA
Lowry, 3.9 attempts, 26%
Leonard, 3.5 attempts, 33%
Green, 3.0 attempts, 37%
Miles, 1.9 attempts, 26%
VanVleet, 1.7 attempts, 25%
Powell, 1.4 attempts, 37%
Anunoby, 0.9 attempts, 28%
Ibaka, 0.8 attempts, 13% (!!!)

Yet another list of guys taking shots you’d really rather they not take. Why Serge Ibaka would be taking nearly one somewhat contested three per game is beyond me, especially considering that success rate. Same goes for someone like Anunoby. Miles has dropped off in his ability to hit those contested looks — he shot 33% on the same shots last season, and is well down from that now. Though it is worth noting that they only accounted for 29% of his three point shots last year, while this year a full 55% of his shots are not of the wide open variety. That’s a dramatic difference in shot selection, and explains a lot of his struggles this year. He is shooting only 35% on wide open threes this year, but 35% is a sight better than 26%.

Again we see VanVleet taking a lot of these less ideal shots, and missing badly on them. While again we see Leonard and Green mostly unfazed by the tougher shot selection. In a surprising twist, Powell is showing up again on the list of tough shot hitters, which is a nice skill set for him to flash, as that’s something the bench needs badly and will go a long way to him keeping that job over Miles.

And yeah, we’ll get to Lowry...

Last Second Shots

Last category, I promise. This one is quick. Unlike the Raptors’ offence, apparently.

Player | 3PA in final 4 seconds of shot clock | 3PT% on late clock 3PA
VanVleet, 0.9 attempts, 24%
Lowry, 0.7 attempts, 23%
Powell, 0.6 attempts, 23%
Leonard, 0.5 attempts, 15%

These are essentially the guys most often getting screwed by some bad offence and having to try (and mostly fail) to bail out the team offence.

Okay, enough is enough. We’ve seen that some players are not taking the shots they should be, and the team efficiency is suffering for it. And the team needs to fix that. But let’s get to the elephant in the room.

Kyle Lowry

Lowry is taking all those pull up shots, and missing a lot. That’s not good. But you may recall, that’s his shot, right? Last season, Lowry took the 9th most pull up threes in the league, and hit 40% of them. This year he’s dropped from 3.4 of those shots per game to 2.7, but more notably has seen his efficiency on those shots drop to 29%.

You may also remember that Lowry started off the season on absolute fire from long range, shooting over 40% through the first 11 games. He cooled off after that, but even after 22 games he was holding up at 36% from downtown. That had dipped to 33% by the time he got hurt. After those first 11 games, he shot 28% from downtown until his injury against Golden State in mid December. He’s shot 30% since coming back from the injury.

A pattern that has existed for a while, and has continued this season, is how dependent Lowry is on solid screens to get clean looks from the perimeter. It’s hard to ignore that the Raptors’ offence fell off a cliff essentially the moment Lowry and Valanciunas got hurt, and hasn’t completely rebounded since Lowry’s return. The Raptors are a borderline top 10 offence since Lowry’s return, were a bottom 10 team while he was out, and were the 2nd ranked offence prior to the injuries.

Looking closer at Lowry’s performance in particular with and without Valanciunas paints a pretty significant picture. On the season, Lowry has a 30% three point percentage without Valanciunas on the court with him. He has a 39% three point percentage with Valanciunas on the court.

Some of this is buoyed by that early season hot streak, but if we break it down by month, the pattern persists. In those first 11 games, Lowry had a 52% 3-point percentage beside JV, and a 34% percentage without him. So even in that early run, all Lowry’s hot shooting was happening beside the big man. Perhaps this is just luck, of course.

Looking at the next segment of games, after the early streak but before the injuries, as Lowry’s shooting fell off, we still see a split, though not as big, in his shooting with and without JV — his three point percentage plummeted to 31% with Valanciunas, but even further to 26% without him. That back injury may have been affecting him even then, but you could still see the impact a solid screener had on his game.

Since that time, Lowry has been settling in at 30% from long range. He doesn’t see the same impact from playing with Greg Monroe, another big body but not the same screen and roll threat — his 3-point percentage is almost identical with and without Monroe since his return from injury.

There’s no guarantee Lowry will find his shooting again, but based on the season to date, his best chance (assuming the team already plans to do whatever they can to make sure his back gets and stays healthy) may be to get some significant minutes playing with the big Lithuanian when he returns.

Lessons

So, what have we learned? Well, Lowry is having a heck of a tough time shooting the ball, and JV may help him there but there are no guarantees on that front. Still, worth trying.

Perhaps a bigger impact of Valanciunas’ return may be an improvement in the offence in general. The offensive struggles over the past couple of months have driven the team to poorer and poorer shot selection, with contested and off-the-dribble three pointers making up too much of the team’s shot spectrum. If the return of a true inside threat brings those ratios back in line with what brought them (relative) success last season, one would expect their improved roster of shooters compared to last season to bring their shooting percentage up significantly.

And if Valanciunas isn’t the answer, it will fall to Nick Nurse to find a way to get his players to find and take better three point shots. Far too high a proportion of their shots seem forced, perhaps as part of a mandate to take more threes. The team needs to find their long distance shots more organically within the offence, without being crowded by defenders or having to create those shots themselves.

And if the team thinks the answer is to go out at the trade deadline and acquire another shooter, the percentages above should make them think twice. The issue seems to be more about the shots rather than the shooters right now, so any upgrade in that area will likely have its effect dampened, and therefore the holes opened up by the outgoing pieces in any trade may actually hurt the team more than the extra shooting helps.

All stats per NBA.com.