The Raptors went away from the lineups that worked, and it cost them Game 4

Game 4 in the Raptors-Wizards series was a little frustrating, to say the least. The game went well enough for the first half, the Wizards got hot in the third quarter, and then the Raptors completely fell apart in the fourth. You read all the grizzly details here.

Let’s take a look at what the Raptors did well early on in the game, then poorly later. And, as always, figure out whether there is anything we can learn from this game.

Off to Great Start

As has been the case of late, the starters came out of the gates ready to play, and built an early lead. Unfortunately, Jonas Valanciunas picked up some early fouls, so we were treated to some of those usually-disastrous mixed lineups that feature Serge Ibaka and Jakob Poeltl together. But lo and behold (thanks to OG Anunoby playing very well, and staying on the court with that pairing rather than the usual substitution of C.J. Miles), they won those minutes too! (If by a narrower margin than the starters were managing.) The DeMar DeRozan-plus-bench unit closed out the quarter without relinquishing any of the lead, and the Raptors hit the locker room up eight points.

Then something odd happened. The Raptors went directly to the Kyle Lowry-plus-bench unit to start the second quarter, a move they absolutely need to make while Fred VanVleet is out, and haven’t been doing quickly enough in past games. The Raptors covered the first five minutes of the second quarter and pushed the lead from 8 to 10.

Unfortunately after that, the line mixing started. Valanciunas and DeRozan checked back in, but Pascal Siakam and Miles stayed in the game. We’ve covered this, but these transitional looks are almost always bad and although sometimes they are needed (like in the first quarter with Valanciunas in foul trouble), it wasn’t needed here. The Raptors survived the short stretch, breaking even (having an extra offensive possession helped), and went back to the starters. Sadly, that was short-lived, as after only one offensive possession, Valanciunas picked up another foul and was relegated to the bench again. Throw in Anunoby twisting his ankle a few minutes later (he would leave for the half but would come back to play in the third), and the lineups were a mish-mash the rest of the way. Overall, Toronto held on, and managed to enter half time with an 11 point lead.

That’s a lead you’ll take at the half. It came from the starters, the Ibaka-Poeltl transitional lineup with OG that worked, and the Lowry+bench unit, for the most part. Other transitional lineups mostly washed out.

But The Third Quarter

So now we reach the part of the game where it went wrong. The Raptors starters came out in the third with an 11 point lead, played seven minutes, and were broken up with a three point lead. That’s not a great run for a unit that has been the strength of the team so far this series.

Washington scored 26 points in those seven minutes, a disastrous rate. The Raptors actually did well to score their 18 points in the same time frame, a good rate as well, just not quite enough. Let’s look at how the Wizards scored their points in those 7 minutes.

2-of-3 inside 10 feet
2-of-5 from midrange
5-of-6 from three point range
3-of-3 from the line (fouled on a three point attempt)
2 turnovers

The Wizards got hot. They only had three looks inside the paint out of 17 possessions (on the season they had 44% of their attempts come from in the paint). They were taking long jumpers, and with the exception of the Raptors committing a foul on a jump shooter once, they were just hitting a crazy number of those shots.

So, that’s a rough stretch. It’s possible that if the Raptors didn’t overreact to it, they would have been fine.

Instead, coach Dwane Casey substituted Miles in for Valanciunas (leaving OG at the 4 beside Ibaka, a look I actually like when matching up small). Delon Wright would also come in for Lowry a minute later, then Siakam for Anunoby as they slowly transitioned to the bench unit. The defense was not much better in this stretch — the Wizards scored 14 points in less than five minutes. But the worrying factor was that the offense Washington was now getting should have made the situation a lot worse. Now, instead of hitting tons of threes and jumpers against the Raptors starters, the Wizards were getting to the rim.

Let’s look at the same breakdown of the Wizards offense for that stretch.

4-of-6 inside 10 feet
2-of-3 from midrange
0-of-0 from three point range
2-of-3 from the line

They hit one more midrange jumper than you’d expect, but the Wizards did not need to rely on hitting jumpers to score on the Raptors anymore. The Raptors overreacted to the hot shooting and gave up the rim, allowing a parade of looks inside.

At this point, the two teams are tied. Now to the fourth.

The Final Frame

Just like in the first half, Casey makes the right call and goes immediately to the Lowry+bench look. After the first five minutes of the quarter, the Raps held onto a two point lead. Then they got it up to an 8-point lead, but then a couple of shots by Washington (a 28-footer by Bradley Beal and a long midranger from John Wall) fell and closed the gap. It was about time to make some substitutions anyway.

DeRozan and Ibaka checked back in for Miles and Siakam. This was... not ideal. This created another of those Poeltl-Ibaka transitional lineups that have never worked well (except for a couple of minutes in the first half of this game, with Anunoby out there to help them out). Surprise of all surprises, the Raptors fell apart. Over the next five minutes, the Raptors scored six points, and gave up 12.

Here’s the glorious Raptors offense during that run:

Lowry miss pull-up three
DeRozan make midrange jumper
DeRozan miss midrange jumper
DeRozan miss midrange jumper
Poeltl make midrange jumper
DeRozan miss short jumper
Ibaka turnover
DeRozan miss layup
DeRozan miss pull-up three pointer
DeRozan make layup
Lowry miss layup
DeRozan miss step-back three pointer

There is now 1:33 left, the Raptors are down by three, and Siakam subs in for Poeltl, then a play later Miles subs in for Siakam. Those small lineups play the final 1:33, over which the Raptors scored two points on four possessions and the Wizards scored seven points.

Now, obviously it is easy to throw stones and criticize the Raptors’ coaching staff for the substitutions made and style of play after the fact. But could they have seen this coming?

Lessons Learned?

The end of this game was frustrating not just because the Raptors lost, but because it was so predictable, upon seeing who was on the court.

The starters had a bit of a rough game, but most of their on-court losses were because of that hot shooting stretch from the Wizards. This is the Raptors’ best lineup (hands down with no Fred VanVleet options to consider) and to not have it play at all for the final 17 minutes is a very bad choice. Of course, we knew this ahead of time.

Lineup | MP | ORTG | DRTG | RTG

KL-DD-OGA-SI-JV: 737 minutes, 114.3 ORTG, 102.8 DRTG, +11.5 net rating

That’s the third most used lineup in the league this season, and with an impressive net rating to boot — of the 44 lineups league wide with 200+ minutes played, the Raptors’ starting lineup ranks ninth in net rating. This is a far cry from past seasons where the starting lineups were break even at best and far more often had significantly negative net ratings.

Heck, we covered some of the stuff they finally did right in this one, like the Lowry plus bench unit.

I didn’t include any single Lowry bench units, as none have played over 21 minutes on the season (due to Lowry’s generally reduced minutes). But those looks have also been a tremendous success whenever used. Lowry has played 163 minutes this year with none of DeRozan, Ibaka or Valanciunas on the court, and has a +22.5 net rating in those minutes. Additionally, looking at those small sample individual lineups, there are seven lineups with at least eight minutes played — and every one of them has a positive net rating.

But the second piece I wrote on the playoff rotation focused on two things: the transitional lineups, and the closing lineup.

Ibaka and Poeltl have played over 200 minutes together, the vast majority of which came with both Lowry and DeRozan, and posted a negative net rating, on the strength of a 114.9 defensive rating. It just hasn’t worked. And this is no residue from early season struggles — since the All-Star break the pairing has been even worse, posting a 125 DRTG and -10 net rating in over 50 minutes played together.

Yeah, that pairing actually played to a break even rating in this one, but most of that is driven by the small sample success they had beside Anunoby. In crunch time, Anunoby was nowhere to be seen, and the pairing crashed and burned as expected.

Speaking of which, what did we have to say about crunch time lineups?

VanVleet and Anunoby both seem to be key defensive players in the clutch. Anunoby’s sample sadly comes from only six games from early in the year, but if there are match-ups where you need a bigger wing defender in crunch time and Fred VanVleet doesn’t make sense on that end, Anunoby has had success when he’s been used, and as we noted, closing with your starters makes sense anyway.

Well, with no VanVleet available, this was Anunoby’s shot! Considering how great he has been in the series so far, it would be shocking if he didn’t play in the fourth at all. (It’s fair to note here that OG may have been struggling with a sore ankle, but we’ll never quite know how much.)

One other thing that stands out is how the offense absolutely dies when Siakam or Poeltl play. With reduced spacing, the defense can key on Lowry and DeRozan too easily. Leaving aside Anunoby’s few games of great success, the key to the offense appears to be playing Valanciunas, and to a lesser extent Ibaka. Meanwhile, the defense with or without Valanciunas has performed about the same in the clutch. Another clue pointing to playing the starters together late, or at least most of them.

The offense dying late when Poeltl or Siakam play? Who could have seen such a thing coming?

And on the final few plays, with the Raptors down by six, playing a small ball unit that relies on guys hitting jumpers to stay afloat, and desperately needing a three pointer, who should the shot come to but Serge Ibaka. Just the guy you want taking that shot late in a game having played 34 minutes on one day of rest, right?

Ibaka in the 4th Q:
30-of-60 from 2-point range
10-of-13 from the free throw line
3-of-19 from three point range

Right. Yes, he’d need to be taking shots if he’s on the court regardless, but with a small unit like that, those shots are the backbone of your offense, with no real threat inside (or to grab offensive rebounds after missed shots).

In theory the point of going small is to be really good defensively. But we’ve seen all year that playing small/mobile in closing minutes with Poeltl at C, or with Ibaka at C, is a disaster overall and on the defensive end. And so of course the Raptors posted a DRTG of 145 in the final seven minutes of the fourth quarter after moving away from the Lowry-plus-bench unit. It’s not entirely surprising, considering most of those minutes were with Poeltl at C, and on the season Poeltl posted a 128 DRTG in his limited clutch time appearances (and the poor pairing numbers for Ibaka and Poeltl were already quoted above).

And then there is the issue of DeRozan’s late game play. The ball movement disappeared, the player movement disappeared, and the offensive success disappeared. There’s a lot of reaction online about how the Raptors reverted to their old selves in crunch time. But it’s not really true. This is just their current selves when they play lineups with no roll threats to build their pick and roll offense out of.

This season, in the fourth quarter, DeRozan had the following numbers playing beside each of these players.

Poeltl and Ibaka: 41% usage, 48% TS%, 92.5 team ORTG
Ibaka and Siakam: 40% usage, 58% TS%, 106.4 team ORTG
Valanciunas and Ibaka: 34% usage, 59% TS%, 118.4 team ORTG

DeRozan quite simply has always had to (or chooses to) do too much playing with those combinations. Give him a non-threat like Poeltl and he takes on too much, getting bad shots for his trouble. Give him a little more space with a small frontline and he can score more effectively on his own, but he still takes on a lot himself and the team only scores moderately well. Give him a real screen setter and roll threat, as well as Ibaka’s shooting, and he doesn’t have to do it all himself, still scores well, and the team rolls.

Sadly, the answer to whether we learned anything from this game is no. The Raptors succeeded in ways they have always succeeded. The Raptors failed in the ways they’ve always failed. Two of the Raptors’ best players didn’t play a single minute in the fourth quarter, and as a result, they lost. The Raptors relied on a look that has always failed them, and as a result, they lost. The Raptors played lineups that encourage their stars to try to win by themselves, the stars tried to win by themselves, and as a result, they lost.

Hopefully, although we didn’t learn anything from this game, someone did. And if that someone could fix the Raptors’ rotation, that would be great.