How have the Raptors most used lineups held up, and can they be improved?

We tackle this idea every season — which combinations of players have worked well for the Raptors this year, which have not, and what lineups should we see (and not see) in the playoff rotation?

Let’s start at the beginning — of the game that is.

The Starters

It is a great relief to be able to write a very short opening section this year. Previous seasons had seen the Raptors trot out starting lineups that were terrible (largely due to starting the worst power forward on the roster) and needed to be fixed before the playoffs. Last season due to Kyle Lowry’s injury and the deadline deal for Serge Ibaka, we hadn’t even gotten a chance to see the starting lineup play together at this stage of the season.

This year, the Raptors managed to stumble onto the right answer less than a month into the season, and have stuck with it since. Oddly enough, the right answer was not one that was considered a candidate entering the season. The starting point guard, shooting guard, power forward and centre were pretty much given, with the question mark coming at small forward.

Much of the discussion was about whether Norman Powell or the sharpshooter C.J. Miles should fill the hole in the starting lineup left by DeMarre Carroll. OG Anunoby wasn’t expected to be ready to play at the start of the season, and no guarantee to even make the rotation when he was back.

But after Powell struggled mightily to start the season, and OG’s early return to the bench unit was a great success, the rookie got his shot and hasn’t looked back since.

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

MP: Minutes Played
ORTG: Offensive Rating, points scored per 100 offensive possessions
DRTG: Defensive Rating, points scored per 100 offensive possession
RTG: Net Rating, ORTG minus DRTG, essential point differential per 100 possessions

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.

As for alternatives to OG, on the season he’s held up as the best option. Put Powell in his place and the lineup has a -4.5 net rating. Miles has gotten only a limited shot there (34 minutes) but has posted only a break-even net rating (+0.8). Delon Wright could be considered a possibility as a big guard, but they’ve struggled with him there too (-8.6 net rating in 61 minutes). The only other option is Fred VanVleet, who has seen just over 100 minutes in that lineup (mostly in closing minutes for each half) and has posted a tremendous +24.6 net rating. But that look is probably best not used right from the outset.

So, let’s say the Raptors starters are set, and we have a nice alternative option for later in games, and move on.

The Bench

We covered this in depth last week, but as a reminder, here are the most used bench lineups this season, along with the most used DeRozan bench unit.

Lineup | MP | ORTG | DRTG | RTG
FVV-DW-CJM-PS-JP: 299 minutes, 118.4 ORTG, 98.2 DRTG, +20.2 RTG
FVV-DW-NP-PS-JP: 99 minutes, 104.8 ORTG, 95.6 DRTG, +9.2 RTG
FVV-DD-CJM-PS-JP: 92 minutes, 121.1 ORTG, 109.9 DRTG, +11.2 RTG

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.

So no worries here. Moving on to some worries.

Transitional Lineups

One pattern that is consistent for the Raptors is staggering their substitutions. Typically Valanciunas and Anunoby will be the first starters to sub off, with Poeltl and Miles often checking in (though VanVleet, Siakam or Wright are also possibilities). Later, Ibaka will check out leaving DeRozan and Lowry playing with the bench front court, before transitioning to the DeRozan bench unit to end the 1st and 3rd quarters.

Similarly, after running the bench unit out for the first few minutes of the 2nd and 4th, typically some combination of Lowry, Ibaka and Valanciunas will check back in, with DeRozan re-joining them later (and VanVleet never checking out).

These transitional lineups are used less than the pure starting or bench units, but collectively make up more than half the minutes played on any given night. And they’ve yielded more varied results than the very successful units we’ve covered so far.

Here are the most used lineups this season that have at least two starters and two bench players in them, plus a few with one substituted bench player as well.

Lineup | MP | ORTG | DRTG | RTG
KL-DD-CJM-PS-JP: 52 minutes, 96.0 ORTG, 110.1 DRTG, -14.1 RTG
KL-DD-OGA-SI-JP: 49 minutes, 106.0 ORTG, 122.8 DRTG, -16.9 RTG
KL-DD-CJM-SI-JP: 48 minutes, 110.7 ORTG, 103.8 DRTG, +6.9 RTG

Those are the only lineups that fit the criteria with significant minutes played. And those are some bad results overall. But given the small samples, let’s try looking at some pairings to see what combinations should be tried and avoided when using transitional lineups.

Top pairings with bench players for each starter that is likely to be in a transitional lineup:

Kyle Lowry | MP | RTG
FVV: 424 minutes, +9.7
Wright: 327 minutes, +7.5
Miles: 375 minutes, +8.2
Siakam: 519 minutes, +9.2
Poeltl: 453 minutes, +7.3

DeMar DeRozan | MP | RTG
FVV: 526 minutes, +12.6
Wright: 494 minutes, +3.6
Miles: 531 minutes, +2.4
Siakam: 661 minutes, +5.3
Poeltl: 599 minutes, +2.6

Serge Ibaka | MP | RTG
FVV: 296 minutes, +9.3
Wright: 307 minutes, +5.0
Miles: 205 minutes, +6.5
Siakam: 160 minutes, +11.1
Poeltl: 212 minutes, -4.2

See an outlier in that list? Yeah, me too. 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.

Alternatives

That pairing needs to stop. But what can it be replaced with? The first obvious option would be to substitute Ibaka off at the same time Poeltl comes in, but that leads to the top lineup outlined above, which has been a disaster as well — the bench frontcourt has yet to prove they can handle playing the minutes that Lowry and DeRozan are trusted to tackle together.

So, the second option is to try another look with Lowry, DeRozan and Ibaka still on the court. Assuming Valanciunas and Anunoby are still checking out, who else could check in? Well, based on those pairing numbers above, Siakam and VanVleet sure look like strong candidates. The issue is that VanVleet already plays a lot of minutes at the back end of each half, but let’s explore what lineups have featured those players anyway.

Lineup | MP | ORTG | DRTG | RTG
KL-DD-NP-PS-SI: 34 minutes, 110.2 ORTG, 97.5 DRTG, +12.8 RTG
KL-FVV-DD-PS-SI: 23 minutes, 98.4 ORTG, 99.8 DRTG, -1.4 RTG
KL-DW-DD-PS-SI: 21 minutes, 119.1 ORTG, 92.4 DRTG, +26.8 RTG

That’s all the options with 20 minutes played, which is a very small sample as it is. But the early returns on these small looks seems to indicate that using VanVleet at the same time as the small frontcourt might leave them too small. In any case, it hasn’t worked in a small sample and it’s hard to see how the rotation would work with VanVleet checking in so early, so we’ll stick with the other two looks. Adding a defensive guard who can shoot and attack a little seems to be an effective way to unlock these small ball looks. Wright would likely be the first choice, but should Powell be called upon at some point we should keep these lineups in mind.

But those small looks have also been used sparingly this year, mostly in situations where it makes sense — to match up with smaller opposition. So if it makes sense in-game, they should certainly try these looks rather than the early Poeltl and Miles minutes. But what if it doesn’t?

At that point we have to question the players subbing off.

Which is what we will do next week leading into the playoffs, along with tackling the closing lineup question as well.

All stats per NBA.com, as of April 2nd, 2018.