Every year I try to put together a playoff rotation based on what we have seen the Toronto Raptors implement throughout the year. This year is particularly tough, as they’ve rarely had a full complement of their preferred rotation players, with the mid-season trade for Thaddeus Young just adding another barrier to the sample size available.
That said, we’ve seen enough things work that we should at least be able to put together some broad rules for what has worked and what has not heading into Round 1 the playoffs against the Philadelphia 76ers.
So let’s get to it.
So we’ve got five players who are clearly the organization’s preferred starting five, even though the fit of those five as a lineup is questionable, and it’s because these are simply the five best players on the team.
- Fred VanVleet
- Gary Trent Jr.
- OG Anunoby
- Pascal Siakam
- Scottie Barnes
Play each 48 minutes and problem solved. Got awful close to that for long stretches this season, but let’s not be that silly until we get pretty deep into the series.
So, with that in mind, there will be some bench lineups. Or at least lineups without some of those guys.
It’s the playoffs, so we don’t need to find a 5-man unit to get all those guys rest at once. And heck, for most of this season, we’ve thankfully seen Nick Nurse try to have a couple of starters out there with the bench at any given time, a strategy that really helps bench players stay in their supporting roles and have an opportunity to excel playing alongside your top players.
So, the real candidates for inclusion in the playoff rotation outside of the above five guys are:
- Precious Achiuwa
- Chris Boucher
- Thaddeus Young
- Khem Birch
You might notice that’s just 4 PF/Cs. And, yeah, it is. Welcome to the Raptors.
As much as I enjoyed Dalano Banton’s early season chaos, and am a big fan of Malachi Flynn, neither have played well enough for long enough to have earned looks in the playoff rotation, unless we see foul or injury trouble (and maybe not even then).
The most common bench unit used this year (particularly late this year) is some combination of starters with Precious, Boucher and Thad all playing at once. It’s real big and with a real wide array of skillsets in spite of the similarly sized players. With Precious and Thad both flashing a little bit of shooting, and Boucher being quite capable of shooting the three in spite of a not great 3PT% this season, it’s a lineup that can create and score in a variety of ways depending on which starters are out there with them, and can defend pretty much anyone.
As for starters to play with the bench... Scottie Barnes has been the guy they have subbed off early in those few games where they’ve been healthy enough to start the small ball starting five. So he tended to be used as a creator with the bench, and I have to imagine that will continue, either in place of one of those three forwards or alongside them.
Generally speaking, there are two guys who really drive success for the team among those starters, and those are Pascal and Fred. Each of them is generally tasked to support a bench unit to buy the other guy rest minutes.
I went and looked at what players each starter most often shared the floor with when playing at least beside Boucher and Precious, the almost-season-long bench pair that the Raptors don’t seem to want to go away from, in spite of Precious’ in-season improvements maybe indicating he could get another crack at starting. Here are the most common teammates for Fred and Pascal in those scenarios, and how well those units have done.
Fred: Scottie (+30), Gary (-26), OG (-7)
Pascal: Gary (+27), Scottie (+52), OG (+9)
So, if we accept that Scottie will be there a lot (as he should), the OG and Fred pairing and the Gary and Pascal pairing make sense, from a relative success perspective.
As a general check...
Pascal with Gary, without Fred and OG: +6.1 net rating
Fred with OG, without Pascal and Gary: +0.4 net rating
Break even lineups are non-ideal, but there’s a lot of noise there with Fred and OG carrying a heavy load early in the season with no Pascal. But break even is not terrible either, and a halfway decent bench unit should be able to build leads while the 76ers try to find rest minutes for Embiid.
The bench rotation actually seems like the easy part. The Raptors know what they want to do, they’ll throw one of Fred or Pascal out there, with some combination of rangy forwards including Precious and Boucher.
The only wrench that might get thrown is if they start Precious to try to hard match his minutes to Embiid’s as a primary defender. Some good signs if that plays out, where Birch could step into the role:
- Boucher’s best pairing among all steady rotation players: Khem Birch (+14 net rating)
- Young’s best pairing among all non-starters: Khem Birch (+9 net rating)
- Pascal with Gary and Khem, without Fred and OG: +1.5 net rating
- Fred with OG and Khem, without Pascal and Gary: +21.2 net rating
All those with smaller samples than the above overall numbers, but still positive signs that the bench shouldn’t suffer if Precious is pulled into the starting lineup.
For the record I am very much a proponent of doing exactly that.
But if the bench units are easy, what’s the hard part?
Right. So, back to these five guys:
- Fred VanVleet
- Gary Trent Jr
- OG Anunoby
- Pascal Siakam
- Scottie Barnes
Those guys have played a lot together, comparatively anyway. More than any other lineup on the team. They’ve played 345 minutes together, the equivalent of just over seven 48-minute games of basketball, and they’ve won their minutes... by 6 points.
That’s .500 ball. Which is great if you want to break even against a .500 team, but the Raptors are trying to match up with the 76ers and MVP candidate Joel Embiid. Embiid and Harden (as proxies for the 76ers starting group) have roughly a +10 net rating since they joined forces, and so you’d expect them to cave in a +0.5 net rating lineup like the Raptors’ starters.
Now, the Raptors never did get a chance to try their small group against the 76ers, so we don’t know for sure whether it would work in this specific matchup, but the overall success of the two lineups is not a great starting point. One thing they did try was starting both Khem and Precious together (a truly baffling decision), in the final two matchups with Philly. The Raptors got eviscerated in those minutes (literally lost the 12 combined 1st Q minutes between the two games by TWENTY SEVEN points, 41-14), so should never do that again, but managed to win both games anyway, so what do I know?
But it gets worse. That small ball starting five is actually a pretty good lineup, as you’d expect with all those good players, outside of one specific scenario.
Unfortunately, that scenario is “at the start of games.”
The group’s overall net rating is +0.5. Their net rating in 1st quarters is -12.9.
That’s... bad. Real bad. And certainly not the roaring-out-the-gates performance you’d like to see from your, uh, starting lineup.
Roughly a third of their overall sample has come in the first quarter, over 100 minutes played together. This is not a blip, and it was a trend throughout the entire season. The group just doesn’t work to start games.
My personal theory is that if you are going to play small, you need to be on an absolute string defensively, and that is tough to do right at the outset of a game, no matter how prepared you are. Some level of in-game rhythm is required to really be on that string.
Offensively, that lineup falls prey to letting Siakam fade into the background as the ball ends up in any of the five players’ hands, and that’s simply not the ideal for an offence with Pascal freaking Siakam available as a creator.
Pascal carries a 25% usage and 22% assist rate on the year. Pascal’s usage dips to 20% with that lineup, and his assist rate to 17%. It’s not him being more of a playmaker — he’s just got the ball less. It’s a process that maybe is helpful to reduce load over the course of 82 games, but in a playoff series it’s not going to hold up.
Later in games they’ve have much more success, as you can tell by their break even overall net rating, and part of that is ramping up Pascal’s usage later in games naturally and part is the defence clicking a little better, so expect to see plenty of this lineup regardless of whether they start. They’ve only been awful to start games.
That said, every time the team has had that lineup available, they have started them. Maybe they go back to that well in spite of the evidence and maybe it succeeds. If it doesn’t, it may dig them a hole too deep to climb out of in a playoff series where they are already the underdog.
So, there are basically two real big picture options to consider.
One is the option the Raptors have defaulted to, both often as a starting lineup when a starter is hurt, and also as the quick substitution they used in the middle of the season when the starters were healthy for a decent stretch, to get Barnes switched off to play with the bench. That’s Khem Birch.
Birch has had... mixed success with the starters. His samples are all small, the bigger ones are not great in terms of results. Most of his minutes with the starters, both overall and in 1st quarters, have washed out to roughly a -5 net rating. That’s really not great. But consider that we are comparing to a starting group that a) is eating up minutes from all five of our best players, and b) are getting caved in at nearly three times that rate in 1st quarters.
There are promising caveats. His success behind OG is much better than without OG, though in smaller samples (as OG was the starter who missed the most time). This is not unexpected, Birch is a low impact guy you plug into lineups where you just don’t want him to hurt you, as compared to a guy like Precious who can cause issues or solve them depending on the night (and stage of the season, to his credit this is far less an issue now than early in the year). But remove OG from the starting lineup and your perimeter defenders are Barnes and Trent, who are both better than expected in their own right and can be pests defensively, but are prone to gambling rather than staying in front of their guys, and that doesn’t play into Birch’s sound positioning approach to defence. A much better fit with Precious’ ability to switch and help from anywhere.
But with OG there, and one of Gary or Barnes out of the lineup, you’ve got a perimeter defence that allows Birch to settle in and just not hurt you.
In any case though, in any context, a starting lineup with Birch, arguably the 9th best player on the team, is going to be something like a break-even lineup. You are still starting with a group that is not ideally suited to go against the best lineups of the opponent. But you are avoiding the wasteful small ball five opening minutes, and buying some minutes in the rotation.
You are also preserving the Precious and Boucher bench pair, which seems to be a priority for the team. I question how much that should be a priority (as playoff series go deep you tend to see your best players dominate the minutes and see less and less of your usual bench looks), but it seems to be one.
If it isn’t however, that’s option 2.
Let’s just start Precious. Birch has seemed fine off the bench, even better in many situations than Precious has.
Precious had some half decent success early in the season starting alongside the other starters, and he was a far inferior player at that point.
On the year, these lineups have been quite good, even with that early season sample playing a large role:
Fred-Gary-Scottie-OG-Precious: +8.4 RTG
Fred-Gary-Scottie-Pascal-Precious: +13.4 RTG
Both those lineups have played over 150 minutes on the year. It works. Throw in the ideal of matching up Precious to Embiid defensively and it seems like an easy answer. The risk is that they haven’t tried a look with OG available - but OG generally makes things better, not worse, so that’s not much of a risk.
We might not see the Raptors start the series like this, but unless their small group pulls off a miracle and completely flips the script (which is an outcome we shouldn’t write off, Nick Nurse has pulled off miracles before), they should end up with this look at some point. The hope is that they don’t drop games they didn’t need to before they get to that point.
All stats per nba.com.