clock menu more-arrow no yes

Filed under:

How to fix the Raptors’ broken starting lineup

New, comments

The Raptors struggle out of the gate, and struggle in third quarters. The starting lineup appears to be a problem. Can it be fixed?

NBA: Chicago Bulls at Toronto Raptors John E. Sokolowski-USA TODAY Sports

It’s become an all-too-familiar refrain this time of year. The Raptors are winning games, but mostly on the back of their bench unit, and the starters are struggling against top opposition.

A couple of years ago, it was Luis Scola starting all year long, with Patrick Patterson coming off the bench. Last year, Patterson again took a back seat to a rookie Pascal Siakam, who had not really found his feet defensively yet (his improvement there this year being one of the biggest positive story lines for the team). Both times, the starting unit found itself wallowing in misery (and significant negative net ratings) while the bench unit, and the starters when they played with Patterson instead of Scola or Siakam, found great success.

And so we (myself quite possibly the loudest) called for a change. And it finally came midway through the season last year, only to be quickly followed by injuries to DeRozan, Patterson, and eventually Lowry, with a trade for Serge Ibaka thrown in for good measure. We never really got to see that experiment play out — would Patterson have fixed the starting lineup? All signs pointed to yes at the time, but we don’t know. We didn’t even get to see Ibaka try too, with Lowry’s poorly timed injury.

So here we are again. Another year, another struggling starting lineup. Another bench unit killing other teams (in spite of the unit being completely different from the ones in years past).

How Broken Is It?

Let’s keep things in context here. How badly have the starters performed this year?

Well, there have actually been a bunch of starting lineups due to injuries, but the two that have played by far the most minutes are Kyle Lowry, DeMar DeRozan, Serge Ibaka and Jonas Valanciunas joined by one of Norman Powell or OG Anunoby.

Lineup | Minutes Played | Offensive Rating | Defensive Rating | Net Rating
KL-DD-NP-SI-JV: 122 MP, 107.2 ORTG, 112.0 DRTG, -4.8 RTG
KL-DD-OG-SI-JV: 91 MP, 113.8 ORTG, 112.7 DRTG, +1.1 RTG

Rather than dealing with two lineups separately, let’s look at the 4-man unit of Lowry, DeRozan, Ibaka and Valanciunas, as that’s really what we are questioning here. The small forward position seems likely to take care of itself, at least in the short term.

Lineup | Minutes Played | Offensive Rating | Defensive Rating | Net Rating
KL-DD-SI-JV: 228 MP, 111.6 ORTG, 111.4 DRTG, +0.2 RTG

So, that’s not great. Certainly, you want better than .500 ball from your starting unit. But let’s also put it in the context of the issues from the past, again using 4-man units, ignoring the revolving door at starting SF.

Lineup | Minutes Played | Offensive Rating | Defensive Rating | Net Rating
2016-17 Siakam Starters: 458 MP, 106.2 ORTG, 113.3 DRTG, -7.0 RTG
2015-16 Scola Starters: 779 MP, 103.2 ORTG, 108.1 DRTG, -4.9 RTG

Even if you ignore the tough schedule the Raptors have faced so far, the starters this year are really not a problem like the past ones have been. They’ve had a couple serious duds of late, but the season as a whole tells a story of a mediocre lineup, not a terrible one.

Now, again, that’s not what you want. The Raptors have about 80% of their team salary in that starting unit, so obviously they want better than break-even basketball from that group.

The starters have struggled about as much defensively as they have over the past couple of seasons. The difference this year is they are able to outscore a lot of those issues, which may not be considered acceptable as a strategy.

But What About Those Duds?

And yes, of late we’ve seen a couple of examples of just how disastrous those slow starts to each half can be, with fairly narrow losses to New York and Indiana ultimately coming down to huge runs given up by the starters.

And it really does seem to be an issue for the starters specifically at the start of each half. Consider the following breakdown for the starting 4-some we talked about above.

Quarter | Minutes Played | Net Rating
1st Q: 98 MP, -8.5 RTG
2nd Q: 44 MP, +29.3 RTG
3rd Q: 68 MP, -13.0 RTG
4th Q: 17 MP, +19.9 RTG

What the heck, right? These guys suck to start each half, and are awesome to close each half. They’ve played a lot more in 1st and 3rd Q’s, as you’d expect, so their awesome play later in halves is just managing to balance out their overall numbers. But why? Why would they play so poorly to start if they are capable of playing so well, and do so, later in each half?

The picture starts to look like maybe this is not a lineup construction issue as is widely assumed. For all the obvious drawbacks that Valanciunas brings defensively, and DeRozan as well, and the sporadic effort of Serge Ibaka when playing power forward, and Lowry’s struggles earlier in the year: those drawbacks are all there in the 2nd and 4th quarters. So what is up with the 1st and 3rd?

And then there’s the recent attempt to fix that by rolling out a different lineup to start the third quarter against the Pacers, after a lacklustre first half effort by the main group. Coach Dwane Casey sent out Siakam in Valanciunas’ place to give the starters some defensive backbone and energy in transition. Of course, it didn’t go well, with that unit giving up a 15-5 run in five minutes, good for a -83 net rating. That’s even worse than the not-so-nice -69 net rating the starters managed in the 1st half.

So there is something rotten in the starting units, and the answer does not seem to be as simple as, say, removing Valanciunas, the obvious choice if any lineup change were to be made.

Finding Alternatives

But let’s at least look at what options the Raptors have. If we accept that Lowry and DeRozan are fixtures in the lineup (as we should), and accept that the starting SF will be largely irrelevant to the success of the group (and that OG has earned some time there for the foreseeable future), and accept that Ibaka is certainly going to start, with the contract he was just given and his status as a borderline star in the league — that is, we accept that the only move to potentially make is to remove Valanciunas from the starting lineup — then what are the possible replacements for him?

First, the obvious lineup change, the one Casey tried against the Pacers to no avail: Bring Siakam in to play PF, move Ibaka to C, and run a small, switchy defensive look. Makes sense — stylistically that lineup couldn’t be much more different than the current one.

What if Siakam Starts?

How has that lineup done this year, overall? We’ll again isolate from the SF position and use the 4-man lineup of Lowry, DeRozan, Ibaka and Siakam to maximize the sample size. Also, somehow those four have played with all three of Valanciunas, Poeltl and Nogueira for short spurts, so we’ll remove those minutes, since we’re not really talking about 3-big lineups.

Lineup | Minutes Played | Offensive Rating | Defensive Rating | Net Rating
KL-DD-PS-SI: 73 MP, 106.0 ORTG, 102.3 DRTG, +3.7 RTG

Well then, the defense looks much better. The offense dips considerably, but not enough to wash out the defensive improvement, and they post a slight positive rating. Not exactly the killer lineup you’d hope for, but at least a small improvement, and if it could hold up over a full season, it would be a significant positive contribution to carry a small positive margin over heavy minutes.

But there’s more context here — remember when we looked at the starting lineup by quarter, and how they somehow started slowly but played well later in halves? Let’s run the same test with this group, although we know the samples will be small.

Quarter | Minutes Played | Net Rating
1st Q: 17 MP, -20.9 RTG
2nd Q: 13 MP, +5.9 RTG
3rd Q: 23 MP, +11.7 RTG
4th Q: 20 MP, +12.7 RTG

Well, the third quarter is better at least, though the Pacers game showed that’s no guarantee. (The lineup had a +28 net rating in the third prior to the Pacers game, but that didn’t stop the collapse.) But sadly the first quarter numbers are actually even worse than the starters have posted. Of course, the above is all small sample theatre, and overall as we showed they’ve washed out to a slight positive, but the trend continues with this lineup — they are struggling to get started. When you consider that the vast majority of those 1st and 3rd quarter minutes would have come after about six minutes, that muddies the waters as well.

In any case, we’ve got a candidate for a maybe-somewhat-better lineup than the starters. There should be concerns about how Ibaka would hold up at C over the long haul. Also, what happens to the rotation at C and PF if they pursue this? One of Valanciunas or Poeltl likely gets buried, and Nogueira can kiss his somewhat sporadic minutes goodbye. Do you completely bury Valanciunas, probably further damaging any limited trade value he might have? Do you do that to Poeltl, restricting his development? And who plays at PF? Do Siakam and OG play a combined 48 minutes there?

It’s complicated with the current roster construction. So even if we won’t write off this possibility entirely, let’s explore one other alternative.

What if Poeltl Starts?

This one makes sense long term. Poeltl seems to be the presumptive heir to Valanciunas’ minutes, and is another top-10 centre pick who has impressed in his first two seasons. He took the backup job from Nogueira last season, and now is performing very well in that role this season, even though the Raptors have been running a deep rotation, especially at centre.

It also makes sense to address the defensive issues for the starting lineup. Although this leaves Ibaka at PF where he has struggled a bit with quicker players, Poeltl is a tremendous help defender, and is at his best when defending the pick and roll in space, a notable weakness of Valanciunas’. It’s also something that can get exploited in the starting lineup by the best guards on the opposing team, especially in conjunction with DeRozan’s general inability to navigate screens, and Lowry’s inconsistent effort in that area.

The good news is, Poeltl has actually gotten a nice chunk of minutes beside the other starters, often when Valanciunas has gotten an early hook or in his place in the 4th when the team is closing and needs defense on the floor. The bad news... well, I’ll show you. Here are Poeltl’s on-court ratings when playing with, a) all three of Lowry, DeRozan and Serge, and, b) both Lowry and DeRozan, with or without Serge, to give a larger sample than the first one. Again I’ve removed the weird 3-big lineup minutes.

Lineup | Minutes Played | Offensive Rating | Defensive Rating | Net Rating
KL-DD-SI-JP: 27 MP, 89.8 ORTG, 115.9 DRTG, -26.1 RTG
KL-DD-JP: 73 MP, 102.8 ORTG, 118.1 DRTG, -15.3 RTG

So... yeah, that’s not great. And 73 minutes is a pretty sizeable sample there. Personally I think that although Poeltl has shown great promise and has really helped that bench unit, he’s just not physically ready for the starting bigs this league can throw at you. For whatever reason, this idea looks less than promising. The defense has somehow been worse than the starting lineup’s has been, and the offense falls off a cliff.

And at the same time, even if you think the results would come around, Poeltl hasn’t exactly shown an ability to stay on the court in those matchups. He racks up seven personal fouls per 36 minutes when playing with both Lowry and DeRozan. That’s a very high rate — the highest rate on the team right now is Valanciunas with 5.0 PF/36, Poeltl right behind him with an overall rate of 4.9 PF/36.

For the record, Nogueira has the same issue. When he’s played with all three of Lowry, DeRozan and Ibaka (he has a decent sample with all three at 39 minutes) he’s fouling at a rate of 9.1 personal fouls per 36 minutes. That’s insane.

What Is The Solution?

All this leaves us with is more questions, and fewer answers. We’ve got a starting lineup that sucks starting but closes well. We’ve got a small ball look that may fix some of the issues, but they have started just as slow, crapped the bed in their most recent chance to fight to be the starting lineup, and come with concerns about the rotation implications. And we’ve got the up and coming Poeltl making a straight positional substitution, which has been a disaster so far this year.

My personal theory? The Raptors lead guards think they have to run the offense through Valanciunas and Ibaka when they are out there, especially to start halves. In the first quarter, Ibaka and Valanciunas carry usage rates of 24% and 23% respectively. In the third quarter, 19% and 25%. But remember from above, those starting lineups have excelled in the second and fourth quarters. In the second quarter, the two bigs have carried usage rates of 15% and 17%, and in the 4th: 16% and 17%.

Although the offense clicks well enough for that group, it clicks better later (it’s roughly league average in the first and third quarters), and asking more of the bigs just takes energy they could be using defensively. (The other big concern for me, the inconsistent energy, from all five guys, defensively). I think a more aggressive approach from the guards early in games could lead more naturally to the appropriate usage for the bigs, where their opportunities feed off of a defense broken trying to contain the guards, rather than force-feeding them looks. The lack of attacking mentality also makes it seem like the team is going through the motions early, which may translate to that missing defensive energy.

My theory aside though, I’ve tried to explain the complexity involved in this topic compared to the problems of the past couple seasons, where there seemed to be an obvious solution to relatively glaring issues. Please feel free to ask any questions about the stats presented above, and please comment about what you think should be done with the Raptors’ starting lineup, whether it is a rotation change or not.

All numbers per