Starting Marcus Morris and Marcus Smart: the change up revisited

In a bumpy season in which disappointment and frustration have abounded, the number of genuinely encouraging moments can likely be counted on one hand. However, there stands one universally recognized point in the Celtics underwhelming 2018-2019 season that stands as an unimpeachably a bright spot. After a disappointing road loss to the mediocre Dallas Mavericks, Brad Stevens turned to something new on November 26th in New Orleans. The Celtics relegated Jaylen Brown and Gordon Hayward to the bench and inserted Marcus Smart and Marcus Morris as full time starters. The Celtics then proceeded to win eight games in a row.

The Celtics are now 27-22 since that eight game winning streak christened the start of the Marcus era in Boston. Fans with a cynical eye will point out that the eight game winning streak came against teams that are all eliminated from playoff contention with two weeks remaining in the regular season. Fans with an even more doubtful view of things might now, with the benefit of hindsight, argue that the move was akin to shuffling deck chairs on the Titanic.

The answer seems to lie more with former than the latter, as the Celtics did seem to benefit from a dead-cat bounce in their soft schedule as the new lineup took hold. That said, it’s also a lineup that’s been better than the Celtics team average and still regularly outscores opponents. The “Marcus starting lineup” is still a +6 in net rating, well above the teams overall rating of +4.3. Compare that to the Opening Night starting lineup, which was a net negative in it’s 303 possessions together.

That said, “good” should not be assumed to be equivalent with “unimprovable,” particularly when the Celtics continue to lose games. Lineups have a trickle-down effect, and just because one lineup does well doesn’t mean it’s the optimal choice. It’s well and fine to outscore opponents, but it does no good if the next part of your lineup is such a tire fire that they can’t hold on to any sizable lead.

My esteemed colleague Daniel tackled this very subject at the beginning of March, but questions about the lineup changes seem destined to both come in like a lion and go out like a lion. Celtics fans have been grumbling about the lack of diversity to the rotation, but got their first real preview of new starting lineup in Friday night’s game against the Pacers. Stevens finally started Baynes and benched Morris, giving the Celtics a starting lineup that has outscored opponents 80-49 in limited minutes per the NBA Wowy tool. Rest for Irving and Horford prevented Celtics fans from getting a glance at that lineup again, but there’s a chance the Celtics revisit that lineup in Monday’s game against the Heat.

This is something that has intruiged me, because I’m someone who credits the Celtics improved lineup performance more to Smart than to Morris. A quick glance at the on/off numbers via NBA Wowy’s data bears out an interesting pattern.

The Celtics overall team pace adjusted average is 4.42, so as you can see here, Morris and Smart being on the floor are largely in line with Celtics averages. Indeed, Smart is Morris’ most frequent teammate on the floor, with the two playing 2708 possessions together this year. However, Morris suffers MUCH more than Smart does when the two are separated, and there’s an overall improvement when neither are on the floor.

Is that right? Are Smart and Morris really hurting the team with their play?

Diversify the Celtics individual on/off numbers and the presence or absence of Smart and Morris and a more complete picture comes into focus.

All of the Celtics rotation players suffer a significant negative effect when playing with Morris in the absence of Smart. The inverse is not true at all, as the Smart only lineups have done very well, and outscore the lineups without either about half the time.

Indeed, it seems like playing so many minutes with Morris has hurt Smart’s on/off numbers this year, as he has fallen from his usual +/- perch near the middle of the team, to near the bottom of the squad. Even with his own improved play and shooting this year, Smart only leads Morris, Jaylen Brown, and Terry Rozier in rotation player +/-.

Gordon Hayward and Brown continue to make strides on the bench and both players seem to be playing themselves into a rhythm. Jayson Tatum has been hit or miss this season, but all three have shown the ability to take over a game this season and all three are in direct competition for positional minutes with Morris. It’s clear now, with the data (and frankly eye test) that reducing Morris’s role and expanding that of Baynes and the other wings is the way for the Celtics to proceed from here on out.

It’s unlikely we see this, as Morris is a respected veteran and Stevens has been extremely conservative with lineup changes all year. However, it’s clear that the Celtics need a clearer identity as the playoffs descend on us, and a shorter rotation will help with winning. I think Morris would be fine with it. After all, he said so himself.