The Boston Celtics are 6-4, with commanding wins over multiple playoff teams. And the data suggests they could be even better by playing more small ball. Assessed below is each Celtics frontcourt pairing through 10 games.
The primary stats used are called "offensive rating" and "defensive rating," which is the number of points per 100 possessions a team scores or allows. "Net Rating" is the differential between "ORTG" and "DRTG." All stats used are from NBAWowy are relatively small sample sizes since it's early in the season.
The Celtics played "small" almost exclusively after the trade deadline last year. But with David Lee and Amir Johnson in the fold, Brad Stevens has used more traditional lineups with two bigs in the game.
But Tyler Zeller, who played in 82 games last season, has played only 62 minutes, due to the minutes crunch. Stevens said after Friday's 106-93 win over Atlanta that he expects Zeller to play "a big role" at some point, but stretching the floor is a primary objective.
"We got a lot of bigs. I don't know how else to say it," said a chuckling Stevens. "We haven't shot it great so you want to play some guys that can stretch the floor and be guarded when the floor is stretched. And that leaves at least one person out."
Despite the plethora of true bigs, Stevens has still went small with Jonas Jerebko at power forward, because they were lethal with him last year, with a +9.5 Net Rating.
The results with Jerebko have been even better early on this season.
Four of the five best scoring lineups all involve Jerebko; the two best involve Kelly Olynyk. Combine, Olynyk and Jerebko make up the best frontcourt pairing on the Celtics, dominating in their 33 minutes with a +72.3 Net Rating.
The Sullinger-Olynyk pairing was potent last season and early returns are promising. But, surprisingly, they've played only 15 minutes together.
Though they've faced bench units more often than starting units, the offensive success of the Jerebko lineups and Sullinger-Olynyk pair is likely a product of their shooting and spacing.
The new starting frontcourt of Sullinger-Johnson has a -2.7 Net Rating over the full year, but in the last seven games it's +2.1. And over the recent three-game winning streak, it's +11, with a 75.0 Defensive Rating.
Zeller has played 30 of his 62 minutes alongside David Lee, Boston's original starting frontcourt, and it's obvious why Stevens pulled the plug on that so quickly, with a -12.3 Net Rating. Even since they've been benched, Boston's worst Offensive Ratings come with them in the game, at 95.6 and 80.6, respectively. So it's not like their averages are weighed down from the first three games they were starting.
The three worst scoring lineups all involve Lee. The Celtics have 13 players who are capable of shooting threes; the only two that can't are Zeller and Lee. So it may not be a coincidence that their offense is at its worst with them, due their inability to stretch the floor.
As Stevens said before, Zeller will have a role on this team. But it might be unfair to expect him to have anything close to last year's role considering how much better they operate playing small. Zeller's role could be matchup-based, perhaps similarily to how he was used on the recent road trip as a stopper against Steven Adams and Dwight Howard.
The data also reveals some intriguing trends based on Jerebko's usage when he's playing as a stretch four alongside just one big or as a small forward, alongside two bigs.
The Celtics are a +40.7 with Jerebko at the four, with an incredible 124.1 Offensive Rating. But when Jerebko downshifts to small forward, their Offensive Rating nosedives to 77.7 and their dominant defense balloons to 100.5, a -22.8.
This could be happening because when the Celtics have small lineups, they typically have five players on the floor that can shoot threes, drive closeouts, and create for teammates.
Zeller and Lee can make plays off the bounce, but they can't space the floor like any of the other Celtics bigs can. Maybe the Celtics are just creating more quality shots when they're stretching the defense with shooting lineups.
It's also possible that having an extra wing in the lineup -- Marcus Smart, Avery Bradley, Jae Crowder, or Evan Turner -- instead of an extra big -- especially Lee -- means there's simply more talent on the floor.
The SportVU data isn't publicly available, but it'd be interesting to see how the Celtics perform creating "open shots" with Jerebko at the four, compared to other pairings.
This play from Sunday's dominant win over Houston is Stevens' read and react motion offense at its finest. Sullinger sets a screen and hovers around the three-point line, stretching Dwight Howard into an area he'd prefer to never cover. This stretches the Rockets defense, leading to a Sullinger drive and dish to Johnson for the open layup.
Jerebko's greatest strength on offense is his ability to stretch the defense and drain open jumpers. But when he's covered by small forwards, who aren't often tasked with protecting the paint like bigs are, they don't give him enough breathing room to either hit shots off the catch or drive closeouts, or use his speed in the screen game.
Here, Jerebko becomes a matchup nightmare against the Bucks.
Jerebko is capable of defending either of their bigs on the floor, but there's little chance they can match his speed either on the drive or the roll. The ball handler, Evan Turner, also had the option of flinging the ball to Kelly Olynyk, who was wide open since the off-ball defender had to help on Jerebko.
Perhaps even more important are the changes that take play defensively, since Jerebko can switch on most pick-and-rolls, especially with paired with Jae Crowder. Here's a clip from last season that showcases their versatility.
Originally it was assumed the Celtics might struggle on the defensive boards with Jerebko at power forward, but lineups with him on the floor lead the team in DREB%. Still, the Celtics could be prone to getting abused on the boards against lineups using two traditional bigs. It's something Stevens needs to balance based on the matchup.
And that's why Lee and Zeller still have value. Zeller can be used in matchups against bigs, and Lee can be a change-of-pace big. Stevens has done a good job of minimizing their roles, without giving them both the Gerald Wallace end-of-the-bench treatment.
The bottom-line is that, at least according to the small sample size of data, so far the Celtics are simply better, more versatile, and more lethal with lineups that stretch the floor. It's early in the season and Stevens is unlikely to make any major changes. But his candid statements to the media suggest he has a desire to play small more frequently, and the data supports that desire.
As the Boston Celtics hit their stride as one of the most feared teams in the NBA, it'll be interesting to watch the changes Brad Stevens makes to maximize their performance.