On the surface, things couldn’t be going much better for the Celtics a quarter of the way through the regular season. The team has overperformed early with an 18-4 record thanks to solid and timely contributions up and down the roster, which have helped the C’s overcome offensive lulls that have cropped up repeatedly over the first few weeks. Defensively, the Celtics had been an early juggernaut, but they’ve slowly fallen back to earth in that department (104 defensive rating in past six games).
Boston’s season defensive rating has risen to 98.3 points allowed per 100 possessions, which is still the best in the league, but it’s about four points higher than it was a couple weeks ago. Some regression shouldn’t be considered surprising -- the Celtics were defending at a level better than the NBA’s top defensive team last year for the first few weeks of the year. However, a closer look at the team's lineup numbers showcases a troubling sign: the C's starting five may be their worst defensive unit.
Out of the Celtics’ 17 most commonly used lineups (more than 15 minutes of court time), Kyrie Irving, Jaylen Brown, Jayson Tatum, Marcus Morris and Al Horford rank dead last in defensive rating. The starting lineup has played the most out of any of Boston’s top lineups this year (85 minutes) and is allowing 117.8 points per 100 possessions, nearly 20 points worse than the team’s season average of 98.3. That 117.8 mark would put them eight points below the NBA's current worst defense (Sacramento Kings at 109 points allowed per 100 possessions).
The team’s offensive rating with that unit has been subpar as well (102.8 points scored per 100 possessions), giving the group a -15 overall net rating, the only negative net rating among the team’s top 17 most-used lineups. The sample size is obviously a bit limited 20 games into the season, but those numbers are alarming nonetheless.
[caption id="attachment_376496" align="alignnone" width="766"] via NBA.com/stats[/caption]
Can those ugly results be attributed to bad luck, a bad game or two, or is there a bigger problem with that grouping that should result in Brad Stevens shaking up his starting five? And who should be the person to sit if a change is made? I dived into recent games and into some video to try to get a better sense of the group’s issues.
The first question I tried to answer is whether the starters’ bad overall numbers are a result of an outlier or a couple of bad stints. I looked at each of their results, game-by-game, and found that neither was the case. This specific grouping has performed far worse than the team defensively on the whole in six of their last seven starts. I compared their overall plus/minus and defensive rating to the team's in all of these games.
8 minutes vs. DET: -14 in 10 point loss. 149.2 defensive rating (124 for team)
13 minutes vs. ORL:+2 in 20 point win. 112 defensive rating (103.5 for team)
3 minutes vs. MIA: (Aron Baynes started over Marcus Morris so discounted this game)
13 minutes vs. DAL: Even in 8 point overtime win. 106.6 defensive rating (94 for team).
15 minutes vs. ATL: -1 in 11 point win. 134.4 defensive rating (104 for team)
9 minutes vs. GSW: -9 in 4 point win. 144.8 defensive rating (89 for team)
12 minutes vs. BKN: +5 in 7 point win 87.5 defensive rating (101 for team)
10 minutes vs. ORL: -10 in 16 point win 136.3 defensive rating (89 for team)
A lot of people have been pointing the finger at the C’s offense for creating early holes for the team to come back from, but the starting unit at the beginning of the first and third quarters have been far worse on defense than offense.
So where exactly have the issues come from? I watched the last few games to try to find some common trends.
Jaylen Brown is struggling with quicker guards: Avery Bradley, Reggie Jackson, and Kent Bazemore are just a few of the shifty guards that have given the Celtics trouble in recent weeks. All of these guys have had their way with Brown at various stretches. Bradley, in particular, worked over Brown during Detroit’s road win on Monday. The 6-foot-7 shooting guard likes to stay tight on opponents when matched up against them off the ball, and Stan Van Gundy took advantage of this by picking Brown off repeatedly with picks and handoffs.
Look at how much separation Bradley got on Brown on the first play of Monday’s a game:
Plays like these are why Andre Drummond scored 22 points with ease on Monday night. Bradley comes off an off-ball screen with Brown trailing him and the ball is given to Bradley on a handoff. Drummond’s defender (in this case Daniel Theis), has to help off the center, which allows Drummond a clean run towards the rim. Bradley either gets a good look and/or Drummond gets in great position against an undersized defender in the 2-on-1 situation. A floater, easy layup, or offensive putback were the usual end results. Rinse, recycle, repeat. Other teams have targeted Brown in the same way in recent weeks and he was put through similar play calls all night long by Detroit.
To be fair to Brown, the Pistons’ execution was tremendous and plenty of teams have trouble chasing around Bradley on screens as we saw in Boston for years. However, Brown’s recent defensive miscues don’t stop there. Take a look at his foot positioning here on the handoff to Bradley.
Bradley is dribbling at full speed towards his strong hand, yet Brown is caught leaning towards his right when Bradley takes the handoff. Bradley gets separation immediately and another easy layup. That's sloppy technique/positioning.
There was also another bad gamble on Monday night when Brown was pressuring Drummond in the backcourt, which allowed Jackson to get a head of steam going down the court. Brown doesn’t recover in time and Jackson finishes the layup. This is the type of mental miscue that we saw a lot of last year, but not much this season -- until recently.
All of this isn’t to say that Brown has been having a bad defensive year. He’s been far better than pretty much anyone anticipated and his overall jump from his rookie year has been tremendous on the whole. Still, teams are starting to prey on this weakness (staying too tight on defenders away from the ball, combined with Brown's inability to get through screens) and attack it more since the C’s have no one else to guard elite speedy scorers in the starting five.
Kyrie Irving is getting targeted: This was what we all expected to happen once the All-Star point guard came to Boston, but teams are going at Irving more on a regular basis. He’s been putting in better effort than most anticipated, but he’s still a weak link in this defense right now. The Mavericks scored their first 11 points of the game on Irving during last Monday’s game simply by placing him in a pick-and-roll and switch situations in the post.
The Pistons similarly targeted Irving when they could, whether it was an on-ball or off-ball situation.
This is part of the package with Irving, but the C’s fellow starters also need to help a bit more in this department. For instance, when Tatum gets picked in that second clip, he needs to fight through that screen harder to try to get back to his man and not leave Irving to face Harris in a post situation.
Rebounding issues: We foresaw this problem before the start of the season. Horford is an undersized center and Morris is a below average rebounder when he plays at power forward. As the team’s starting frontcourt, they have rebounded the ball at a horrendous rate, and this starting unit is grabbing just 70 percent of all available defensive rebounds, nearly 10 percentage points below the team’s average. Tatum and Brown have been rebounding fine for their size all year, but the lack of a true center (Aron Baynes or Daniel Theis) down low has hurt this lineup a lot. The tradeoff was supposed to be defensive versatility and an above-average offensive numbers, but that clearly has not been the case just yet.
Contesting/Pressure issues: Opponents are shooting 47 percent from 3-point range against this starting five. Some of that is luck, but some of it is lazily contesting shots on the perimeter (Marcus Morris is probably the biggest offender). With the C’s lacking good foot speed at the 3-5 spots on the floor, they often give opponents a bigger cushion than they should. Irving isn't exactly reliable at contesting shots from the perimeter either.
For instance, on this play, Morris played great defense on Harris in the post and deflected the ball to the perimeter. He then essentially conceded an open look at a 3-pointer to an above-average shooter.
Morris isn’t back to full strength physically (at least he doesn’t look like it from a speed perspective) so that’s part of the package for him right now on defense. It has hurt the starting unit on the whole.
The Celtics lack of perimeter defenders (Brown is the only plus defender right now in that department) has also led to a lack of ball pressure and turnovers on opponents. The Celtics force an average of 14.8 turnovers per game, but their starting unit’s defensive turnover rate has only produced a 10.2 turnover rate per game (when stretched out to 48 minutes) for opponents.
Put it all together and you have a group that can’t rebound well, isn’t forcing turnovers and is allowing a lot of good looks from inside and the perimeter. That’s the recipe for a leaky defense, as has been the case over the last couple weeks for the Celtics starters.
Start Aron Baynes more: The C’s offense takes a hit when Baynes is on the floor in place of Morris, but Boston’s defensive rating is absurdly low (87.5) in 81 minutes when the big man shares the floor with Horford, Tatum, Brown and Irving. Baynes certainly provides better rebounding and stronger rim protection than Morris. When playing teams with a true center (like the Pistons on Monday night) Baynes is probably the right play. Not sure why Stevens only went with him for 13 minutes on Monday night.
Start Marcus Smart: The Celtics are putting Brown in a lot of tough spots defensively and he’s been getting in trouble due to them. Starting Smart at the two (and pushing Brown to small forward and Tatum to power forward) helps take care of quicker wings giving the C’s problems. They would still be vulnerable on the rebounding front but would be very versatile on the whole. However, this would leave a bench loaded with bigs (Morris, Baynes, Theis) and I'm not sure Stevens wants to manage that rotation. He also loves bringing Smart as energy off the bench, so it's hard to see this happening.
Wildcard option: I don’t hate the idea of starting a guy like Theis instead of Morris, who has played very well, on the whole, all season long. He’s probably the most mobile C’s big, is a strong rebounder who crashes the glass well and has shown good awareness on the whole. I feel like C’s could get away with him in this group with the other offensive options they have.
Ride out the problems: If and when Morris gets fully healthy, this original starting five is one of the best groups (on paper) that Stevens has all year. The Celtics are still 6-1 in games they have started so why not work through the kinks now and try to make it work, with a few adjustments in scheming? I would not blame the head coach one bit if he goes with this option. Some more effort from the veterans and some smarter technique from Brown would go a long way.
All things considered, it’s still early in the season but this is a situation to keep an eye on. My gut says we'll be seeing more of Baynes in the near future with the starting group to help counteract the problem. Stevens told the Boston Sports Journal before the year even started that he was going to be flexible with his starting five combinations all year. Unless this group shows improvement quickly, this is one unit he may be better off staying away from altogether in the wrong matchups.