After dropping Game 2 in convincing fashion, the Celtics showed up with plenty adjustments to take back the series lead with a victory in Game 3. The Celtics beat the Warriors 116-100, pouring in a fantastic offensive performance after a stinker in Game 2.
Using win probability data from our friends at Inpredictable, we can see the Celtics were in control for a majority of the game:
After Game 2, my feeling was that the Warriors had the advantage at the moment. I thought there was some low hanging fruit for the Celtics to pick offensively, but their troubles with Curry were going to be harder to correct. While the ladder became true, the Celtics showed up offensively with an impressive combination of adjustments and simply playing better.
Below is a breakdown of the offensive and defensive ratings as well as the four factors on both sides of the floor for the Celtics per CleaningTheGlass. The biggest shift in Game 3 was the Celtics offense, where they skyrocketed from 2nd percentile in Game 2 to 92nd percentile in Game 3:
Below we're going to discuss how the Celtics had so much more success, where the Warriors need to adjust, and adjustments the Celtics can make as well.
The pressure point has been found
A lot of what Boston has done in the previous two series has been finding a pressure point offensively and continuously exploiting it. Against the Bucks, it was involving Grayson Allen and Bobby Portis in every action. Against the Heat it was early offense and quicker kickouts with better spacing. In Game 3 against the Warriors, the Celtics figured out they can relentlessly attack anyone besides Andrew Wiggins downhill quickly and get positive results.
Going from the Bucks to the Heat to the Warriors are wildly different challenges offensively. It was expected that the Celtics would need time to adjust. The Bucks defense operates in shutting down the paint through extremely good rim protectors combined with intentional funneling and collapsing with length. It took the Celtics to really dip into 5-out to have more success against them. The Heat defense protects the rim in an aggressive way a little differently. They pull help over from one pass away and sit right in the gap at all times, knowing they have the ground coverage to make the closeout. Again, the Celtics needed to be more intentional with their spacing and attacks but also hunted the right matchups and pushed the pace in transition. The Warriors operate differently, funneling drives to some extent but also using their connectivity and quickness to prevent high quality rim attempts and close out to shooters. The Warriors don't have the defensive talent or size, but they do have some elite pieces with overall versatility and IQ.
The key for Boston was the drives. The Celtics came out in attack mode, looking to get two feet in the paint and make plays far more often than they had in the first two games:
Now, this didn't just happen because Boston felt like it. In theory, teams would drive every single possession if they had the chance. Getting paint touches is one of the guiding principles of most NBA offenses, and drives are obviously a major part of that. Boston wasn't just driving to the rim more because they felt like it, but there were some key elements that unlocked their ability to drive more often.
The first key point is attacking the pressure points. Boston was intentionally hunting the Warriors guards in ball screens and off-ball screens because they knew they couldn't stay in front of them. Jaylen Brown and Jayson Tatum were the main proprietors, but Smart was also excellent at getting downhill consistently.
It's not just about who they're attacking, but it's about where they're attacking. The Celtics did a much better job of nailing the subtleties of help manipulation to create easier finishes at the rim and more extended help. In the clip above, look at where everyone is located. The best rim protector, Kevon Looney, is planted one pass away and far away from the rim. There are shooters in the deep corners with Marcus Smart in the slot. This was not on accident.
In the next clip, look at where Draymond is placed. The best rim protector on the floor ends up nowhere near the rim on the drive because Horford is planted in the slot. Again, this was not on accident.
The changes are small but they make a big difference. Here, Horford is on the wing but takes a single step towards the top of the key to create a slightly longer closeout on the Jaylen Brown drive. A quick decision from Brown leads to a good shot. Also, look at where Robert Williams is if the drive gets deeper. He's lower and further away from the rim, just to make that rotation slightly more challenging if needed.
The Celtics intentionally blitzed the paint while making the responsibilities of Draymond Green much more difficult. Not only did they have success finishing against him (5/9 against him within 5 feet), but they had a ton of success finishing against the entire Warriors team:
Is all of this sustainable? Probably not, but it was an impressive showing that exposed some real holes in the Warriors defense.
The Celtics stars were outstanding
Jaylen Brown came out in the first quarter and had arguably the best 12 minutes I have ever seen him play. His growth in this series as a downhill attacker and playmaker have been exciting to witness. In Game 2, the biggest adjustment made by the Warriors was putting Draymond Green onto Jaylen Brown. Green cut off a lot of driving angles and prevented Brown's on-ball creation to a large extent. That all changed in Game 3.
Jaylen Brown came out and attacked Draymond Green. Green couldn't fully stay in front of Brown with a full head of steam in a way that was shocking to witness. I mean, watch Brown call off a screen designed to get Curry matched up against him just to go at the one of the best defensive players ever:
His process was different tonight. He's the Celtics version of a "scheme breaker". While Tatum is obviously the best player, Jaylen Brown has been the most consistent isolation matchup killer in this series. He has a wild combination of burst and bend for someone of his size that makes it incredibly difficult to stay in front of him. To reiterate this, here's another clip of him driving on Draymond:
Jaylen Brown breaking the code on the Draymond Green matchup early was a series shifting event, but his playmaking growth could be a career shifting event. Brown was excellent as a passer and decision-maker in Game 3. He leveraged his downhill gravity beautifully with well timed passes on target out to the perimeter.
The Warriors bring help later and less forcefully than opponents in the Celtics last two series. They are also generally smaller and less athletic. Once Brown was able to figure that out, it was tough for the Warriors defense to stop him.
Jayson Tatum was also excellent in the game. In my Game 2 article, I criticized his stop and hold mentality for having negative effects on the Celtics offensive efforts and thwarting some of their advantage creation and drive and kick prowess. The script was completely flipped in this one, with Tatum having a ton of success getting downhill consistently against smaller guards. While he doesn't have the burst or shift of Jaylen Brown, he leveraged his elite handle and strength combination to get in the paint quickly and consistently.
Tatum has also just been ridiculous as a passer in this series. While he's still finding the optimal blend between tough shot making and distribution, Tatum has continuously made plays for others in this series. The past month has been somewhat of a skip pass and cross court kickout clinic from him.
I think the extent to which Tatum has impacted the series so far may be going a bit under the radar. While the individual scoring and efficiency haven't quite been there on a consistent basis, he's still easily the Celtics best offensive player in my opinion because of the masterful passing. From Todd Whitehead of Synergy, Tatum has created a wild 45% of Celtics points while on the floor at a staggering PPP of slightly over 1.2:
The combination of leveraging the star wings while also optimizing what's going on around them has given the Celtics the lead in the series and could make them favorites the rest of the way.
Stephen Curry: the NBA's 3 cup problem
The 3 cup problem is an impossible mathematical puzzle. It starts with one cup facing up (the one in the middle) with the other two facing down. The player then has 6 moves to get all cups facing up, but must flip exactly 2 cups every move. This fun little puzzle is exactly what defending Steph Curry is like. Flip one cup up and another one is flipped down. During the course of a game, it's literally impossible to shut everything off (get all the cups flipped).
There are a variety of ways to defend a screen, but the most common NBA coverages are as follows: drop, switch, flat coverage, hard show/hedge. The Celtics have elected to play a drop, which has received wide criticism. In this coverage, the center drops down below the level of the screen, limiting the screen to a 2 on 2 and keeping the action from needing extra help from other defenders. While this may not look great, it could be the best coverage the Celtics have. Curry has gotten the Celtics out of switching by effectively attacking Al Horford and Grant Williams when out on the floor and consistently collapsing the Celtics defense. A flat hedge or hard show essentially brings two to the level of the ball with different levels of aggression, which is exactly what the Warriors want.
The Warriors are a movement oriented offensive machine that's nearly impossible to stop once they get going. It's their movement and randomness that demolished a great Mavericks defense in the previous round. Their movement is created screening and cuts, but a crucial element to unlock advantages is the opponent bringing two to the ball on screening actions. By dropping the Celtics have generally kept themselves out of the Warriors offensive blender. Watch the possession below, which the Celtics finally aggressively bring two to the ball with Draymond out of the game but the Warriors get exactly what they want:
While the drop may be empowering Curry and Klay, everyone else has been contained. Not only that, but what makes the Warriors so incredible offensively has been somewhat devolved to Curry and Klay making shots off the dribble. Those possessions may result in plenty of plays that look bad on the surface, but overall taking everyone else out of the game and flow (especially Draymond) has produced decent results.
While I don't think the drop is necessarily as bad as others think it is, it does need to be adjusted. This meme from Game 1 needs to be adjusted a little, but is still quite effective in the story telling process:
While I understand the theory behind the drop coverage, this is simply too deep. Giving up a 3 this open to the best shooter ever is not great process but it's also not too far away.
While the depth of the drop certainly deserves criticism, Curry makes legit cheat code level plays that you have to live with sometimes. Do you really expect Horford to be up to touch on a screen set 35 feet from the rim?
Don't be surprised if the Celtics come out in Game 4 playing drop yet again, just with a higher iteration. Watch here at the end of the 3rd quarter, as Marcus Smart and Ime Udoka are telling Robert Williams to get up higher on the ball screen.
While some possessions may look bad, think about the bigger picture. The Celtics are essentially daring the Warriors to run a high ball screen every play with Curry shooting deep off the dribble 3s because they know it takes them out of their offense. Is this the most optimal solution? Maybe not, but don't let the aesthetically displeasing results of a few plays cloud your view of the entire tactic.
Lineup tinkering and searching for optimization
While the Celtics took a step forward in optimizing their Finals lineup, it felt like the Warriors took a step backward. The Celtics have found something with playing more with only one big on the floor. In Game 1, Robert Williams and Al Horford played 31 possessions together. In Game 2, that number was 24 possessions (but remember there were almost 11 minutes of garbage time). In Game 3, the number drastically reduced to 19. Ime Udoka took Robert Williams out early at the beginning of each half for three reasons. First, it allowed Williams to eventually sub back in for Horford and increase the spacing and allow the Celtics to play smaller. Second, it limited the time he was on the floor with Steph Curry, who was hunting him in ball screens. Third, it effectively eliminated Daniel Theis from the rotation, who just isn't good enough to be playing at this stage.
The counteract the Celtics playing smaller, the Warriors decided to also play smaller. Kevon Looney had his minutes shrunk to just 16:49 in Game 3. With the Warriors losing for much of the game, Steve Kerr elected for more offensively oriented lineups with Gary Payton II only playing 11:25. These two decisions put more on the plates of Otto Porter and Jordan Poole, both of whom were questionable defensively in Game 3.
One problem I've had with the Warriors all season is the constant push/pull of many of their lineup decisions. Kerr is often left with the assignment of sacrificing something. More defense gives them less shooting and creation. More rim protection and rebounding creates a more clogged paint. More shot creation leads to weak point of attack defense. While this is generally an issue with every team, the Warriors take it to the extreme for a title contender with how poor Poole is defensively, how limited Gary Payton II is as a shooter, and the athletic limitations of Kevon Looney.
I understand how difficult the decisions are, but I was confused by the decision from Kerr to go away from Looney and Payton so early. Each player is crucial to the Warriors defensive effort of plugging holes that the Celtics have figured out how to create. They are also great screeners and ball movers offensively. Importantly for Looney, he is able to punish the Celtics smaller lineups with rebounds on both ends of the floor while providing more rim protection. I'd be surprised if we don't see more of each in Game 3.
Quick shoutout: Robert Williams
Despite clearly laboring still from his torn meniscus (reminder: his surgery was only two months ago), Robert Williams was outstanding defensively in Game 3. The Warriors are simplify terrified of him around the rim. Having Williams effectively roam on the backline while figuring out what the Warriors are doing and being earlier on rotations had a major impact in Game 3. He was everywhere.
Series outlook
First of all, something I didn't discuss is the Curry foot injury. If he can't play or isn't quite right, the series is over. Curry is of unbelievable importance to the hopes of this Warriors team. However, with him healthy the Warriors are still right there. The Celtics still haven't figured out the Curry issue, and I'm worried about an overcorrection from their side. The struggles from the Warriors came on the defensive side of the ball, but simply playing their defensive personnel more should help. They can also mix in different coverages, stunt on ball screens like they did against Luka, work harder to not give switches on screens, and bring help earlier and higher on drives.
Each side still has plenty of adjustments to figure out, which makes this series very exciting the rest of the way. I see this as a fairly even fight the rest of the way, but considering Boston has a 1 game lead they should clearly be considered the favorites as of right now. However, the evaluation process here is still extremely fluid. This is far from over.
Comments