Where Did the Portland Trail Blazers' Defense Go and Will it Come Back for the Playoffs?

There was a time when Nicolas Batum was widely considered the Blazers' best perimeter defender. His lanky limbs would smother three positions and occasionally a fourth. Anytime an opposing player started to get a rhythm, the Blazers would send their own version of the Stifle Tower to put out the fire. He still plays this role on occasion but at some point he lost his crown to Wesley Matthews.

Now, it's still a healthy debate. Both players are so different on the defensive end that you could never definitively prove who's the best. However, if Matthews were ever trying to bolster his case, this past month has been pretty convincing. On March 5th, when Matthews went down, the Blazers had a defensive rating of 99.3 (according to stats.NBA.com) good for third best in the league. Since that fateful day, the Blazers have a defensive rating of 106.9, ranked 24th over that time frame.

Wowzers.

Obviously, the Blazers were going to get worse on both ends after losing Matthews, but nobody could have expected the Blazers to drop that dramatically. This is especially true since his replacements aren't complete defensive liabilities. In fact, Arron Afflalo and Dorell Wright are both considered at least average defenders and C.J. McCollum has really improved on that end.

Watching the games, the most notable difference has been the ability to track shooters. Matthews was an expert at navigating off-ball screens, staying lock step with his man. There's a reason he was the guy to break Korver's streak of consecutive games with a three pointer last season. While the effort is certainly there, Afflalo and company often struggle to stay connected to shooters.

Notice how Afflalo is a half step behind as Klay Thompson gets the dribble handoff from Andrew Bogut. That has a cascade of effects that leads to an easy bucket. Bogut sees that Klay already has an advantage against Arron and abandons his screen to roll hard to the rim. He doesn't even wait to make contact with Afflalo before dashing to the front of the iron. As a result, he gets behind Robin Lopez.

The way the Blazers' defense works is to drop the big men back on the pick and roll and on dribble hand offs like this. The idea is that the big man can stay between the ball handler and the basket and between the roll man and the basket. He essentially takes away the layup from both players and the Blazers' guard is tasked with contesting the pull up jump shot.

Obviously, as you can see from the clip, this scheme breaks down as soon as the roll man gets behind the rim protector. When Thompson makes the pass, Lopez can't slide over and cut off Bogut because he's already been beat.

Here Afflalo does a better job of staying connected to Klay. As a result, Bogut actually has to set and hold the screen. That half second delay means he's farther behind the play and Lopez can stay in front of both Klay and Bogut. If Klay slows down to give Bogut more time, then Afflalo can recover allowing Lopez to return to his original man. At this level, the game really is about half steps and half seconds. With Wesley gone, the Blazers give away a few more feet a little bit more often.

This concern is certainly felt by all the players. First off, the guy who's guarding the shooter knows he's struggling to stay with him. In fact, they probably know it better than anybody else. This can cause guys to overplay and give up even easier shots.

Both Lillard and Afflalo are so concerned with Klay and Curry using screens to get open around the perimeter that they give up layups.

Second, the weakside Blazers have been going into crisis mode more often, leaving shooters more aggressively than they have in the past. This is one of the areas where trust is a big key to any great defense. Players need to know that if they stay within the scheme that everyone else is going to do their jobs. If that trust gets strained, then players begin to try and do too much leading to other problems. Since Wesley went down, the Blazers have been contesting 16% of opponents' three point attempts, down from 21% before the injury (data pulled from NBAsavant.com). That's due to both the individual defenders not staying with their men but also the need, or perceived need, for additional weakside help. When one part of the defense gets weakened, the whole system strains and bends.

Wesley's absence also affected the teams ability to control space. Dropping big men back and funneling the ball into them only works if you can actually funnel the ball in certain directions. Perimeter defenders are tasked with forcing guys certain directions all the time but it's easiest to see when defending a pick and roll.

As Stephen Curry catches the ball, LaMarcus Aldridge sets up on the right side of the screen. You can see Steve Blake shading Curry that direction but he can't get the job done. Curry is too good and uses a simple fake and a quick first step to dribble over the screen. Since Aldridge was positioned on the opposite side of the screen he has no hope of contesting Curry's jumper in time.

The Blazers have one of the most conservative defensive schemes in the league. It's successful only when it fits together. Since they don't force many turnovers or have the players to make athletic, disruptive plays, they prevent points by staying solid and forcing the toughest types of shots. All of this requires great communication and the Blazers have screwed up more often without one of their longest tenured Blazers on the floor.

Batum absolutely dies on a Bogut pick and then tries to call for a switch. Lillard doesn't hear it and is preoccupied trying to stay with the always dangerous Thompson. Of all the shots in the world, a wide open Curry might be the worst one to allow.

To be fair, the Warriors are really, really good. They do this to everyone, but Portland has been making these kind of mistakes against less intimidating teams like Utah or Brooklyn as well. But is all that really enough to drop Portland from the third best defense all the way to the 24th?

After digging a little deeper, the answer appears to be no. I already mentioned that the team is contesting fewer threes. This is something you'd expect from a weaker perimeter defense that is being forced to overcompensate more often. However, if this were a significant problem you would also expect opponents to attempt a lot more threes. That hasn't happened. Opponents are taking about the same percentage of shots from behind the arc and at the rim. All in all, the Blazers' strategy to force midrange shots hasn't fallen apart.

Other fundamental explanations for the drop off also fall short. The schedule hasn't been appreciably tougher since Matthews' injury. The average offensive rating of teams post-Wesley was 103.1 compared to 103.5 before. Portland has actually rebounded at a higher rate and opponents aren't getting any more shots in the first eight seconds of the shot clock when teams tend to score at higher rates. Essentially, teams are getting pretty much the same shots, they're just making more of them.

I hate to say I called it, but........I kinda called it.

Back in January, I wrote an article where I failed to find a reasonable explanation for why Blazers opponents were shooting so poorly from three point range. Without an explanation, I argued it was likely that opponents' shooting percentage would increase, undermining the Blazers' defense.

And boy has it ever. Since Matthews' injury, Portland opponents are shooting 39% from three. That's 28th in the league over that time frame and way up from the paltry 32% opponents were shooting up until that point. This could be explained partly by the lower contest rate but opponents are actually shooting an unconscionable 43% on contested threes. That's dead last in the league and, frankly, very unlucky.

There are a couple things to take away from this. First, even when healthy, Portland probably isn't a top three (or maybe even top five) defensive team. When the team possessed one of those elite rankings they were propped up by poor opponent shooting that proved unsustainable. Right now, after all the injuries and the regression back to the mean, Portland has the 10th best defense in the league. That seems like a pretty reasonable floor considering all the things that have happened this year. Moving forward, top five to top eight seems like a fair description of this team's potential on that end. That's certainly good enough to be a contender but it won't make the Blazers the kind of force they appeared to be in the beginning of the year.

Second, the team isn't as dependent on Wesley Matthews as it may appear. Of course the team is worse, but it is mostly a coincidence that his injury coincided with opponents shooting better from three. Given the contested shooting percentages, there would have to be something unique about the way Matthews contests shots that he chose not to share with his teammates. Even then, it would be a stretch to attribute such a large change to one player. I'm all for giving Wesley his due and recognizing the interconnectedness of the sport, but we need to make sure not to overreact to random events.

That's the biggest take away here. Yes, the Blazers' defense has looked like a sieve that got blasted with a shotgun this past month or so. Yes, Afflalo hasn't looked like the lockdown defender he was billed as. And yes, the loss of Wright has hurt the team even further in the perimeter defense department. But a large portion of the Blazers' recent struggles can be attributed to randomness and, by definition, no one knows if that will continue into the playoffs. I'm not sure I would go so far as to predict a first round victory, but the team is much better than they've shown recently.

These same questions about randomness and defense are particularly striking with the Memphis matchup. As Dave's wonderful breakdown shows, the Grizzlies shot 34% from three on the year and over 50% against the Blazers. How much of this was random versus something inherent to the matchup?

I, along with the rest of the Blazers Edge staff, will be re-watching the Grizzlies games trying to answer those very questions but a quick look at the data suggests a good portion of it was random. The Grizzlies don't take many threes and when they do they're usually open. Over the entire year, only 13% of the Grizzlies' three point attempts were contested and they made about 28% of those shots. Against the Blazers, 16% of their threes were contested but they shot a blistering six for nine (small sample size alert!). They also shot a completely unsustainable 48% on the 46 uncontested threes they took against Portland. All of this points to the matchup being closer than it appeared during the regular season.

Outside shooting is just one part of the defense which is just one part of the entire series. There are too many factors to project how the games would have been different if the Grizzlies had shot their typical percentages. It's also important to remember that a seven game series is still an extremely small sample and these types of random variations could easily pop up for either side.

What we do know is that the Blazers' defense is worse than it was before Welsey's injury but better than it has seemed lately. The Grizzlies had our number during the regular season but it's unlikely their hot shooting will continue. Memphis is certainly the favorite but the matchup might be closer than people think. We'll continue refining our series breakdown over the next few days but that's a pretty intriguing place to start.