In search of offense
What adjustments can Golden State make to scrape together enough of a scoring punch versus Toronto’s swarming defense
Sabotage their roster with injuries, rip out their soul with the cruelest of injuries, and pit them against their most ferocious adversary yet…fine, but the Warriors aren’t going to die easily.
The narrative out of Game 5 was a Golden State fight for survival only describable through the franchise’s years of rising to the occasion when their dynasty seemed most dire. But even in the aftermath of an all-time Warriors victory and Kevin Durant’s crushing injury that may change the entire NBA landscape, Golden State is still flailing for its life in the deep end against a Raptors squad that has never flinched on this giant stage while looking up straight-up better than this depleted Warriors group for practically the entire series.
It’s unfair and shallow in respect to Toronto to say this series has been skewed by Golden State’s health; no Finals victory is a fluke, and if the Raptors win the championship, no one in or around the organization will give a damn what amount of luck and or number of breaks factored into that achievement. But the Warriors’ health can be an overarching pivot point of this series without selling the Raptors short. Put a fully equipped Golden State team out there and Toronto would still have a fair shot, but the Warriors have been a shell of themselves all series long. Durant has/had a case as the best basketball player on the planet; his presence would’ve recast everything about the series. But even beyond his individual merits, as soon as the Warriors signed Durant they leaned into an extremely top-heavy roster that sapped them of quality rotational pieces, a reality that’s crippled them throughout the series. Golden State isn’t just missing the generational talents of Durant, they’re missing a minutes placeholder at the wing who isn’t a paralyzing liability.
This is now essentially a markedly worse version of the Warriors’ 2016 team, one with less depth and one with far less quality depth considering the age-related decline of Livingston, Bogut and Iguodala. Subtract the net of Golden State’s pure advantages in offensive playing style and team 3-point shooting back in 2016 as the rest of the league has caught and adopted similar styles, and this version of the Warriors can’t even tread water versus that 2016 squad.
But still, a past versus present Warriors comparison doesn’t account for the Raptors and their destructive defensive power. And the Raptors need to be accounted for because they’ve done one hell of a job hijacking a Golden State offense that has run roughshod over the league for years via its fusion of revolutionary shooting with anticipatory passers interwoven into a whirring offensive system. Toronto has held Golden State to a 110.1 offensive rating through five games thus far, 6.4 points per 100 possessions worse than their regular season mark, per Cleaning the Glass (the Raptors held Milwaukee and Philly 7.5 points and 7.0 points per 100 below their regular season marks, respectively). In three of those five games, the Warriors’ half-court offensive rating has been 90 or below, about the same as the league’s worst half-court offense this season in the New York Knickerbockers.
Threading the line between Golden State’s well-oiled offensive sophistication and its sludgy personnel undercurrent has always been a challenge that has splintered defenses under the weight of trying to harness two styles at once. The lack of shooting and direct scoring ability from Golden State’s non-Splash Bros without — and at times even with — Kevin Durant in the lineup screams susceptibility. But the Warriors have roasted foes anyway for years with agile, quick-hitting screening actions that become amplified with their attention to weak side actions and wealth of visionary passing across the roster.
The Raptors, however, have found the balance between helping off non-shooters while still remaining wary of what they can do as screen setters off the ball. Nick Nurse and the Raptors were brilliant in Game 1 with how they treated adjacent non-shooters, and that’s continued for the most part throughout the series. Add in how Toronto has steadily grappled with defending Curry, bumping and grabbing him incessantly, and the Raptors has reduced Golden State’s offense into a bit of a slog. All series long, the astute, reactive playmaking, screening and cutting that made the Warriors into a hegemon has been stalled. Where the Warriors have generally feasted on dunks and layups, the Raptors have constructed resistance with heady, rangy defenders willing to make second and third efforts.
As the Warriors try to keep their dynasty chugging for a second and possibly third straight game, how can they cobble together efficient offense (aside from going absolutely bananas again from 3 like they did in Game 5: 20-of-42, 47.6 percent)?
The low-hanging fruit: Lineups and turnovers
Golden State has struggled all series to sop up minutes from its band of misfits that comprise the end of the team’s rotation. Shaun Livingston has been a two-way wallflower all series save for a few stints in Game 4, where he enjoyed the spoils of running off Curry backscreens (which Toronto has since started switching). Jonas Jerebko was a punching bag before Kerr shortened the rotation, and Jordan Bell hasn’t faired much better. Alphonso McKinnie tries hard, but he’s been taken to school by Leonard every stint they’ve been matched up, and his greatest offensive contributions come as a flying baseline crasher when the Raptors inevitably ignore his existence spotting up in the corners. Andre Iguodala’s cagey playmaking has dried up versus Toronto’s restricting help of non-shooters, and his shot is completely broken.
Quinn Cook seems due for a healthy uptick in minutes in Game 6 simply be being a merely competent shooter. Kerr wrestled with supplanting a deteriorating, grimy-looking Iguodala in favor of Cook down the stretch of Game 5, an encapsulation of just how dire the Warriors are for a dude to even take — not make, but take — an open 3 at this point.
The residual effects of Cook being a respectable spot-up shooter may go a long way for the Warriors, but that comes with a defensive tradeoff, where Cook has run smackdab into more picks and been more a negative than he’s been cited for. Still, considering Toronto’s been reticent to mismatch hunt with Leonard, more minutes for Cook seems tenable even if its saps Golden State of long and heady help defense and rebounding. Cook also unlocks more makeshift Draymond-at-the-5 lineups Kerr has been hesitant to deploy without Durant; perhaps that’s healthy skepticism on Kerr’s part considering Golden State’s dearth of two-way wings and how well a tough-as-nails Kevon Looney has held up for two straight postseasons now. But with Looney grimacing in pain on every possession and DeMarcus Cousins a defensive turnstile, there are combinations of Cook, Iguodala, Livingston and McKinnie that Kerr can patch together around the Curry/Klay/Draymond trio. Thus far, Green-at-5 lineups without Durant this postseason are an ugly minus-6.5 points per 100 possessions on 479 total possessions, with the offense at a morbid 98.5 points per 100, per Cleaning the Glass. The “but” however is that just 17 of those 479 possessions have included both Curry and Thompson alongside Green.
Whatever lineup machinations Kerr uses in Game 6 and possibly beyond, it’s an ancillary factor to the team’s turnover woes in terms of its offensive output. Golden State’s turnover rate through five games is an inexplicable 17.4 percent, a full turnover per 100 possessions worse than Atlanta’s league-worst 16.4 percent turnover rate in the regular season. Toronto’s defensive efficacy warrants praise with how they’ve straddled the line of roving off non-shooters and how they’ve mugged Curry off-ball the last two games. All series long the Raptors have prolonged possessions for Golden State; they’ve run Curry off the line and into a swath of waiting help defenders, and have forced extra passes through teensy corridors during connect-the-dots sequences that have typically been of the blissful ping-ping-dunk variety throughout Golden State’s reign.
But roughly a quarter or maybe a third of the Warriors’ turnovers have been the apoplectic, nonchalant variety that have hindered them for the entirety of the Golden Era. Now would be a good time to cut these in half at least.
Toronto has strangled Golden State’s normally fluid ball and player movement that has zipped a half-second ahead of defenses for years with tic-tac-toe passing. The Raptors have been more proactive with its treatment of non-shooters to avoid falling behind, and even when they have, dudes like Leonard and Siakam can sop up space in a hurry. Ditto for Gasol’s fleet and savvy emergency rotations. It’s clogged up passing and driving lanes.
It may time for Kerr to relinquish his zeal for egalitarian play and lean more into relying on power screening combinations featuring Curry, Green and Thompson. The Curry-Green pick-and-roll has wrecked the Raptors all series; Zach Lowe said on his most recent podcast, the Curry-Green PNR is scoring 1.35 points per possession, a whopping figure. The chemistry woven within the Curry-Green partnership has reached the pantheon where a defense’s response doesn’t matter. Trap Curry and Green slips into open space, while the two push the pick higher and higher on the floor to further strain a defense. Drop back and Curry reigns fire or uses the threat of his pull-up to careen into the lane. Sometimes Curry and Green simply leverage the defense against itself, where Green slips out of the screen a beat early into vacant space or Curry gets the on-ball defender leaning toward the screen before darting back the other, rejecting it all together.
Toronto has slotted Leonard or Siakam on Green for the most part, setting up a logical situation, but it hasn’t happened. Ditto for blitzes of Curry. Perhaps Nick Nurse has been skittish about sending multiple defenders so far above the 3-point line out of fear the Curry-Draymond synergy will spurn them and feast in downhill numbers-advantages situations. But he hasn’t discovered a more effective alternative; Toronto’s coverages have been a bit perplexing all series versus the Curry-Green combo, almost like they’ve been searching out a happy-medium between showing and aggressive dropping depending on how high the pick is. They haven’t come close to finding it. But maybe any coverage is doomed to fail versus Curry and Green, any defense rendered moot. In the words of Cersei and Jamie Lannister: “Nothing else matters, only us.”
Curry-Thompson screening actions can be equally as frightening for defenses, though in much lower volume. Any time Curry and Thompson intersect their paths on the court, panic is liable to ensue as defenses fall over themselves to ensure neither gets even a sliver of space.
Toronto fumbled through some of those same slip-ups early in the series, but with the increased off-ball clutching and bumping as well as a greater willingness to simply switch Curry-Thompson actions, they’ve cleaned it up since Game 2. And without Durant, Golden State’s best lineups even are basically Curry, Thompson and three rickety shooters; the Raptors have become more brazen hugging up on the Splash Bros. and closing out hard, knowing reinforcements lie in the wake that can prolong a possession.
But for a Golden State squad desperate for shot creation, Curry and Thompson milling around each other in the same area is still one of the best options available. Even a teensy error by Toronto can result in three points. The Curry-Thompson pick-and-roll is an ace still tucked away in Kerr’s back pocket. It’s almost a given at this point, Thompson will slip the pick early, but defenses still falter, consumed by terror.
The Warriors have been the best in the league for years at summoning some level of respect and gravity among its battalion of non-shooters. That’s what smart, timely screening can do….well as long as two of the best shooters in history are also on the roster to up the ante for the defense every possession. But it’s true, playing Golden State’s non-shooters and providing help defense is different than versus other teams, a cool tidbit Zach Lowe got Meyers Leonard to explain after the Warriors swept Portland in the conference finals. Toronto hasn’t let Golden State off the hook, however, like so many teams before them.
Yet, there are tricks to manufacture gravity for non-shooters. Various weakside screening actions, (i.e. pin downs, flares, pin-ins, backscreens) is the most obvious method, and the Warriors may want to find ways to dabble in some more when Curry or Klay is on the weakside simply to try to coax a second defender into abandoning help responsibilities.
Another gimmick: Put those non-shooters directly into the on-ball action. That concept essentially saved their season in Game 5 when Curry hit what was still a ridiculous 3 with Golden State on life support.
Part of the reason that Curry attempt materialized at all is because of the token action that precedes it with Draymond and Iguodala playing catch and temporarily holding the attention of their defenders.
Kerr — who has been damn impressive with his ATOs all series, floaties that have kept Golden State’s offense above the water — has unveiled this same set as well as cousins of it throughout the series.
Nearly all of them have come by design, however, (ATOs, after a Toronto free throw, etc.); it’s difficult to map out the floor like that in real-time. But even if Kerr runs a set like this after a Toronto bucket here and there where the Warriors mosey into the loose contours of the false action to provide a split second of gravity, it could be the difference in a few possessions which generally have amounted to the difference in a game’s outcome in this series.
A less sophisticated and certainly less used gravity-manufacturing gambit is swarming together to form almost a bunch formation. It’s so deceitful, I snicker just about any time I see it impose its usually unintentional effects.
Organizing the heist is straight-forward: Plop Curry or Thompson in an area of the floor and send two disguised screeners to mill around them like vultures. The worry of a Splash Brother breaking loose off a screen the moment a defender looks to offer aid elsewhere can hold the attention of that defender for an ever so slight moment that may afford a scoring opportunity to unfold elsewhere.
It feels cheap, almost feels dirty. But those buckets feel oh so good.
- Screen-the-Screener (STS) action
Screen-the-screener action is pretty self-explanatory. Where a basic screen (ball screen, pin down, etc.) would occur, there’s simply a screen before it as well, a way of making defenses fight through more screens while also accounting for more moving parts. STS pick-and-rolls are the most common version.
Golden State dabbled in some screen-the-screener (STS) pick-and-rolls in Game 2, but they haven’t really gone back to at all since. It’s difficult to say why. Maybe it’s because the Curry-Green pick-and-roll has inflicted major damage without all the elaborate extra scheming. Or maybe it’s out of fear Toronto will just switch the screener addendum; presuming Golden State targets Gasol or Ibaka with the initial screen to put them behind, Toronto could theoretically switch that first phase and disintegrate the entire premise. Of course, Golden State could essentially flip-flop the sequence of the initial screen and have Gasol’s or Ibaka’s man set the screen, putting Toronto in a sticky spot to either actively switch Gasol or Ibaka into PNR coverage or risk Siakam/Leonard/(insert Tortono defender) getting clipped slightly and arriving to the fray a little to late as Curry pulls one from deep.
All of this counter-recounter rejiggering comes with the caveat that such rigmarole is often more effective in theory than in actual practice. It requires both screens to hit flush with proper screen setup maneuvering by Curry. Van Vleet has been pretty good getting into Curry at the point of attack, and Toronto can overcompensate defending the actual ball screen, bumrushing Curry on closeouts and in rearview pursuit with so many help options off non-shooters.
- Veer pins
The veer pindown concept was made famous by the Lob City Clippers, with DeAndre Jordan meandering up to set a pick for Chris Paul before aborting into a pindown for JJ Redick. (Here’s an example with the Pacers from this season)
The play worked so well with the Clippers because opposing bigs often played drop coverage versus Paul-Jordan pick-and-rolls knowing Jordan only posed a threat as a rim roller. Well, when that defending big is hanging back in the paint or at the free-throw line as Jordan veers off course into a pin down for Redick, a open 3 for a money shooter materializes.
In the case of the Warriors, the veer pin is best applied as a counter to aggressive, trapping pick-and-roll coverages versus Curry. As two defenders swarm Curry, the ball screener bails, toggling into heat-seeking missile mode as they look to smack Klay’s defender with a pick.
There are a couple moments every game for Golden State where the chance to zip into one of these babies on the fly emerges. They’re often of this ilk, a rejected dribble-handoff (DHO) or pindown by Curry or Thompson where they cut through and cause defenders to retreat a step or two. It’s the perfect chance to find the other and lay a pick on their man to spring a 3.