Why the Warriors are (somewhat) good right now, Steph Curry's season in context, and James Wiseman's rookie season.
The good, the bad, and the Biedrins.
AUTHOR’S NOTE: I wrote this on April 26, before the Warriors’ disaster of a game against the Dallas Mavericks the next night.
So why are the Warriors so fun to watch right now?
Well, Steph Curry is as good as he’s ever been. Even though Steph is shooting a few ticks below his 2015-16 rate (50.4% FG, 45.4% 3P, 90.8% FT in 2015-16, 48.9% FG, 42.7% 3P, 91.7% FT), I’m not sure that the 2015-16 version of Steph is a better player than he is right now. Maybe Steph has lost a little bit of burst — I can’t quantify that — but whatever he’s lost in physical ability, he’s made up for in composure, his ability to warp opposing defenses on and off-ball, and his ability to create efficient looks for himself, particularly at the basket, in lineups starved for spacing.
I think it’s fair to say that most Warriors fans and NBA observers would have expected and accepted it if Steph’s efficiency went down some on this iteration of the Warriors. Instead, Steph’s 66.1% TS is the third highest rate of his career after 2015-16 (66.9%) and 2017-18 (67.5%). What’s most impressive to me is that Steph’s 2P FG% — 57.4% — is only two percentage points short of his career-high 59.5% 2P FG in 2017-18. Here’s where it gets nuts; Steph’s 65.6% FG in the restricted area is the third highest percentage of his career, behind only 2014-15 (66.5%) and 2015-16 (65.9%), this despite Steph playing in lineups with 5th percentile spacing (as of April 11th). By now, I’m sure you’ve seen the countless screenshots on Twitter of Steph getting double or triple teamed above the three point line, so I shouldn’t have to explain to you how impressive it is that Steph is finishing at an elite rate in the restricted area on this Warriors team.
But wait, there’s more! In 2014-15, 39.5% of his makes in the restricted area were assisted. In 2015-16, that number goes down to 36%, but after the Warriors get KD, that number starts increasing in a big way: 43% FGM AST in 16-17, 48.7% FGM AST in 17-18, 48.3% in 18-19. But this year, without Klay Thompson and Kevin Durant to space the floor and create gaps for cutters, that number is back down to 41.3% FGM AST. So in short, despite Steph having the worst spacing he’s had in the Steve Kerr era, he’s finishing in the restricted at a near career-high rate and making more unassisted shots in the restricted area than he has since his back-to-back MVP seasons.
If you’re reading this, you’re probably a fairly knowledgeable Warriors fan, or at least one who cares enough about the team to have read nearly 2000 words about them, so I’ll assume you’re well aware that Steph Curry has made 85 three pointers in the month of April, which is the all time record for most threes made in a month by a single player. There’s two also still two games left in April because Jesus fucking Christ, Steph. Not surprisingly, Steph’s supernova run coincided with two notable events: James Wiseman tearing his MCL and Kelly Oubre missing games and returning in a bench role.
James Wiseman :(
It’s hard to write about Wiseman this year without feeling like I’m being unnecessarily cruel, but it has to be done, especially if we want to talk about why the Warriors are so fun to watch right now and what that means for this team moving forward. On April 6, a redditor made a long and scathing topic that laid out how disastrous Wiseman’s minutes were and coined the infamous nickname, Duolingo Kwame Brown. I’m linking to the thread here because I’m going to refer to it repeatedly. Through April 6, Steph Curry was posting a 119.8 ORTG in minutes without James Wiseman, which would be #1 in the league, just ahead of the Clippers’ 119.1 ORTG. Throw James Wiseman into the mix and... that number craters to a 103.4 ORTG, which is 8 points below league average this year. The lineup of Steph/Wiggins/Oubre/Dray/Wiseman rocked a 98.9 ORTG. To quote our redditor friend, “That second lineup, scoring at a rate below 1978 league-average, two years before the 3pt line was introduced? That's the Warriors starting lineup.” In a nice, and perhaps, foretelling twist, lineups of Steph/Dray/Wiggins without Wiseman and Oubre had a scintillating 132.2 ORTG. That’s, um... really fucking good.
In the Steve Kerr era, only three players have managed to play more than 100 minutes with Steph Curry and rock a negative net rating in a two man combo. Two of them play on this Warriors team; Kelly Oubre (-0.5 net rating with Steph) and James Wiseman (-8.1 net rating with Steph). The only other player in the Kerr era to have a negative net rating with Steph is... Anderson Varejao (-6.6 net rating with Steph) in the 2015-16 season, which may or may not have indicated that Vareajo playing fourth quarter minutes in the 2016 Finals was a bad idea!!! The only players to have gotten a worse net rating in any two man combo with Steph to play >100 minutes either played with rookie Curry or were out of the league the next year; Dan Gadzuric, Rodney Carney, Jordan Crawford, Kelenna Azubuike, and Mikki Moore.
And while I wish I were done, I’m not, there’s still one more impact stat I want to note: James Wiseman ended his rookie year with a -4.0 OBPM on 55% TS. For reference, Andris Biedrins in 2012-13 took 22 shots on 45% TS in 495 minutes and had a -4.8 OBPM. In Jordan Poole’s dumpster fire of a rookie season, he had a reasonable claim to being the worst player in the NBA with a -4.4 OBPM and 45% TS. So Wiseman, despite being 10 percentage points more efficient than late career Andris Biedrins and rookie year Jordan Poole, was only marginally less atrocious than they were per OBPM. This is, um, not good. But, something I think is interesting and perhaps, hopeful, is that on April 6, the day the aforementioned reddit post was made, Wiseman had a -4.4 OBPM. Wiseman played two and a half more games after that post was made and his OBPM increased by .4 percentage points. Small sample size? Totally. But Wiseman did provide solid minutes and production against Milwaukee (13 points on 6/13 shooting with totally acceptable defense) and Washington (9/11 shooting for 18 points in 23 minutes). If you’ll recall, Steve Kerr had recently elected to actually... run pick and rolls with James Wiseman, which speaks to a point I want to make here: Wiseman playing as bad as he did was entirely avoidable.
Why it’s not all Wiseman’s fault
I was not a James Wiseman fan pre-draft — I felt apprehension towards pretty much every top prospect, but the only thing I knew with any level of certainty is that I did not want the Warriors to pick Wiseman — but he has obvious physical traits that could have made him useful on offense, had Kerr been “chasing wins” all season. Part of that utility, of course, is predicated on putting Wiseman in optimal situations, but it’s fair to say that Kerr did pretty much the exact opposite of that for most of the season. There’s reasons for that, some of which I understand, some of which I really don’t, but let’s be real, Wiseman’s rookie season has been an abject failure and that is in large part because of Steve Kerr.
Wiseman started out the year shooting it with confidence and on volume from three. He ended the year having shot 37 three point attempts on 32.4% and... 81 mid range jumpers on 33.3%. On the year, Wiseman shot 35.3% on jumpshots (all the shooting splits are here). Although the small sample size on three's makes it hard to say with confidence that the 33% 3P would hold up on high volume, there’s simply no reason for Wiseman to shoot nearly three times as many mid range jumpers as he did threes, especially when their percentage difference is marginal. Wiseman even took 42 hook shots this year and connected... on 28.6% of them. I really don’t care to know whether these mid range jumpers and hook shots were shots that Wiseman wanted to take or if Kerr wanted those shots, because at the end of the day it’s on Kerr to either tell his player to stop taking bad shots OR to stop putting his player in situations that lead to bad shots.
Wiseman shot it at 75.4% from the restricted area, which is actually quite good! As a point of reference, Bam Adebayo is at 74.7% on the year, DeAndre Jordan (who Wiseman is often compared to) is at 80.8%, Andre Drumond (another frequent Wiseman comparison) is at a disgusting 52.1%, and Anthony Davis is at 76.5% on the year. But here’s where things get a little weird — of all the players to have been a roll man on the pick and roll for the Warriors this year, Wiseman has the lowest points per possession (PPP) and he shoots only 52% on this play type, which would indicate that a significant amount of Wiseman’s PNR shots are occurring outside of the restricted area. The shooting splits would back this up — Wiseman took 83 shots in the paint non-restricted area this year at a 31.3% clip. He is also only a 62.8% free throw shooter, so there’s a significant body of work here demonstrating that Wiseman doesn’t have great shooting touch right now, so when you look at his percentages across various shot types and locations, it seems even more imperative that his diet of shots be constricted to 3’s where the PPP yields are better than the midrange on a similar percentage (albeit smaller sample size at the 3 point line) and dunks, whenever possible.
At this point in time, James Wiseman can’t be trusted to make complicated reads and while Wiseman’s low PPP in PNR might lead you to believe that the PNR is fool’s gold, I offer a different conclusion; the James Wiseman PNR would look a whole lot better if Steve Kerr had committed to playing lineups with good spacing.
Warriors fans will note that in the absence of Wiseman and temporary absence/return of Kelly Oubre, that Steve Kerr has downsized out of necessity and played lineups with a lot more shooting. James Wiseman, unfortunately did not have the luxury of playing in lineups with real spacing. While I’m pretty sure that this Steph supernova explosion would not have happened with Wiseman, I do wish we’d gotten to see Wiseman play in lineups that are maximized for chasing wins. As it stands, Wiseman played 598 of his 836 minutes with Andrew Wiggins (-15.5 net rating in a two man combo), 533 minutes with Steph ( -8.1 net rating) 527 of his minutes with Oubre (-10.8 net rating in a two man combo), and 481 of his minutes with Draymond Green (-8.4 net rating in a two man combo).
Here are the 5 man lineups that Wiseman is in that received more than 20 minutes this year (I’ve bolded the positive ones):
247 of Wiseman’s minutes with Steph/Wiggins/Oubre/Dray/Wiseman (-12.8 net rating)
47 minutes with Steph/Wiggins/Bazemore/Dray/Wiseman (-6.2 net rating)
44 minutes with Steph/Bazemore/Oubre/Dray/Wiseman (+27.8 net rating!)
38 minutes with Steph/Wiggins/Oubre/JTA/Wiseman (-6.7 net rating, almost all in the Brooklyn and Milwaukee games when Draymond was recovering from Covid-19)
33 minutes with Poole/Wiggins/Oubre/Dray/Wiseman (-22.1 net rating)
24 minutes with Poole/Wiggins/Oubre/JTA/Wiseman (-43.1 net rating)
24 minutes with Steph/Wiggins/Oubre/Paschall/Wiseman (-26 net rating)
I personally consider the real shooters on this team, this is to say the ones that teams actually close out hard on, to be Steph, Jordan Poole, Mychal Mulder, and Damion Lee. Kent Bazemore is shooting it at a career high 41.7% clip, but that’s 6 points above his career average 35.6% and Andrew Wiggins is seeing a similar #StephBump of 38.2% from his 34% career average (this was written before the April 27th Mavericks game) , so for the sake of this thought exercise, I’m not counting them as “shooters.” If you filter for more than 5 minutes of total playing time, these are the lineups that James Wiseman played with this year that included two shooters (I’ve bolded the ones with a positive net rating):
16 minutes of Mannion/Poole/Mulder/Paschall/Wiseman (+51.7 net rating)
15 minutes of Steph/Mulder/Wiggins/Green/Wiseman (-33.1 net rating)
14 minutes of Steph/Lee/Wiggins/Draymond/Wiseman (+26.5 net rating)
13 minutes of Wanamaker/Poole/Mulder/Lee/Wiseman (+42.9 net rating)
12 minutes of Steph/Lee/Oubre/Green/Wiseman (-18.1 net rating)
10 minutes of Steph/Mulder/Bazemore/Green/Wiseman (-39.7 net rating)
9 minutes of Mannion/Poole/Mulder/JTA/Wiseman (+35.5 net rating)
8 minutes of Mannion/Poole/Lee/JTA/Wiseman (+54.5 net rating)
7 minutes of Poole/Wiggins/Lee/Bazemore/Wiseman (+76.9 net rating)
7 minutes of Steph/Mulder/Lee/Green/Wiseman (+39 net rating)
6 minutes of Mannion/Lee/Mulder/Wiggins/Wiseman (-34.1 net rating)
6 minutes of Poole/Bazemore/Lee/Oubre/Wiseman (+13.3 net rating)
6 minutes of Steph/Lee/Bazemore/Wiggins/Wiseman (-53.8 net rating)
6 minutes of Steph/Lee/Oubre/JTA/Wiseman (+26.7 net rating).
Am I scraping the bottom of the barrel here? Yes, absolutely! That’s exactly what I’m doing! I’ve had to go to great and ridiculous lengths to find the lineups that James Wiseman played in that contained LITERALLY JUST TWO CREDIBLE NBA LEVEL 3 POINT SHOOTERS. You’ll note that he didn’t play more than 16 minutes in a single lineup with TWO CREDIBLE NBA LEVEL THREE POINT SHOOTERS! THAT’S NOT EVEN AN ENTIRE HALF OF AN NBA GAME!!! It’s mind boggling stuff, truly. So when you do the rough math, you’ll see that Wiseman played only 135 minutes this entire season in lineups with two or more shooters, which would account for 16.1% of Wiseman’s total minutes on the year (keep in mind I didn’t go to lineups to receive less than 5 minutes so this is a rough estimate, probably off by a percentage point at most). You’ll note also looking at the above lineups, that when I got into the barrel scraping lineups that received less than 20 and more than 5 minutes on the year, a fair amount of those lineups look like garbage time lineups.
The vast majority of the lineups that James Wiseman played in this year were antiquated lineups with exactly one credible NBA level shooter on the court at any given time. With all that in mind, I think it’s safe to say that while Wiseman’s offensive impact was horrendous this year, it is not at all entirely his fault and Steve Kerr consistently put him in positions to fail. If you’ll recall, the entire reason that I did this deep dive just now on the lineup numbers was to back up my thesis that Wiseman’s low PPP as the PNR roll man was likely connected to bad spacing. While I don’t have the capacity to look at Wiseman’s playtype PPP numbers in those specific lineups, it’s glaringly obvious that the bad spacing lineups with Wiseman performed terribly and that the lineups with requisite levels of spacing in the modern NBA didn’t get enough minutes to really show anything but statistical noise.
As I said before, Kerr made choices to play these lineups with Wiseman and to spend 40 or so games trying to let Wiseman run Zaza Pachulia style dribble hand-offs (DHO’s) and encouraging, or at least tolerating (for god knows what reason) feckless postups and mid range jumpers. These tactical decisions did not lead to wins, nor did they make James Wiseman look desirable to other teams. If you believed pre-draft that Wiseman was going to be a star, or at least provide a positive impact in a Javale McGee type of rim running role, well, that didn’t happen and Wiseman’s minutes indisputably contributed to several losses this year.
What happened instead is that Wiseman looked worse as the season progressed, save for the few games before his meniscus tear — the Chicago game at home where Kerr unleashed the Steph/Wiseman PNR — and Wiseman’s trade value has likely tanked Kerr’s inability to either win games with Wiseman or help him show positive, linear progress throughout the year will in turn make it harder for the Warriors to build their team next year.
In my next post, I’m going to go deeper on some of Steve Kerr’s coaching decisions this year and try to make sense of what was and was not defensible. See you there!