Nylon Calculus Week 8 in Review: Displaced centers and luck

Lord Tim Duncan had his jersey retired, and we’re reminded yet again of the unrelenting passage of time. There are a wealth of amazingly talented players out there, and I do hope we can probably appreciate them all. They can disappear within the blink of a season, as NBA history is littered with the ghosts of players like Bill Walton and Gilbert Arenas. Sure, we have the indestructible LeBron James, but we don’t have a clear expiration date for any of these guys. Not everyone can have Tim Duncan’s career.

And with that, let’s take a look back at the last week in basketball….

New Collective Bargaining Agreement: No lockout imminent

This is good news, because it appears the threat of a lockout is gone and we’ve had no draconian changes to the NBA, like a higher age requirement or the players getting a smaller slice of the pie. You can see a breakdown of the changes here.

Teams will now be able to extend their own players an extra year on some deals, provided they meet requirements, and an “age 36 rule” about contracts has been changed to 38, which by sheer coincidence would benefit union president Chris Paul and vice president LeBron James. I do like the change with minimum and rookie-scale contracts though: they’ll be tied to the cap, like max salaries are, so they’ll rise faster now. Plus, and I find this most interesting, there will now be two-way contracts for the NBA and the D-league, whereby teams will get two more roster spots for players who have a toe in each league. They’ll be paid more than pure D-leaguers too.

Read More: Six creative ways NBA teams are adapting the pick-and-roll

For an environment concerned about the elite players and owned by billionaires, any attention paid to the poorest basketball players is great — remember, these guys can’t work for 40 years in their field, so they need big paydays to justify the risk in employment.

However, as Kevin Pelton noted, we could see some negative unintended consequences. Players could be coerced into forcing trades earlier, when they’re still on their rookie contracts, because they’ll still be eligible for the new extra year of the extension later down the line. But generally speaking, it appears there was no obviously detrimental item added to the agreement, and the NBA is flourishing. Let’s hope that continues.

Goodbye

I’m not sure what to say, but I had to comment on Craig Sager’s death. What’s touched me the most is his positivity, and how so many people he encountered had something great to say about him. He was always a delightful presence, and he was always easy to spot from the stands no matter how bad your seats were. I think Gregg Popovich said it best (credit goes to USA for reproducing the quote):

"“On a day like this, basketball has to take a backseat as we all think about somebody who was very unique, very special. Whether you really knew Craig or not, you got the feeling that he was a special person in a lot of different ways. And right now, I just feel for his family. To talk about him being a professional and good at what he did is a tremendous understatement.“All of us who knew him understood that that fact was what he was all about as far as work was concerned, but he was a way better person than he was a worker – even though he was amazing in that regard.“He loved people. He enjoyed pregames, postgames. He loved all the people around him, and everybody felt that. So, the most amazing part of him is his courage. What he’s endured, the fight that he’s put up, the courage that he’s displayed during this situation is beyond my comprehension. If any of us could display half the courage he has to stay on this planet – to live every life as if it’s his last – we’d be well off.“I’ll miss him very much.”"

Flightless Griffin

Blake Griffin will miss three to six weeks for arthroscopic surgery on his right knee, which should obviously hurt an already reeling team that’s fighting for higher seeding in the west. The common thinking is that the Clippers will weather the storm just fine because they have in the past when missing one of their stars. But they have a weak bench, and without Griffin they’ll be even more susceptible to ruin if another injury occurs — imagine this team for a few games without both Blake and Chris Paul.

The silver lining is that this event could boost Chris Paul’s MVP chances because he’ll have noticeably less help, and if the team does well without Griffin it’ll be cited in Paul’s case. That’s enough to make a difference in a wide MVP race. Plus, as Kevin Pelton pointed out, Paul’s individual stats increase without the All-Star power forward. As long as Blake isn’t out for longer than expected, and no other key player goes out, they should glide for a few weeks dropping only two or three more games than they otherwise would have. It surprises many people how wins are actually lost when significant players are lost, like when Zach Lowe mentions Reggie Jackson here, but the math commonly unfurls like that. The real issue is Griffin’s future.

The last Noel in Philadelphia

Nerlens Noel may as well still be injured — he was been benched by his team after comments about his frustration with the playing time and the center logjam. And now it’s all but official he’ll be traded. This is unfortunate, and he’s not being favored over Jahlil Okafor, who is by all objective standards a worse player and a poorer fit next to Joel Embiid, purely because of the money. Noel’s rookie contract ends first, and they probably don’t want to pay him now that they have Embiid. I can understand the reasoning there, but this could end up hurting them. If Embiid suffers another major injury — please no — then they’ll have to depend on Jahlil Okafor, who I don’t believe is an NBA caliber starter and will consistently submarine their defense. Plus, even Embiid endorses Noel, pointing out that they haven’t even tried the duo together and how well they could perform on defense.

The lie in Dallas

Dallas just blew-out the Kings by 20 points, and it’s an encouraging sign, but this season has been a disaster for them. Coming into that Sacramento game, the Mavericks were 6-20 and looking at the bottom of the Western Conference. That’s a major concern following a season where they won 42 and followed with a big free agent signing in the summer. As it is with many under- or over-performers, some of this can be explained by fluky opponent 3-point percentage. Opponents are shooting nearly 40 percent against the,, and if one regressed that percentage to league average, they would be improved by up to 3.4 points per game — depending on the assumptions made on offensive rebounds after misses, fast breaks after said misses, etc. And yes, I’ve discussed this before with Dallas, but it’s not likely those opponent stats will hold up for the rest of the season. You can cycle through their SportVU nearest defender stats; they actually have a higher proportion of tightly guarded not, and not wide open ones.

Again, even though their defense has been unlucky, Dallas’ offense has been the biggest disappointment. Maybe this will surprise people who jumped the gun and vindicated the Harrison Barnes signing early, but Harrison’s game of midrange volume with few assists has not been beneficial to the team as a whole. The fact that Dallas has been better on offense with him off the court, despite playing with the starters, should be a pretty good indictment of this 20 point per game scorer. His defensive on/off number is wild, but using some intelligent adjustments and his value is actually below average. The nine-game losing streak they suffered has derailed their season, and the discussion now about the team concerns how they could sell off their assets to build for the future. Yet with a bit of luck, they might have been still in the playoff race at this point.

Houston Rockets’ ascent slowed down

Unfortunately, another Western Conference team has lost a key member: Clint Capela has a broken fibula and will be out for several weeks. For a team realizing its potential, this is saddening. However, at least the center market has a surplus. As we’ve seen from the cases of Denver, Philadelphia, and others, quality big men are forced to the bench because there aren’t enough minutes, so Houston should be operating from a position of power here. And hey, it’s a chance to pick-up another 3-point shooter: I don’t think the Nets want to hold onto Brook Lopez, and there’s little interest in him league-wide, but Houston could go all-in on offense and start Brook until Capela returns. (I’m mostly joking here. Mostly.).

Frontcourt musical chairs in Denver

The Nuggets have struggled with juggling two big men with tantalizing potential — the sophomore Nikola Jokic, who looked like a star last season through advanced stats, and Jusuf Nurkic, who was out much of last season with injuries. Denver began the year starting both guys, and then they oddly benched the superior Jokic for the more traditional power forward in Kenneth Faried for a stretch of games. After that, Darrell Arthur another big man, was used next to Nurkic for three games.

However, they switched to Jokic as the starting center with two wing players at forward (Wilson Chandler and Danilo Gallinari) last Thursday, immediately resulting in two solid wins over the Knicks and Blazers. Both games were at home and those opponents aren’t strong, but the point margin suggests improved play — and this is what every armchair coach has been saying about starting Jokic and playing small more often.

As the Nuggets continue to use that lineup, we have to consider that these performances will be the norm. Jokic’s value was depressed by being out of position and the middle of the court was clogged with two big men. Denver’s perimeter players were effectively playing on a smaller court, but with Nikola driving lanes are open and his incredible passing skills can be put to use. We don’t have to create some Gordian knot about second-year players, politics, his personality, and the team to explain his poor start to the season — he’s not a power forward, and he played so well as a rookie that a bit of regression is normal anyway. The future of the team is with him at center, and the organization needs to act on that fact.

If you haven’t seen much of Nikola Jokic and consider him a tier or two below other young prospects, I’d implore you to watch Denver footage immediately. He already has few peers among passing centers. His ability to find guys at the rim while in the high post is uncanny. Here’s an example with Gallinari and a bullet pass. Jokic has decent range too, so when he runs a give-and-go his defender has to stay with him. He’s at 44 percent on his long 2-pointers, and he’s still developing NBA 3-point range. If his defender jukes at a nearby teammate, Jokic can nail that jump shot — and when defenses have to worry about pressuring his shot, the court opens up. When he can make a pass like the one below to Kenneth Faried, you can see why some Nuggets fans are so excited. (Since my week in review is late this week, Jokic’s 27-15-9 game against Dallas isn’t technically covered by the week. But yes, we have more evidence already of his value at center.)

Team skill in outperforming point differential

“There is something fascinating about science. One gets such wholesale returns of conjecture out of such a trifling investment of fact.” – Mark Twain

There’s been some discussion recently about explaining wins. Traditionally, analysts use a Pythagorean expectation, which is just a way of translating point differential to a win percentage. But two teams have, once again, defied their numbers, which has caused people to wonder what the basic point differential stats are missing. Memphis currently has a negative point differential, yet they’re 18-11 — and they have a long track record of outperforming their margin of victory. Minnesota, meanwhile, has been disappointing, and by point differential they “should” have won three more games than they currently have. They too have a history with Pythagorean expectation. Are the numbers actually missing some information, or is this all much ado about noise?

To look at this question in-depth, I set up a simple test. Consider “luck” as the difference between actual wins and expected wins due to point differential. If there’s an actual team skill in luck, then we should see a correlation between luck during the first half of the season to luck during the second half of the season. This was easy to grab from stats.NBA.com, and you can see the results below from 1997 to 2016. There’s no clear relationship, and even if there is one it doesn’t appear strong.

This is similar to what I did for 3-point defensive luck in that I chose January 20 as the demarcation line, adjusting the date for 2012 and ignoring 1999 altogether. This means a team’s season isn’t perfectly split by 41 games, but you can adjust “luck” by making it a per game stat. The correlation there is close to zero: 0.043. I can make strength of schedule and other adjustments, but remember this is about a win differential — it’s about outperforming expectations, not how good you are. That correlation isn’t actually zero though, so perhaps we can have a more intelligent expected win measure. We can’t explain everything — the differences appear to be mostly noise — but a small improvement might be possible.

The Memphis Grizzlies are the case in point. Since 2010, their seasons have mainly resided in the first quadrant: during the first half of the season they have positive luck, and this carries over into the second half of the season with one exception. The Timberwolves, meanwhile, during the same time-frame are a third quadrant team: they generally have negative luck that carries over from one half of the season to the other. We may be falling victim to the Wyatt Earp effect, but for a small subset of teams it does appear like “luck” exists as Memphis being able to repeat this feat continuously makes it quite unlikely to happen purely by random noise.

There are many plausible explanations. Kevin Ferrigan just wrote about one cause, and I’ve dabbled in explanations for a while. Here’s one that’s intuitive for basketball fans: teams with good benches, especially deep benches, can run up the score during garbage time, when games are basically already decided; and vice versa. For another quick test, let’s look at full season luck versus average bench BPM, which I’ve defined using the number of non-starts for every player.

You can see the results below. It’s noisy, and it’s not perfect, but part of the theory holds up: there’s a slight negative relationship between win luck and bench strength. That doesn’t explain Minnesota and Memphis, sure, but it (partly) explains why Golden State had significantly more wins last season than San Antonio even though, surprisingly, they both had very similar overall point differentials.

There are a few other explanations too. One that I’ve held for a while was offered in a recent article about the Memphis Grizzlies: clutch play is all about a slow pace, the lack of fast break opportunities due to teams taking fewer risks, and defense, thus teams that naturally play that way have an advantage and win more games. After I wrote this, the Grizzlies lost a close, overtime game versus the Celtics. While there’s some evidence they have supernatural abilities to win, it’s still not likely to hold up forever.

Next: Is Reggie Jackson disrupting the Pistons offense?

You’ll note that description describes the Grizzlies pretty well. Of course, we might be reaching for anything that looks like a pattern to organize the chaos, but there’s still potential out there to modify our understanding of how point differential and team strength translates into wins. This is about the first step, identification, in my own little experimentation hub of my weekly article, but there are some paths now for the future and for greater team metrics. We just need to realize the benefits could be pretty small; luck does not appear to be an actual skill in most cases.