Nylon Calculus Week 5 in Review: Giving thanks for stats.NBA.com

We’re already done with about a fifth of the NBA season, as outlandish as that seems. The NBA map is solidifying, and projections are giving way to totals. We have a better sense of what the season will be, and the standings and leaderboards are a bit more plausible.

The Golden State Warriors have overtaken the Los Angeles Clippers by Basketball-Reference’s SRS, and the Los Angeles Lakers have fallen back to Earth. Russell Westbrook, however, is flirting with a triple-double, which shouldn’t be possible this far into the season. And he may not even finish top three in the MVP voting.

What else has been going on with the league?

Redesign at stats.NBA.com

The relentlessly thorough stats.NBA.com got a redesign, and there are some crucial changes. Firstly, navigation is easier thanks to an improved menu, links to popular pages on the side, and a better search engine. The site itself was a bit of a labyrinth of stats, so any bit of help is appreciated. My favorite addition though will probably be the change in how they display video linked stats. For example, go to Marc Gasol’s page and click on his made 3-pointers. It’ll take you to a page that displays a shot chart, a list of every made 3-pter this season, and an embedded video playlist that shows every trey. You can do this for every linked stat, from steals to rebounds.

Read More: The best single game performances of the 2016-17 NBA season

Pages now come with rankings, as you can see here. It’s a useful addition because too many people have resorted to manually counting when citing a ranking. But I’m annoyed by the death of the separate page feature and the ubiquitous “load more” button infesting yet another page. If you want to cycle through entire lists, you either have to keep clicking the load more button while it slowly loads or find another method to complete the task.

People have lamented on the state of the online box-score — ESPN, for example, only lists the players currently on the court or the starters, and you have to manually click “see more” twice to look at every player — so I’d recommend at least trying their box-score. You can cycle through the traditional box-score layout to advanced stats to miscellaneous to SportVU stats. Most helpfully, the box-score stats directly link to video clips a few minutes after the game ends, so you can watch, say, every shot Kevin Love made in his 34 point 1st quarter explosion or something less obvious, like every Mason Plumlee assist in the game the Portland Trail Blazers had against the New Orleans Pelicans.

I know people still have reservations about the site and its layout, but its breadth is fantastic and the video library is a treasure — why should a sports league give free access to every play from this season just a mere minutes after games finish? From SportVU stats to the advanced lineup filters here — you can see how 2-3-4-or-5 man combos do with or without various players — the information is a treasure trove. So, please, cherish what we have and enjoy it while it’s here. I recently pessimistically commented on the absence of play-by-play video links, but even those are coming back.

Semantic Antics: Greg Oden’s draft legacy

Every so often, a discussion arises about either Greg Oden or number one pick draft busts, as the two are forever entangled. As a Portland Trail Blazer fan, I’m biased in my support of Oden as a draft pick, but I think this point is objective: he was a bust because he was too injured to play, not because he wasn’t good enough. I believe that’s a crucial distinction. A “bust” label should denote a player who should have never been taken number one, or close to it, because the player’s quality was never that great. We need another term for guys who had the talent but not the opportunity due to health — we should at least re-categorize them as injury busts.

The greatest offense in NBA history

Currently, Golden State’s offensive rating is something that we have never seen before. Not only do they have the best offensive rating ever, at least according to Basketball-Reference, but they have the best rating relative to the league average too. You need that adjustment because an offense may not necessarily be better if you don’t account for the environment of the league itself. Given some of the advances in the game and the large swath of the world that we now pull players from, one has a very strong argument for this being the greatest offense in NBA history. I know we all tend to live in the moment and ignore history, but for now let’s celebrate this team and what they could mean for the history books.

Looking at the greatest NBA offenses below, you have a number of Steve Nash offenses, either with Dirk Nowitzki or with the Suns; the late-era Jordan Bulls are represented twice; a peak Karl Malone-John Stockton team makes it, surprisingly; there’s a Larry Bird Celtics squad of course; an infamous high-octane Nuggets team with Alex English and Dan Issel that scored 100 in every game and allowed 100 in every game; the famed early 00’s Kings make the list, thanks to the league average adjustment; and now it appears we’ll have multiple Stephen Curry-era Golden State teams.

This incarnation is tipping the scales thanks to the purest scoring attribute: they’re making their shots at absurd rates. Their true shooting percentage is well north of 60 percent, which is ridiculous. Only 16 players cleared 61 TS% last season, for instance (with at least 100 field goals attempted), and two of those guys played for Golden State. And based on their personnel and how well they did last season, I’d expect this to continue unabated.

Table: greatest NBA offenses, 1955-2017, source: b-ref

Down in Dallas

Last season was a disappointment for Dallas, which makes this current one painful. It’s not a good sign when you have to click over to the third page on Real Plus-Minus to find the highest ranked Maverick. Their offense is what has tanked, and they’ve been challenging the historically incompetent 76ers for the worst mark in the league. Sure, they haven’t had Dirk Nowitzki for most of the games, and as a team they’re collectively under-performing on their 3-pointers, but they’ve sunk so far to the bottom that it’s tough to realistically imagine them competing for a playoff spot. It could be their first terrible season since 1999. Even when they missed the playoffs in 2014 they were still 0.500. They’ve had remarkable consistency, but everything fails with time.

On the bright side, the Mavericks are experiencing some bad luck, which never lasts. Opponents are shooting 41 percent from behind the arc, and that could translate to nearly 5 points per game, depending on what assumptions you make. Some of their own players could make more of their 3-pointters too. But the primary issue is their offense, and over the start of the season it’s been centered on Harrison Barnes in the mid-range area, where he’s been one of the most prolific shooters from that distance. You can see some video clips here. And needless to say, revolving around Dirk Nowitzki’s mid-range game was a good idea, but Harrison Barnes is no Dirk Nowitzki and consequently their offense is sputtering and contending for last place.

Zach LaVine, current God of Dunking

Last week, a video surfaced of Zach LaVine dunking from the foul line … while doing a 360. I’m sure virtually everyone reading this has seen the dunk, but I thought it merited a comment nonetheless. While James White has had a number of foul-line dunk variations, Zach LaVine’s style itself is mesmerizing and he sells the dunks as well as anyone with his execution. And, perhaps, lost in all this was an in-game dunk he had against the Phoenix Suns, where he viciously posterized Alex Len with a candidate for Dunk of the NBA Season, as seen below.

Early Defensive Player of the Year discussion

Awards discussions often occur way too early, so I thought I’d tackle this one before arguments proliferate. Typically, players with the most impressive basic defensive stats, like blocks and rebounds or players from the most impressive defensive teams get all the attention. No one stands out in that respect, even Hassan Whiteside, who’s lagging behind Anthony Davis blocks per game.

The advanced numbers, so far, see this as a two-man race at the top between Draymond Green and Rudy Gobert. Both players look elite on defense in both ESPN’s Real Plus-Minus metric and the box-score measure Box Plus-Minus. Atlanta, who owns the best defensive rating in the league, is led by Paul Millsap, not Dwight Howard, according to most advanced stats. The Clippers, who have the second best mark, have two of the other leaders in both metrics: Chris Paul and DeAndre Jordan. I wouldn’t expect Paul’s defensive rating to hold, however; it’s the result of a flukishly good defensive rating from the team, which I doubt will hold for the entire season. Kawhi Leonard, as odd as it seems, is far from defensive leaderboards. The Spurs have a middle-of-the-pack defensive rating and the opponents have confusingly scored over 11 points per 100 possessions better with the Claw on the court. I don’t expect that trend to hold either, but even DBPM, which is box-score only, doesn’t favor him.

Of course, those defensive measures are not ideal, especially at this point in the season. And I have to mention a subtle point with large implications for some players: DBPM uses interaction stats, like TRB%*AST%, that DRPM does not. That may not sound important, or it may sound like gobbledygook, but the result is that Russell Westbrook is fifth in DBPM, and I doubt that even Oklahoma City fans would try to justify that ranking. This is where BPM and RPM diverge, besides the plus-minus influence. We’re going to see advanced stats cited frequently in the coming months, but we have to understand how they’re calculated and what their strengths and weaknesses are — or else we’re all just flittering about with nonsense,

Block efficiency measure

Continuing my examination of foul types in the play-by-play files, I thought I’d make a more accurate block statistic that utilizes the available information. This is an efficiency measure, and measuring efficiency is about rates: how many successes have you had versus the total number of opportunities? This isn’t opaque, because there’s the straightforward example of field-goals made versus field-goals attempted — that’s efficiency. But for other stats, like blocks, this is tougher to judge, especially when you don’t have SportVU stats to lean on.

For an easy to understand measure, take the number of blocks a player had and divide it by the number of goaltending violations, shooting fouls, and blocking fouls (I also included a 1 in the denominator to sidestep division by zero). This is not perfect, because many of those fouls can occur when a player is not going for a block, but it’s a better indication of how many successful blocks a player has compared to the negative actions that can occur when going for a block.

You can see a leaderboard below for 2016, and one player obviously stands out: Hassan Whiteside has a block efficiency ratio over two, and the nearest guy is not much over 1.5. The filter here is minutes, with a minimum of 1000, so big men are clearly favored by the measure (I didn’t have a blocks minimum.) Their high foul rates don’t sink them. You do have some perimeter-oriented players, however; the Defensive Player of the Year, Kawhi Leonard, isn’t a shocking inclusion, but Kevin Durant is at least not one you’d expect. He has curiously low foul rates. But the most surprising player has to be Marvin Williams, who averaged a scant 1 block per game. The lowest ranked player, by the way, is Evan Fournier, thanks to a bizarrely low total of two blocks.

Table: 2016 block efficiency

Not to be outdone, Anthony Davis actually had a higher efficiency in 2015, and Hassan Whiteside’s was actually a lot lower: 1.54. Rudy Gobert and Pau Gasol, again, do very well, and the list is completely dominated by big men. John Henson had few shooting fouls, and thus made the list. No one else is a big surprise, and Tim Duncan has a much better ratio — his final season was actually a bit of an outlier. The highest ranked perimeter player, by the way, was LeBron James, while Ramon Sessions was the lowest ranked overall.

Table: 2015 block efficiency

Looking at the 2014 leaders, Lord Tim Duncan reigns supreme. Anthony Davis and Serge Ibaka, two high-block total masters, rank high due to sheer force. But Terrence Jones, Pau Gasol, and Andrea Bargnani rank high too — that can’t be a good sign for a defensive measure. Gary Neal ranked last, as small guards dominate the bottom ranks.

Table: 2014 block efficiency

A block efficiency measure is a fun little stat, but there’s one major issue: shooting fouls are not all the same. Fouling the jump shooter is not the same as stopping a Dwight Howard dunk by sending him to the line. Plus, fouls are a measure of defensive activity in some instances, where slower defenders, like Pau Gasol, are too far out of position to even foul.

But data-mining might be able to offer a solution: what happens when we look at shooting foul locations? That’ll be the next topic in Week in Review….