Nylon Notebook: Wrapping Up The NBA Finals

The Warriors are champs. There will be plenty of time to digest, discover and debate why in the coming days, and I’m sure we’ll do a bunch of it here. Much will be made of how this proves a given pet theory, or necessitates going a certain direction in next week’s draft. But really what it proves is, to paraphrase the victorious coach, play really good offense and really good defense, and you give your chance to win. Before getting around to unwrapping some more big picture topics over the coming weeks, here’s a season-ending Nylon Notebook full of a few smaller points to wrap things up.

Lebron Redefines “Dominant”

I generally try to avoid bombast, and big declaratory statements, but there is one thing these Finals did not do. LeBron’s reputation is not tarnished[1. STEAMING HOT TAKES ONLY HERE.]. His legacy, if anything, grew. Had I a vote for MVP, he would have gotten mine for both quantitative and qualitative reasons.

His control over the proceedings was immense. He possessed the ball for around 11:15 of game action during Game 6, bringing his average for the series to just under 12 minutes per game for the series. The league leader in the regular season averaged around 8:15. Only six other players[2. Steph Curry, James Harden, Jeff Teague, Chris Paul, Derrick Rose, and Kyrie Irving.] had more total time of possession in the entire postseason than LeBron had in just the Finals. That possession was as much a defensive weapon as offensive stratagem. Parodoxically running stagnant offense allowed the Cavs to maintain excellent defensive floor balance, which in turn limited Golden State’s powerful fast break attack for much of the early parts of the series. Further, averaging only 3.5 turnovers per game considering his possession and usage is flatly amazing. Those plays in closer to league average hands would have produced around six or seven turnovers per game. If you thought Cleveland’s offense was a slog, add in three more empty trips followed by Warriors’ run outs per night and suddenly it becomes a much more one-sided whitewash of an affair.

While the usual suspects will undoubtedly make hay over his frank admission that without Irving and Love the Cavs simply “ran out of talent” against Golden State, it’s not really arguable this is exactly what happened. Prior to Game Four, I wrote about just how much of the load he carried, and how poorly possessions which didn’t go through LeBron tended to end up. That didn’t really change in Games 5 and 6, really, at all. Shots coming as a result of the LeBroffense resulted in 51.3% eFG over the series while everything else ended up at a putrid 30.7%. The only thing keeping the Cavs offense even marginally competitive was the fact that over 60% of all field goal attempts were as a result of the LeBron shooting, passing, or directly from an offensive rebound of a LeBron miss. Below is an illustration of just how this compares with the playoff runs of some of the other ball-dominant wings who have made the Finals since the turn of the century:

Of course, this extraordinary usage came at a cost in terms of wear and fatigue on LeBron. I don’t want to overstate any possible causal relationship, but James had an eFG of 50.6% in the first 6 minutes of quarters during the Finals, while during the second part of each period, or in the overtime sessions of Games 1 and 2, he had a 37.4% eFG. Even with some of those late game foibles, I don’t think we have the statistical vocabulary to demonstrate exactly how sui generis  LeBron’s command over how this series proceeded really was.

Pace Matters

This series drove home just how imperfect a measure of speed of pace, the “Pace” stat can be. Prior to the Game 6 closeout, the two “fastest” paced games according to Basketball Reference were Games Two and Three. It certainly doesn’t match up well with the feel of the series for the pace of play to actually have dropped when Golden State made their winning adjustment to the five-out, Draymond Green-at-center starting lineup. Simply counting by possessions ends up giving us a lot of confounding factors, like offensive rebounds and turnovers which can slow or increase the number of possessions without necessarily altering the speed of the game as defined more colloquially. Inpredictable’s sortable average possession lengths based on prior events is a good start, but it’s still packing a lot about intentionality and defensive effectiveness into one number in a way that’s difficult to untangle. A better analytical understanding of time as a resource on the floor, and especially time under the 24-second clock is a great research problem for somebody to take on. It just strikes me that even though these two plays (Cavs and Warriors) end with similar shots after similar amounts of time, one team is playing faster, doing more by any reasonable approximation. This isn’t a better or worse thing, merely descriptive of how quickly each team is moving.

As an aside, one of the reasons the post-up game isn’t a huge force in the modern game is how slow it can be, not in terms of “Pace” more in relation to the amount of viable options a team has on a given play. Running a strong side isolation post up is basically deciding “hey we don’t need these” about the five seconds it takes to enter the ball into the post and clear space for the big man to go to work. I would posit that intentionally operating on a 20 second shot clock while the opposition has all 24 available is giving something away.

The Little Edges

At risk of making a mountain out of a few data points, the few plays which were most telling about the series were those times when Cleveland played stout defense for 18 seconds, until the ball found it’s way to Iguodala or Livingston, who proceeded to make a calm, confident play and get at least a decent shot. Those situations only arise a few times per game, but the difference between a playmaker like one of those players ending up with the live grenade and when Cleveland faced similar situations with Matthew Dellavedova or Iman Shumpert on the ball was fairly important. On aggregate, that might be worth fractions of points per possession. But over enough possessions, that’s why one team is better than another. The gap between the Warriors and a league average offense is right around two baskets per game, and it’s all the little subtleties and differences in player skills which make up those gaps. As an experiment, it would be interesting to have people watch a game with no visible scoreboard or other clues and have them guess at the margin. My intuition is that most would be well off the mark. Which is why aggregates and analytics are a necessary companion to the eyetest which tends to remember the great plays and forget the misses and turnovers in similar situations.

Steph

At some point this offseason, we’ll break down exactly how dynamic and unusual Stephen Curry’s scoring abilities are. I’ve talked previously about how devastating his ability to shoot threes off the bounce can be. Another nuance which greatly impacted this series is his ability to eviscerate switches. This one skill went a long way towards eliminating  Cleveland’s ability to play Mozgov-Thompson lineups for extended minutes. Trapping Curry allowed the Warriors to continually play in 4-on-3 scenarios with the ball in the middle of the floor, but trapping either aggressively or simply putting multiple defenders near him was the only good option. The other possible choice for defending high ball screens would be to switch the big man defender out onto Curry. For teams with mobile bigs (a la Tristan Thompson) this isn’t usually a terrible option. Frequently in this situation the ball-handler’s eyes get big as he lines up the slower, larger defender, unleashes a flurry of dribble moves, and pulls up for a semi-contested 18 footer. In fact, on the nearly 4200 attempts this season when a player 6’5 or under took a pull-up jumper guarded by a player 6’9 or larger, they had an eFG of 42.1%. It doesn’t take great math skills to figure that just over .8 points per shot is a great for a defense.

But that’s on aggregate. Among the 38 players (mostly point guards) with over 40 such attempts, Curry was second to only Chris Paul[7. 56.3% for the Point God, who’s elbow pull up game is defense-breaking in its own right. Meanwhile, Aaron Brooks scored 19 points on 49 opportunites for an eFG of 19.4%, which would have fit right in with the Cavs for much of the Finals.] with a 54.6% eFG on 130 such attempts. [7. Among the 7 players with an eFG of greater than 50% on these shots and 41 or more attempts, 5 are members of the point guard elite with Mike Conley, Tony Parker and Damian Lillard joined by more surprising names Brandon Knight and Mo Williams on the list.] Much as Houston found out with Terrence Jones, there isn’t really a big man in the league who has shown the ability to hound Curry into shots that aren’t effective and efficient on any sort of consistent basis. With the flipside being if you crowd him to take away the jumper, Curry’s quickness, ball-handling and clever finishing allows him to continue to take advantage of the mismatch. In a way, the needing to put a second player on the ball beyond the three point line has replaced the post-up as this era’s best method of putting the defense into difficult rotations and help-and-recover scenarios.

Iggy Pop

As I said above, I would have voted for James as the Finals MVP. Still, it’s not hard to understand why Andre Iguodala got the nod. His insertion into the starting lineup was probably the turning point in the series[4. Though the cumulative effects of fatigue on the shorthanded Cavs can’t be downplayed when contrasting their overall performance coming following the extended rest, post-Conference Finals with how they looked once back in the every-other-day grind of a playoff series.], not just because of how it changed matchups, but Iguodala’s own performance. Using Kevin Ferrigan’s Daily RAPM Estimate, Iggy’s + 23.2 was easily the top individual mark for the Finals[5, Though this, like any single-baseline metric is always likely to undervalue a player in a situation like LeBron found himself whereby the marginal value even relatively inefficient shots for a superstar is pretty high because of how poorly his teammates are likely to do with those same shots themselves.].

Of course, his most important contribution was his work defensively on James. As LeBron noted in his presser, it’s never just one-on-one defense. Iggy often had help. But not a whole lot of help, and seldom enough to completely break down the Warriors overall defense in a way to allow LeBron’s extremely limited supported cast to get the openings they would have needed to be effective. It’s mostly reminiscent of Ben Wallace’s play in the 2004 Finals when Shaquille O’Neal “destroyed” the Pistons to the tune of 27 and 11 on 63% shooting but with Wallace getting very little defensive help, the rest of the Detroit defenders were able to stick close enough to the rest of the Lakers to force/allow 39.2% eFG from everyone else over those five games. The parallels are obvious, except Shaq at least had Kobe Bryant who presumably could create his own shot if needed[6. Of course letting Shaq feast was also a psychological shot at Bryant, who wanted his turn too.]. Not hard-doubling James and creating the same openings which aggressive blitzes and traps of Curry allowed Iguodala himself, alongside Draymond Green, Shawn Livingston and others to eventually exploit was a key part of Golden State’s strategy all series long. And that tactic falls apart without Iggy at least competing and providing some resistance, requiring James to burn so much energy he wore down over games as noted above.

As a final note, thanks to everyone for reading and following along with Nylon Calculus during our inaugural season this year. We’re hoping to hit the ground running this offseason with draft models, free agent valuation discussions and win projections for next season as well as some more long-term research-based material to get us through the long summer break. Much appreciate your attention, and questions, comments, suggestions and pitches are always welcome.