Reproducibility

Posted by Ian Levy on November 14, 2012 under Commentary | Read the First Comment

Accuracy

Early-season euphoria has come to a screeching halt in Dallas. A 4-1 start has quickly become a 0.500 record with games against the Pacers, Lakers, Knicks, Sixers and Bulls all looming before the end of the month. Through the first five games of the season, the offense was a beautiful and unexpected surprise, singing at a tune of 111.3 points per 100 possessions, the third highest in the league and a full 10 points higher than last year’s average. Three straight losses later, the Mavs’ offensive efficiency has dropped to 103.6 points per 100 possessions. On some level this decline is to be expected, although not because of the stiff defensive opposition routinely applied by the Knicks, Bobcats and Timberwolves. Along with their personnel, the Mavs’ offense has changed significantly this season, and that doesn’t even consider the challenges of compensating for the seven-foot hole left by Dirk Nowitzki and his creaky knee. The early season offensive potency was based on some elements that were unsustainable, specifically an obscene shooting percentage on three-pointers.

Over the last three games the Mavericks’ offense has had less opportunities to get out in transition and less success with the pick-and-roll; a direct line to more isolations and forced spot-up jumpshots. With the news yesterday that Dirk Nowitzki’s rehab is not running as smoothly as hoped, this new shape to the Mavericks’ offense may need to be less of a stop-gap and more of a permanent solution. So what has changed about the offense this season, and which of those changes are likely to offer a chance for sustainable success?

The graph below shows the breakdown of the Mavericks’ offensive possessions from last season. For each possession type, the percentage of the offense it consistuted is graphed on the left axis, the points per possession average is shown on the right.

Mavs12

Looking at this simple graph actually can tell us quite a bit about how the Mavericks went about scoring last season. By far the most common possession type was a spot-up jumpshot, a fairly efficient option with the team averaging 0.97 points per possession. Spot-ups, of course, are not an organic occurrence. They require some other sort of offensive action to create the opportunity. Judging from the rest of the Maverick’s possession distribution, it appears that the most common actions used to create spot-up opportunities were post-ups and pick-and-rolls. Those possession types were also weapons in their own right. The Mavericks averaged 0.83 points per possession on post-ups. In the pick-and-roll, ball handlers averaged 0.84 points per possession; plays finished by the screener averaged 0.94.

The graph below shows the Mavericks offense so far this season.

Mavs13

The most striking thing about this comparison is how much better the offense has been on the whole. With two exceptions, the Mavericks have been more efficient this season in every offensive possession type. In particular their spot-up shooting has been incredible. The balance has also been much more advantageous. More possessions in transition means more high quality scoring chances. The Mavericks also have several talented pick-and-roll big men and are seeing a greater share of their offensive possessions going to the screener in pick-and-rolls.

Returning to the two exceptions I mentioned above, we do find some significance – they come on offensive rebounds and from shots by the ball handler in the pick-and-roll. The low performance on offensive rebounds is frustrating and basically feels like giving away free points, although this may be something that will improve just by simple regression to the mean. The pick-and-roll is very concerning though; Dallas ball-handlers are averaging just 0.76 points per possession in the pick-and-roll. In the first few games those struggles hadn’t yet materialized and everything clicked much more smoothly.

That 0.76 points per possession number only includes possessions that are finished by the ball handler with either a shot attempt, shooting foul or turnover. But such a low number is an indicator that the Mavs’ ball-handlers are struggling to penetrate into the teeth of the defense. It’s a problem for their own offensive efficiency but also reduces their ability to move the defense and create open shots for teammates. O.J. Mayo’s incredible shooting has covered up some of the concerns, but over the course of a long season they will continue to rear their ugly head. The high-quality looks that result from spot-ups and cuts may not come as frequently if the Mavericks’ can’t maintain a solid pick-and-roll attack.

If we look at some of the individual offensive possession data those concerns are reinforced. The graph below covers last season and shows the same information as above, but broken down by individual offensive outcome – player and possession type. I’ve only included the 15 most common half-court offensive outcomes, no transition or offensive rebound possessions.

Mavs12I

Looking at just the team graphs may give you the sense that the Mavericks offense has just undergone some minor stylistic changes, which completely obscures the reality on the ground. Thanks to roster turnover and injuries, other than the appearances by Vince Carter and Rodrigue Beaubois, none of these offensive outcomes are currently available to the Mavericks. They have quite literally had to build a new offense from scratch.

Last year, as is the case this year, the offense is driven by efficient outside shooting. However you can also see some of the offensive options they Mavs had last season, legitimate scoring options on their own, that also helped create open spot-up opportunities. Specifically I’m looking at Nowitzki isolating and working in the post, as well as the efficient pick-and-roll scoring from Jason Terry, Carter and Delonte West. Not appearing on this list is Jason Kidd’s work in the pick-and-roll; he doesn’t appear because he rarely finished pick-and-roll possessions, but had a transcendent knack to create open looks for his teammates in those situations, even without really threatening to score himself.

The graph below shows the same information for this season.

Mavs13i

Chris Kaman and Carter are doing their best Nowitzki impressions in the post, but the Mavs’ three primary pick-and-roll ball handlers are all really struggling to score on those scenarios. Carter has been very good in seasons past and it’s likely that his performance will level out. However, Darren Collison and Mayo and have never been very effective in the pick-and-roll and asking that to continue as the centerpiece of shot-creation for the Mavericks’ offense is not a viable option for long-term success. If Collison and Mayo continue their struggles in the pick-and-roll, offensive possessions will be chewed up on low-efficiency scoring opportunities and defenses will adjust, dragging down the scoring averages of screeners and spot-up shooters. If that scenario sounds far-fetched I would encourage you to watch some video of Collison and Mayo running offense for the Pacers and Grizzlies last season.

Looking at the two of them, issues seem to come from different places. Mayo’s problem in the pick-and-roll is often related to decision making. He’s more of a scorer in those situations, an acknowledged and accepted stylistic choice, but he frequently pulls up for a jumper when there is an opening to drive or forces himself into the teeth of the defense when a more cautious approach appears reasonable. Collison, on the other hand, seems to be fighting a long term battle with debilitating passivity in the pick-and-roll. He’s been better for sure this season, but he has a consistent propensity to defer to screeners. I’m not talking about choosing to pass when shooting and passing are both feasible options. I’m talking about playing the pick-and-roll in a way that makes passing the only option.

Over the course of last season with Indiana, the angles Collison took coming around screens became shallower and shallower to the point where almost no pressure was being put on the defense. The result was a largely ineffective pick-and-roll attack that had one of the most lopsided balances towards the screener of any team in the league. The Pacers’ offense was able to maintain a reasonable level of efficiency because of the isolation and post-up prowess of Roy Hibbert and David West, But their ceiling was made much lower because they couldn’t count on Collison to move the defense with his penetration.

Although the first few games were a sparkling offensive gift, the Mavericks are mostly in the same boat. They have some very talented offensive players but no one with the ability to single-handedly move a defense and force adjustments. Over the next few weeks it’s likely that their offensive efficiency will continue to level out as open shots become scarce and statistical normalization catches up with them. The quickest cure for these problems is 50 milligrams of healthy Nowitzki, but that’s a salve that may be out of reach for quite a while. In the meantime, squeezing the most out of every possession, pushing the ball in transition, attacking off the dribble at every opportunity and continuing to make tough shots are the only real band-aids the Mavericks have.

In addition to his work for The Two Man Game, Ian Levy is the author of Hickory High, and contributor to Indy CornrowsHardwood Paroxysm and Hoopspeak. You can follow Ian on Twitter at @HickoryHigh.

Statistical support for this story from NBA.com and mySynergySports.com