Straws That Stir The Drink: Half Court Creators

Earlier this week, I looked at how good half court offense relies to some degree on a team having one or more creators. The common expression is “ability to break a defense down,” which in practical terms means forcing a second and possibly even third defender to guard the ball, leaving others open to make less demanding plays. Against modern exacting and drilled NBA defensive schemes, somebody or combination of bodies has to do something to break that initial shell.

So, while talk of “alpha dogs” and “true number one options” can be tedious, there is probably something to the notion that having someone[1. To borrow a piece of soccer terminology] to take the game by the scruff of the neck is if not essential, extremely useful to having a quality offense. As with many situationally good things, this can get taken too far, and this sort of “heroball” often becomes a goal in and of itself rather than an instrument to pry open a defense to allow the application of more refined approaches. Identifying that line is all but impossible in a vacuum, and plenty difficult even knowing the full context of a situation.

The obvious example of this difficulty is that of Allen Iverson on his best Philadelphia teams. His 2001 MVP award is now seen as a symbol of an antiquated system of rewarding “Yay Points!” at the expense of things which actually contribute to victory. And to an extent, that argument is unassailable. Iverson finished tenth in win shares that season at 11.8, well below the 14.9 of the man many consider the “true MVP” from that season, Shaquille O’Neal. AI was a conscienceless gunner, leading the league in Usage Rate at just under 36% despite a True Shooting Percentage right at the league average for that season, and didn’t do much else to help the team aside leading the league in steals[4. Which for an otherwise offense challenged team turns out to be a darn useful thing!].

Looking more deeply, on that squad having that many shots even at league average was a benefit. Despite a roster with only one other marginally creative player to start the year[5. A 32-year old Toni Kukoc, who was traded to Atlanta as part of the Dikembe Mutombo trade.], and none by the conclusion, Philadelphia had an above average offense that season. This was accomplished in large part because Iverson took so much of the burden and allowed Aaron McKie to shoot open jumpers and everyone else to crash the offensive boards for putbacks. Take Iverson’s dynamism from that team, and they could easily have been the worst scoring team in modern NBA history. Did Iverson take some, a lot, way too many ill-advised shots? Almost certainly. Where the 76ers still better off for them? Probably yes as well.[10. And if this all sounds a bit like Russell Westbrook last year, that’s a decent comparison, though Westbrook had a much tougher conference a league as a whole to contend with, so no Finals appearance or MVP award for him.]

All of which is more to demonstrate there is no magic number at which “this is too much individual creation.” Teams need enough, but shouldn’t do too much of it.

That cutting bit of analysis leads to the real meat of what I’m after here: which player’s had the largest burdens of shot creation in 2014/15, and how did they perform under those conditions?

To answer that question, I took a similar approach to the one used on the team level, looking to plays ending with a given player in isolation, pick-and-roll ball handler or post up situations as defined by the public Synergy Sports stats on NBA.com. This remains an approximation of creation[2. It’s both under and overinclusive, due to the vagaries of play coding and the definitional squishiness of “creating offense.], but will work for a high level view. I defined each player’s “burden” as the proportion of available plays used in these situations, while the points per possession is an aggregate of the three categories[2. Where available. Since data isn’t displayed on NBA.com for players with under ten plays used in a given play type, there is no data included for things like Reggie Jackson post ups or Al Jefferson plays as a PNR ball handler.] Here’s a quick cross-section of all players with at least 300 created plays in 2014/15:

Aside from “James Harden Good. Rondo, LaVine, Burke and Carter-Williams, woof” it’s tough to draw too many instant conclusions from this chart. Worth remembering NBA average PPP on half-court created plays was .816, so the most active players in terms of burden (Westbrook, Dwyane Wade, Kobe Bryant and Reggie Jackson in Detroit) all produced right around league average in those situations. Looking into a little more detail demonstrates, again, how much Harden had to do and how superbly well he performed doing it:

Most created plays and fifth in PPP among players with at least 200 used? Pretty, pretty good. On the other hand, some of Manu Ginobili’s decline can seen in his high propensity for turnovers on created plays, highest among qualified players. Though Manu’s turnover rate wasn’t nearly as ugly as that of Dante Exum who turned the ball over on almost 37% of his created plays, but didn’t have quite enough attempts (147 total) to qualify for the overall analysis.