I do certainly think Scott Brooks plays a big role in the mismanagement of Westbrook and Durant, which is one of the reasons I think he should be fired. I don't think your point about Durant's regular season shot attempts with/without Westbrook tells the whole story though. After all, OKC's problem isn't in the regular season but rather in the playoffs, where ball movement ceases to exist (to the naked eye).
Let's look at 2011-12 (Westbrook healthy, playoff career high minutes at SG, trip to Finals), 2012-13 (no Westbrook) and 2013-14 (healthy Westbrook, MVP Durant, yet frustrating WCF loss). The reason I attempt to look at all three is because I think my perception of the Westbrook/Durant duo is being misinterpreted. I think Westbrook helps Durant as a player regardless of the position he plays. This is shown by the differential in the percentage of assisted baskets between Durant's 2012-13 playoff campaign and his 2013-14 campaign. Of his two-point FGM, 23.3% were assisted in 2012-13. In 2013-14, 53%. A similar differential can be seen in his percentage of three-pointers assisted. These stats suggest Westbrook takes pressure off Durant, creating more open/less contested shots for him. Westbrook's prowess as a scorer and an offensive threat are the same whether he plays PG or SG. When he's off the floor, Durant is forced into isolation.
My argument is more that Westbrook would help Durant even more so and be a more efficient player at SG. Now look at the 2011-12 playoffs. Westbrook played 33% of his minutes at SG, the highest percentage of his playoff career (and it's not even close; next highest percentage is 8% in 2013-14). During the playoff campaign, Westbrook shot career playoff highs in FG% and eFG% outside of his rookie year. Only his TS% was lower in 2011-12 than it was this year, which can be attributed to a far higher FT% this year (something that has little to do with him playing PG vs SG). Most importantly, Durant also found unique success with Westbrook playing more SG during the 2011-12 playoffs. His eFG% was 57%, 5.9% higher than his second career playoff high (which happened to be during this year's playoffs).
There's an argument to be made that the Thunder might just not be as good as they were in 2011-12 with Harden. Westbrook and Durant are, however, yet they played much less efficiently this year than they did in 2011-12. The argument I'm trying to make when I say "create more opportunities for Durant" has nothing to do with shot attempts. Durant already shoots about 20 times a game and there's only so many shots to go around. It has to do with efficiency and offensive fluidity, something that the Thunder's playoff offense lacks despite boasting three Olympians, one of which is the MVP.
On your second point -- two-PG lineups that prioritize P&R and the fast break -- how do you think Rondo, Westbrook and Durant would fare on the court if Durant played more of a Dirk role instead of a T-Mac role (meaning greater participation from him in the P&R instead of Rondo/Westbrook operating with McGary/Adams)? I'm asking you seriously because you seem to have impressive knowledge regarding offensive schemes and trends in the NBA.
Ah, I see what you're saying, and I have to agree with it. In regards with Rondo, Westbrook and Durant on the floor at the same time, I think that sort of depends on the rest of the roster.
My quick questions about that lineup in this hypothetical come regarding how the team deals with the fact that defenders are going to play off Westbrook and/or Rondo (whichever has the ball) in order to help in the paint against a pick and roll featuring Durant, and I don't think KD's strengths are really emphasized by being the roll man (although I'm not sure there's enough evidence to prove that one way or the other). Either way, while you're going to have an open Rondo or an open Westbrook regularly, that doesn't seem like that's not going to be something you can exploit consistently, since they're streaky outside shooters.
The problem, then, is sort of what plagues OKC right now -- they don't have consistent second and third options when they run into trouble after their first set. I'm not sure that changes all that much if you have Durant setting screens.
On paper, I'd probably look more towards something like the Princeton offense for a team with those guys at the 1,2, and 4, rather than using Durant to set screens. That would be a scary team to watch.
Interesting, that's really cool. I did some basic research on the Princeton offense and came up with this team:
http://espn.go.com/nba/tradeMachine?tradeId=qyjdlxqBOS trades: Rajon Rondo, Keith Bogans, Kelly Olynyk, Chris Babb, 2015 LAC 1st rounder
OKC trades: Serge Ibaka, Kendrick Perkins
http://espn.go.com/nba/tradeMachine?tradeId=kpdgpzlNYK trades: P. Prigioni, Tim Hardaway, Jr.
OKC trades: R. Jackson
Rondo/S. Christon/Telfair/P. Prigioni
Westbrook/J. Lamb/A. Morrow/[A. Roberson]
T. Hardaway, Jr./[Durant]/A. Roberson/Perry Jones III
Durant/McGary/N. Collison/G. Jerrett
S. Adams/K. Olynyk/H. Thabeet
That's 17 guys, so they'd need to find a way to trim down the roster. How do you think that would work? Hardaway is a hell of a shooter and stats suggest he's a strong finisher (68% FG% within 3 feet of the rim, which accounts for 20% of his total FGA). He's very good off the catch, especially from deep: from 20-29 feet, about 84% of Hardaway's makes (shooting about 36%) are assisted. He's also at least an average passer with solid court vision.