OKLAHOMA CITY -- On the night defense made an unexpected appearance in the Western Conference finals, the Thunder didn't play enough of it at the start. A 23-point deficit proved too much to overcome in Oklahoma City's 93-87 Game 3 loss to Dallas, for which we can blame in equal measure bad shooting and bad lineups.
More on that in a minute, but let's spend a brief moment pondering the positives. This was a huge missed opportunity for the Thunder, because two things that hadn't gone right for them in the first two games did in a big way Saturday night.
First, the defense of Serge Ibaka and Nick Collison hounded Dirk Nowitzki into one of his worst games in ages; he shot just 7-of-21 with seven turnovers. To give you an idea of the rarity of this event, Nowitzki hadn't missed two-thirds of his shots in a game since April 2, a span of 18 games, and hadn't had a seven-turnover game since the second week of the season. This was a precious gift from the basketball gods, and the Thunder smashed it like Humpty Dumpty.
Second, Russell Westbrook rebounded from his Game 2 fourth-quarter benching by carrying the Thunder offense with a game-high 30 points. The biggest story heading into the night turned into a total nonstory by the end of it, as Westbrook once again showcased the resiliency and hard-headedness that are both his best and worst traits as a player.
Those two major positives should have translated into a win, and the fact that it didn't may very well cost the Thunder the series -- Dallas has now regained home-court advantage, and Game 3 winners in a tied series prevail in 76 percent of them.
For that, we return to the bad shooting and the lineup issues.
First, the Thunder couldn't throw it in the ocean. They shot 36.5 percent overall and 1-of-17 on 3s; insert your own Bricktown joke here. Oklahoma City missed its first 16 3-point attempts, and the odds of that happening to a team that shot 34.7 percent in the regular season are about 1-in-1,000. (OK, 1 in 914.9, to be exact.)
Some of them were poor looks but several were wide open; particularly hurtful were two wide-open fourth-quarter 3s by the normally deadly Daequan Cook that didn't find the mark, sucking the life out of the Thunder's comeback bid. A wild 3 missed by Westbrook -- his worst decision of the night -- with just under three minutes left wasn't helpful, either. And Durant, unusually, missed all eight of his 3-point tries.
Of course, there was another team on the court that had something to do with this. The Mavs defended better and Shawn Marion in particular did a fantastic job against Durant, who shot only 7-of-22 overall and, once again, had trouble getting the ball at times against a defense focused on denying his catches. Oklahoma City ran its first two plays of the game for Durant and he didn't get a finger on the ball either time.
Another, underrated story was the defense played Jason Terry on James Harden, one night after Harden lit him up for 23 points off the bench. Harden was just 2-for-9 and had only seven points in 35 minutes; without his scoring to supplement the two All-Stars, Durant and Westbrook finished as the only Thunder players in double figures.
That was enough material for a loss right there; if Harden and Durant combine to shoot 9-for-31 with no 3s, the Thunder probably aren't winning. The fact they defended Nowitzki so unusually well helped keep things close, though.
And that's where the other issue crops up, and it's one that bears much closer watching heading into Game 4: lineups.
The Thunder's starting lineup once again dug a huge hole for them, falling behind by double-digits early. It was the third straight game this series that Oklahoma City trailed with its starting five on the floor.
In the three games, Oklahoma City has been outscored 39-17 in the opening minutes of the first quarter until a reserve checks into the game. Overall, the deficit to begin each half is 57-31. Take those minutes away and the Thunder are comfortably outscoring Dallas.
To put a finer point on things, let's replace the word "starters" with "Kendrick Perkins." In his 82 minutes on the court, the Thunder have been outscored by a whopping 32 points. In his 62 minutes on the bench, the advantage has titled nearly as strongly the other way: The Thunder are plus-23.
And those slow starts? They've magically ended at the exact second Perkins departed. Oklahoma City trailed by five in Game 1 before he went out, by nine in Game 2, and by 15 in this one -- 29 points worth of deficits to make up the rest of the night. The Thunder overcame it in Game 2 by scoring on nine straight possessions right after Perk went to the bench; in the other two games, the hill was too big to climb.
Of course, this could just be noise. Sometimes plus-minus fluctuates randomly for reasons that have little to do with a player's performance. After watching the three games, however, this isn't one of those cases. For starters, Perkins has made a limited statistical contribution, with only 11 points and 14 rebounds in the three games. His main function seemed to be going out of his way to run into Tyson Chandler off the ball.
This isn't much different from what Perkins did in the Memphis series, but there's a key difference. Against Memphis, he had a function -- he could bang in the post against big, burly Marc Gasol, the type of assignment for which Perk was born.
He's too slow to stay out there unless Brendan Haywood is playing for the Mavs. Chandler is running circles around him off the ball, constantly forcing him to be a half-step out of position to help when guards enter the lane or Marion dives in off the ball. And it's shown in the results -- the Mavs carve them up with Perk on the floor.
I don't want to diminish Perkins' value in general. The Thunder people will tell you in gushing detail about his crunching screens, and what a smart, communicating defender he is, and how his physicality sets a tone. Against the right opponent, this can be enormously valuable just ask Orlando.
The problem is that the Mavs aren't that opponent, and Perkins played 30 minutes anyway. That was the most of any Oklahoma City big man. Collison and Ibaka played only 24 minutes, and Nazr Mohammed didn't play at all.
One can also argue that Perkins' weaknesses are magnified because he plays with another non-scorer, Thabo Sefolosha. Like Perkins, Sefolosha's strengths are wasted in this series because Dallas lacks a great wing scorer. Meanwhile, it's too easy for opponents to load up on Durant and Westbrook at the start of games, which is exactly what's happened in each of the first three contests until Collison and Harden come in.
All of which creates a thorny issue for the Thunder. Their best team going forward has Perkins at center, especially a year from now when his surgically repaired knee is in better shape. I still believe that.
But their best chance of winning this series is with Perkins playing a much more limited role. Thus, they have to risk upsetting their centerpiece big man -- much as Scott Brooks risked upsetting Westbrook by benching him at the end of Game 2 -- and deploying small lineups and Ibaka-Collison combinations much more readily. If they aren't willing to yank Perkins from the starting lineup entirely (they could even start Mohammed to avoid ruffling the bench rotation), at the very least they need to cut his 30 minutes from Saturday to a few scraps at the start of each half.
Again, I don't want to pin all this on Perk; he wasn't the one shooting all those bricks. But the cold shooting was a one-game phenomenon; the Perkins effect has been all series. Until or unless it changes, the Thunder's playoff run is unlikely to continue much longer.