Yeah, Russell Westbrook should probably be MVP
Check out this incredible, astounding, extraordinary fact, via Deadspin's Albert Burneko:
Playing around on my own with some numbers available at NBA.com, I'm concluding with some amount of confidence that over 50 percent of Thunder possessions culminate within one touch of Russ when he's on the foor. (You can scroll down to check that minutiae).
That's absurd. That kind of load-bearing is pretty much unprecedented.
I don't like to dip into the old "if X team didn't have Y player" line of argument, because it inevitably hinges on subjective interpretations, mostly to favor whoever it is who plays with the worse teammates (if you take away LeBron the Cavs still have Kyrie and Love, but if you take away James Harden the Rockets lose so much more, is usually something like how it goes).
Every team's best player is inordinately important, and no team can just lose a star (except maybe the Warriors). But in this case, Russell Westbrook is, actually, measurably and not insignificantly more inordinate than his peers.
Without him, the Thunder would not only be much, much worse, they would basically be unrecognizable. They would have to replace the influencing factor in something like half their possessions. That is a remarkably unique feature among individual NBA seasons, historically, and certainly counts for something in this MVP debate.
Of course, it wouldn't mean so much if his numbers weren't so impressive. And it is just as true that his numbers wouldn't be so impressive if he weren't required to do this much. There's a bit of chicken and egg to it.
Averaging over 30 points and a triple-double per game is stupendous, but if you believe in advanced metrics, you can't just hand the award to Russ because he fills up the stat sheet.
Fortunately, these days there are a number of very good advanced catch-all metrics that are publicly available for us. And by ESPN's real plus-minus wins he edges second-place LeBron considerably*1. He leads the league by BasketballReference's box plus-minus (plus-14.3). He's also in a distinct 1-2 with LeBron in terms of net ratings, versus a far-trailing Harden and Kawhi Leonard. He does rank fourth in ESPN's real plus-minus (plus-6.35) and, notably, is 10th by BasketballReference's win shares (9.4). But overall, his advanced profile is undeniably MVP-worthy.
And let's talk about his game, aesthetically. Westbrook is ferocious, as determined and indomitable a basketball player as there is alive. The moment in which he decides to score, bursts toward the rim, and then puts the ball in the basket is as exhilarating a thing as there is in the game. His dominance is one-of-a-kind in today's NBA.
The actual physical manifestation of his usage rate is an awesome thing to behold, and we should appreciate the unparalleled, historic enormity of his play for Oklahoma City.
In the end, the fact is in basketball "value" is a mix of efficiency and volume. Russell Westbrook is not and will probably never be the most efficient player in the league. Before the season I was dismissive of his MVP case because I did not think he could reach the efficient levels of a player like LeBron, or Harden or Steph Curry or Kevin Durant. It's just not his game.
But what I didn't anticipate is how outrageously Westbrook was capable of swelling his volume. That he has done that, and not sacrificed anymore of his good-not-great efficiency, is a thing to marvel at.
He cannot be ignored as the best player in the NBA this season, and thus, at least for now, the rightful MVP winner.
*1: 13.76, to James' 13.45, a .31 difference which is almost three times as significant as the difference between LeBron and third-placed Jimmy Butler (.11).
*2: Via @HPbasketball. Westbrook Thunder +2.7 On, -10.9 Off, +13.6; LeBron +7.9 On, -7.2 Off, +15.1; Harden +6.7 On, +5.2 Off, +1.5; Leonard +9.3 On, +8.2 Off, +1.1.
Playing around withe usage concepts
Now, because of the nature of and differences in usage rate and assist percentage*1, we can't really just add the two. But to get an idea of how many Thunder possessions end no more than one touch removed from Russ, we can attempt to get a bit granular:
• The Thunder have taken 4,205 field goal attempts while Westbrook is on the floor.*2 He has taken 1,527 field goal attempts this season.*3
• The Thunder made 977 assists with Westbrook on the floor this season. He has 633.
• The Thunder have turned the ball over 683 times with Westbrook on the floor. He has 345.
• The Thunder have taken 1,265 free-throw attempts with Westbrook on the floor. Because three-point shooting fouls are pretty rare, I'm just going to divide this number by two and call it close enough.*4 So that's 632.5 free-throw trips with him on the court, and he has 679 free throw attempts, divided by two, for 339.5 trips.
• Here is where I'm gonna go a little more out on the limb. NBA.com also tracks "potential assists", which are passes that result in shots, generally.*5 Westbrook has 1,200 of those, and if you subtract his assists on made shots from that total, you're left with another 567 possessions, missed shots in this case, that culminated one touch from Russ.
• So the percentage of Thunder shot attempts, assists, would-be assists, turnovers and free throw trips (kinda) for Westbrook comes to (3411.5 Westbrook events divided by 6497.5 possession outcomes) 52.5 percent.
By this rough accounting, 52.5 percent of all Thunder possessions with Westbrook on the court end with him taking a shot, assisting a made bucket, assisting a missed shot, a trip to the free throw line or a turnover. This doesn't even account for Russ passes that result in the next guy committing a turnover or earning a free-throw trip, so on some level it's understating things.
This is all rough, but I think it's safe to say over 50 percent of Thunder possessions culminate within one touch of Russell Westbrook when he's on the floor. That's a dumbfounding figure.
*1: Most obviously, they're percentages of different things. For a deeper look at usage rate's idiosyncrasies, see here.
*2: Totals here.
*3: See here.
*4: In theory, you could go through the play-by-play records of every Thunder game and get an exact number for that figure. I ain't doing that.
*5: See here.