This season, Derrick Rose has not been his old MVP self. This is not a revelation, I'm not telling anyone noteworthy information by making that observation. Rose has shown flashes of brilliance here and there, but mostly he's been okay to pretty bad depending on the night. His overall numbers and efficiency have all trended down over the course of the year. I'm still not quite ready to panic regarding Rose's struggles, though. Some of this is based on positive things Rose has been able to do and part of it might be wishful extrapolation from a select few (some would say cherry-picked) numbers. Let's get into it.
A few weeks ago, I decided to try to build a better version of John Hollinger's Game Score to evaluate individual performance. Basically, I created a simple linear weights metric that estimates +/- impact per game, or per minute if you prefer. (I wrote about it over at my website. I won't get into the details here, if you want them, they are here.) One of the neat things you can do with this metric, which I call Daily RAPM Estimate, or DRE, is examine game by game performance or look at performance trends over different time splits. I did this with Rose's game log and his month by month splits.
Here is what Rose's game to game estimated impact on the margin of victory or defeat looks like through the first 50 games (games in which he was inactive come in as 0s):
There are quite a few very negative games in there and many fewer big positive games to balance them out. Overall, Rose's estimated +/- impact per game, by this measure, is -2.65. That is obviously quite bad. (For context, the league worst in DRE per game, as of February 2nd, was Lance Stephenson at -3.23.) That game by game chart is pretty up and down, and it's tough to see any real trend other than ups and downs, so let's take a look at the month to month splits to see how Rose is trending:
Yikes. First, caveats. There were only 2 games in October and there's been just one in February at the time I am writing this. So the numbers for those months are very small samples. Still, even if we look just at November, December, and January, the numbers are on a clear negative trend. Some of this is Rose playing more minutes and therefore, his negative estimate impact increases with those increased minutes, but even if you look at his DRE per minute, the trend is the same.
Now, DRE is a linear weights metric and a very simple one at that. There's no adjustment for usage or anything that would really give the benefit of the doubt to a player, like Rose, who draws a lot of defensive attention and takes a lot of shots. It should come as no surprise, then, that Rose's Real Plus Minus is -.9 per 100 possessions, which given his 30.8 minutes played per game and the Bulls' pace, would make Rose's Real Plus Minus impact per game roughly -.54. So he's been significantly better than this simple metric would suggest, and really, that should come as no surprise. Real Plus Minus is much more sophisticated than DRE. Regardless, Rose's production has been sliding since the start of the year and that's a bit worrying.
On the other hand, diving into some of the SportVU numbers, Rose looks like he might be close to back physically, at least in terms of his driving ability. Amongst players with 25 games played, Rose is 5th in points per 48 minutes scored on drives. Amongst those in that top 5 group, only LeBron has a better FG% on drives. Rose is shooting better on drives than James Harden, Jeff Teague, Monta Ellis, much better than Russell Westbrook, and roughly even with Damian Lillard.
After some tweaking the raw data in Excel, Rose's productivity on the drive this year is even more pronounced. Amongst players with at least 200 drives, Rose's 1.26 points scored by his team per drive is 7th in the league.
It's interesting to note that Jimmy is 9th, in a virtual dead heat with Rose and the rest of the best producing drivers in the league. So with two of the top driving guards in the league, why aren't the Bulls better?
Well, amongst the 58 players in the league with at least 200 drives, Rose is 29th in drives per 48 minutes (11.3), and Jimmy is second to last (6.3). The Bulls' starting guards, despite being super effective when they drive, are simply not doing it that much. It's unlikely this is a system thing, as Aaron Brooks is 7th on the list, driving a whopping 15.5 times per 48 minutes. Brooks' drives are much less effective, producing 1.08 points per drive for the Bulls, ranking him 48th out of 58 qualifying players.
What's the big difference with Brooks vs. Rose and Butler? Brooks gets to play with a spaced out floor much more often, so he has driving lanes available to him. I don't want to turn this into another Pau bashfest, but it might be nice for the Bulls' starting lineup to get some of that spacing stuff, to open up lanes for Jimmy and Derrick, considering how good they already are at driving, and that's just not going to happen with a Noah-Pau frontcourt and Derrick & Jimmy having relatively suspect jumpers.
Some of Derrick's failure to drive as much, to my eyes, is his settling for 3 point shots before even trying to work inside for a better shot. Generally speaking, that's a bad habit for him that you might want to discourage. On the other hand, giving Rose the biggest benefit of the doubt possible, he might be coasting through this whole regular season knowing that the playoffs are what matter. Getting through this season healthy, without taking too many unnecessary shots to his body, is probably his priority. Settling for a few extra jumpers at the expense of additional drives to the basket helps accomplish that. Given how packed the paint has been for him for large parts of this season, it's hard to blame him.
In addition, Rose had his best 3 point shooting month in January. His 34.4% was a pretty big uptick over his previous shooting and if he is able to be a little more discerning about the shots he takes, he might be able to get that number up to around 35-36% come playoff time. Sure, that's probably being really optimistic, but even if Rose can just hold his percentage close to his January "hot" spell, that'd be fantastic news for the Bulls. This is a guy who won an MVP shooting just 33% on 4.8 3s a game. He doesn't need to be Stephen Curry, he just needs his jumper to be a threat.
Despite his relatively crummy efficiency, Rose is still one of the game's most dangerous forces driving to the basket. In the back of his head, Rose has to know this. Maybe that's where his seemingly delusional confidence this season comes from. Come playoff time, I expect (hope?) to see him drive with greater frequency and his love affair with 3 pointers to wane. If that happens, the Bulls' ceiling becomes much higher than their recent dismal spell would indicate.