James Harden is a superstar

[lorak]

Harden's on/off net is -2.1 this season. Both offense and defense are better with him on the bench. Any explanation?

[Krodis]

Harden's on/off net is -2.1 this season. Both offense and defense are better with him on the bench. Any explanation?

The highest minute not-Harden lineup, Douglas-Delfino-Parsons-Morris-Smith, inexplicably has a ORTG of 151, and mostly because of offensive rebounds and extreme three point shooting. I'd say it's just a fluke. Only 53.5 minutes.

[SideshowBob]

What I'm really looking forward to is how he takes the next step in 2014. We've seen most of the superstar wings make a jump around the age of 24 (Wade in 06, Kobe in 03, James in 09, Durant in 13, etc.) and then take another step forward around 27/28 and hit their extended peaks (Wade 09-11, Kobe 06-09, James 13-??).

My first thought is that I don't see clear separation between Harden at 23 and any of those other guys at 23 on the offensive end (in fact I think he's arguably a little better). The net offensive impact of the combination of skillset and athleticism is about the same for all of these guys in my book, though obviously you could make nuanced arguments for ranking one above the other (James dwarfs the rest in terms of production and is able to give you more minutes; Bryant and Durant's games were already far more refined; Wade's game already translated better to playoff defenses than the others), but they're all in the same realm (Harden included).

So, based on trends, I fully expect Harden to take the leap. Given a year to understand and build experience in a new role as well as develop chemistry, an offseason to continue to develop his game (jumpshooting tends to improve over time, I expect him to work on his mid-range game, he's going to be less turnover prone as he builds chemistry), playing experience usually helps develop game-management skills (an underrated aspect of lead wings' games), Rockets will probably look to improve spacing, etc. etc. The question is, how much can he be expected to improve? And where can he be expected to land, relative to those guys, at the age of 24?

[lorak]

Harden's on/off net is -2.1 this season. Both offense and defense are better with him on the bench. Any explanation?

The highest minute not-Harden lineup, Douglas-Delfino-Parsons-Morris-Smith, inexplicably has a ORTG of 151, and mostly because of offensive rebounds and extreme three point shooting. I'd say it's just a fluke. Only 53.5 minutes.

One lineup with Harden played similar amount of minutes and has similar ORTG and ORB%.
Besides overall Rockets played 633 minutes without Harden and they don't have any decent guard from the bench. So it's really strange.

[colts18]

Harden's on/off net is -2.1 this season. Both offense and defense are better with him on the bench. Any explanation?

This explains it. Garbage time.

The only explanation I can get to is that Houston’s reserves have been extraordinarily good in garbage time. Whether they’re up by enough or down by enough to play all-reserve lineups for the entire 4th quarter the Rockets have been very good in those situations. Using a tool I’ve developed to look at specific lineup combinations it appears that my hypothesis is correct. Assuming that we can define garbage time for the Rockets as any minutes that did not include either James Harden or Chandler Parsons (those 2 play by far the most minutes and Delfino is the only legitimate backup to either player) the Rockets performance has been very good.

The way to read this graphic is that the Rockets have played 214.1 minutes with the combo of Harden and Parsons both off the court (those 2 being off court is the criteria for on-court in the tool). During those 214.1 minutes they’ve outscored opponents by 56 points, which, if extrapolated by a per-minute basis out to a 48 minute game the Rockets would outscore opponents by 12.55 points per game. With either Harden or Parsons on the court the Rockets outscore opponents by 1.97 points per game so the net difference would be 10.59.

I’m encouraged by NBA.com’s release of all of their historical statistical information to possibly develop better metrics because if the metrics we’re currently using to evaluate players can be skewed this heavily by garbage time stats then there’s a lot of room for improvement.

http://www.spreadsheet-sports.com/the-c ... n-the-nba/

[lorak]

^
I don't like definition of garbage time as time without Harden and Parsons. It should be something like last X minutes of the game, when team is leading 15 (or 20) pts.

Besides, any other examples of players with superstar status, whose on/off net impact was around -2?

[Krodis]

I mean, 633 minutes without Harden is a pretty small sample size. And the Rockets have the 4th best ORTG overall, so it's not like they're performing poorly with Harden on the court. I somehow don't think the Rockets sans Harden are a 110 ORTG team, so my reflex is to say it's just fluky small sample size data. I mean, of the other real decent offensive players on the team, Lin and Parsons' offensive on/off numbers are worse than Harden's.

The only alternative explanation to it being a small sample size fluke is that Carlos Delfino is carrying the team >.>

[ardee]

What I'm really looking forward to is how he takes the next step in 2014. We've seen most of the superstar wings make a jump around the age of 24 (Wade in 06, Kobe in 03, James in 09, Durant in 13, etc.) and then take another step forward around 27/28 and hit their extended peaks (Wade 09-11, Kobe 06-09, James 13-??).

My first thought is that I don't see clear separation between Harden at 23 and any of those other guys at 23 on the offensive end (in fact I think he's arguably a little better). The net offensive impact of the combination of skillset and athleticism is about the same for all of these guys in my book, though obviously you could make nuanced arguments for ranking one above the other (James dwarfs the rest in terms of production and is able to give you more minutes; Bryant and Durant's games were already far more refined; Wade's game already translated better to playoff defenses than the others), but they're all in the same realm (Harden included).

So, based on trends, I fully expect Harden to take the leap. Given a year to understand and build experience in a new role as well as develop chemistry, an offseason to continue to develop his game (jumpshooting tends to improve over time, I expect him to work on his mid-range game, he's going to be less turnover prone as he builds chemistry), playing experience usually helps develop game-management skills (an underrated aspect of lead wings' games), Rockets will probably look to improve spacing, etc. etc. The question is, how much can he be expected to improve? And where can he be expected to land, relative to those guys, at the age of 24?

It depends on his team-mates. If Lin can play to his potential like he did last night, keeping defenses honest, Harden will have even more space to work with.

So, I think he'll improve his mid-range game. If he does, he'll be an assassin from 3 all the way to the rim, which means some very good things could come. I would like to think he can cut down on his turnovers, but his shot creation is so vital for Houston and unless they sign another star this offseason (which we can't predict), he's going to continue to hold the ball as much as he does. 3.4-3.6 will be acceptable for now.

Statistically, something like 27.5-28 ppg, 5.5 rpg, 6-6.5 apg on 47-48% FG, 37-39% 3P, and 62-63% TS should be in play if Parsons and Lin improve as well. This guy's final peak, i.e., the age 27-28 one you mentioned, will be right up there with Wade, Kobe and T-Mac's.

[lorak]

I mean, 633 minutes without Harden is a pretty small sample size.

Almost every superstar's off minutes this seasons are in 500-700 range, but yet their on/off net impact is far greater than Harden's. But only in Harden's case some strange fluke screw results? I don't buy it.

[Krodis]

But it's not like the Rockets' performance with Harden ON the court is bad, especially offensively. It's just that the Rockets' numbers with Harden OFF the court are higher than most teams. Now, do you really think the Rockets are a Top 3 or 4 offensive team in the NBA playing their bench unit? Because that's the explanation you seem to be coming to if you're taking the raw on/off numbers at face value.

[SweetTouch]

Correct me if I'm wrong, but didn't Durant have the same problem regarding ON and OFF stats early in his career

[Krodis]

Durant's problems were worse because they were much worse with him on the floor and the team stunk to begin with.

Harden has had really good +/- numbers before this year, too.

[SideshowBob]

Durant was a net -0.6 ON/OFF in 2011, and Bryant was a net -2.0 in 2002

[Krodis]

Yao Ming had bad on/off numbers in 04-05. They did a complete 180 the next year.

Sometimes it just is flukey. Especially raw on/off, especially with sample sizes of a season or smaller.

[lorak]

But Ming's net on/off is explained by Mutombo who was coming off the bench, Durant's by Harden, and Kobe's by Shaq injury. But what is explanation of this year Harden's on/off net?

[SideshowBob]

I think it's just an adjusting to a new team thing, and aside from that, we'd have to take a more in-depth look at the kind of lineups Houston's putting out there.

He's in a completely different role right now and was pretty much thrown into it with just a few days before the start of the season. Let's take a look at his ON court +/- per 100 possessions month by month (ignoring the one game sample from October).

Nov: -0.9 in 538.0 MP
Dec: +4.0 in 561.7 MP
Jan: -1.3 in 654.1 MP
Feb: +11.4 in 272.0 MP

So we're seeing his on-court numbers jump around quite a bit, and I'd wager we'd see the same with the OFF court numbers (I would've posted those, but it's going to take a little while to calculate). I honestly think adjustment may be a major issue. I would take another look at the numbers at the end of the season (preferably in splits, to gauge how he changed over the season), and then I'd expect them to jump next season once he's had more time to fit in here.

(cue mysticbb coming in here and effortlessly providing a better explanation)

[fatal9]

Rockets have a lot of three point shooting off their bench. Their bench actually has almost as many 3PM as the starters. Their bench overall is top 5 in the NBA imo so not surprising they would have a good off rating without Harden. They don't have a 6th man type anchoring the bench (maybe Delfino, he's been really good) but they have really good depth and 3 pt shooting, and play in an uptempo system with great spacing where they are always going to get good looks on the perimeter. Plus, those guys could also just be getting hot during garbage time when defense isn't that good, not hard to picture considering they are one of the best (maybe the best?) 3pt shooting bench in the league.

Wasn't Chris Paul a net negative in on/off (or not very impressive) earlier in the year before he missed games and they began losing? I remember people using that as an argument against him for MVP but it was more due to the bench being good than Paul not making MVP level impact. Take out Paul and the system falls apart, players have to do more than they are capable of, roles aren't optimal and team loses.

I don't touch raw +/-, almost useless to me, too flawed, you have to control for different lineups for +/- to be meaningful.

edit: SSB's post above shows how erratic his ON rating (corrected) has been on a monthly basis, something you can expect with a team that depends heavily on 3 pt shooting. Their starters also haven't been consistent over the year either, Lin looked bad early in the year, Parsons has been inconsistent after a good start, Harden has also been up and down with his 3 pt shooting. Seems like they are finding a groove lately though.

edit #2: Just checked league wide bench numbers. Rockets bench is #2 in 3PA and 3PM (Knicks are #1). In terms of percentage of overall bench field goal attempts that are threes, they lead the league. So Rockets bench depends more on 3pt shooting than any other bench when you consider both volume and the % of bench offense that results in shooting a three. I just think in this situation you're going to get really funky and erratic +/- numbers. 4 of the 5 Rocket starters are net negatives in terms of on/off.

[SideshowBob]

Note that what I posted only included the ON part of ON/OFF.

[lorak]

I don't touch raw +/-, almost useless to me, too flawed, you have to control for different lineups for +/- to be meaningful.

The same applies to with/without, yet it was big part of RPOY and peak projects. Should we redo them without using that kind of data? ;]

Anyway, we all know (or should know...) +/- limitations, but most odd result could be explained (like Ming by Mutombo, Kobe by Shaq and KD by Harden) and thus our analysis based on this kind of data is valuable. But in case of Harden 2013 it's more difficult to find explanation. Bench 3P is probably good one, but what it says us about Harden? For example isn't he (his numbers) overrated because he plays in very pro offensive environment? So good for offense that even bench keeps very high level of offensive production.

[Doctor MJ]

Harden's on/off net is -2.1 this season. Both offense and defense are better with him on the bench. Any explanation?

It's definitely a good question to ask. Not that we should jump on the on/off quickly if there's no explanation, but it definitely warrants analysis. Wish we hadn't suffered a loss of APM data.

I tread very cautiously here in particular because Harden had such good numbers in OKC. Saying anything along the lines that Harden gets in his teammates way seems at the very least hyperbole. That said I'm certainly not going to claim Harden's carrying a bunch of no-talent hack.

It will be very interesting how the team evolves next year.