Graphic Offense: A Look At Usage and PER Mid-Season

What does an offense looks like? I can give you some ideas from watching the teams. Houston moves really fast and really consistently and never underperforms or overperforms. Cleveland can slow to a crawl or detonate underneath a powder keg. The Lakers are largely impacted by Kobe, it feels like Gasol doesn’t get enough touches and Odom has been negatively impacted by Artest’s arrival, though overall they’re the biggest juggernaut in the league. Chicago’s a damn trainwreck.

But how do they look through numbers? Last year in the playoffs I broke down how San Antonio looked in the regular season and post-season with a Parker-heavy attack. I decided to take a similar look at all the teams and where they are as of right now, just past the halfway mark of the season, using data from HoopData.com.

(NOTE: Feel free to hotlink any of these images or still them for yourself for comparison. The Flickr galleries can be found here and here. Please throw me a link, of course, because, you know, I’ll find you.)

The first thing we’ll examine is just the usage of all 30 teams. As a primer, here’s the definition of usage:

(FGA + (FTA * .44) + TO * 40 * League Pace)/(Mins/Team Pace)

which is slightly different from Basketball-Reference’s formula:

100 * ((FGA + 0.44 * FTA + TOV) * (Tm MP / 5)) / (MP * (Tm FGA + 0.44 * Tm FTA + Tm TOV))

So now that we’ve got that cleared up, here’s some things to keep in mind.

1. All data is sorted clockedwise by Usage. Before, I’ve tried using a traditional positional model (PG-SG-SF-PF-C, repeat) to try and get a visual look at where team’s emphasis is. But with bench play so varied between teams, it makes it impossible to get much from it, and that’s before all the positional issues that come with bench rotations.I went with this model so that there would be a more standard shape to the data that you could compare (Houston’s fatter than Miami, for example), and to make the data more apparent.

2. This data isn’t filtered for a minutes requirement, nor for current players that have since been jettisoned. This is why you’ll see players with almost zero playing time with high usages. Keep that in mind when comparing things. Mike James can use up a ton of possessions because he’s only on the floor for about thirty seconds. Then again, in some instances, you can say that maybe one reason a player doesn’t see floor time is the number of possessions he absorbs.

3. I have to say this every time, and I will say it again. Usage is not a value stat. If you shoot a ton but make them at a ridiculous rate, you’re not considered a bad player. The best player on a team has the highest usage a very high percentage of the time. The next set of charts will take a little different look at what they do with the ball (but even then there are limitations to the comparison.

Here are the usage charts.

The Hawks are almost exactly what you’d predict them to be. You’ll notice the extremely high level of the two volume scorers, and the low percentage of Josh Smith and Al Horford, Horford in particular. Mo Evans seems to be keeping his muzzle on, and you have to admire Bibby in particular keeping himself out of the trigger-spot. This is going to be more fuel for the anti-Joe crowd, in ATL and elsewhere.

There are instances where you want more of a “fuller” or “fatter”distribution. Then there are times when you want to skinny-up the offense. The Celtics probably need to skinny it up a bit. Notice #3, there? Sheed. And House. With higher usages than Ray Allen. That’s pretty damning, as is Kendrick Perkins, even with his limited offensive set, not getting many opportunities with as high a FG% as he has. But still, that’s a pretty healthy distribution chart, with two of the best players as the highest.

Woah, there Flip. Simmer down a bit. You’ve been pretty good. Give it a choke-back. Just a bit. Just to see how it feels. Felton’s really low on this, and Diaw being all the way at the tail of the curl is pretty insane. Given Acie’s situation, that’s a heck of a lot of usage, too, but, well, it’s the Cats. People can rag on Diop all they want, but his usage is right where it should be. Knowing your role is important, and Larry Brown’s teams get that.

The fact that Noah’s such an impact player despite not needing the ball is pretty impressive. Then again, you have to wonder how good he could be if he wasn’t a liability with the ball in his hands to produce. Two of the Bulls’ starters are on the back side of the halfway mark. That’s a bad sign. You need guys who are willing to take shots in the starting spot. Of course, Tyrus Thomas is willing to, and that’s not a good thing. The steep decline on this, with the players involved means basically Rose is the only one with an adequate distribution. Wait till you see their PER overlayed.

There is absolutely NOTHING surprising about this chart. Every single player is pretty much where you’d expect them to be, outside of Anthony Parker, who will probably rocket up over the next four weeks. Shaq’s not shy, Mo’s second fiddle-ish, Z’s still in the same spot, and Jacks0n-Hickson-Williams are in high-usage situations. And LEBRON, LEBRON, LEBRON, all day long.

That’s a healthy team, right there. Superstar getting buckets, two primary supporting members with high percentages, support team, and the aging point guard keeping his shots to himself.  For the life of me, I can’t figure out what makes Drew Gooden consistently push his way to the front half. He did this with SAS last year. Can’t figure out why, though. Barea being the halfway mark is pretty spot-on for this team. Gotta be perplexed by Beaubois here. Rookies should not be outshooting Jason Terry on a veteran-heavy-rotation team. Some shots? Good. That many? Eeeee.

From the department of “things you don’t need to be told by this chart”, I give you J.R. Smith.  This is a remarkably shallow distribution for what is considered the second best team in the West. It does follow the same pattern as some of the best teams, though. Billups is pretty much the difference maker.

Tayshaun’s injury certainly plays a part in this, and otherwise, Bynum and Gordon are in a dog fight for who can dish up the most attempts. It’s a really shallow distribution, which fits their record. Villanueva’s is probably pretty high, but hard to argue with the money he was given.

Anthony Morrow is one of the three best shooters on this team, and he is ninth in usage. That’s a pretty good starting spot for what’s wrong with this team. Biedrins injury certainly jacks with this. I feel like Devean George should still be behind Raja Bell, even though Bell’s only played one game. But that’s just me.

Holy MOLY, that’s a fatty fat fat distribution. Take a second and LOL at Tracy using his limited minutes to work his way into the halfway mark. Or Pops Menash Bonsu. Everything else matches up with what you’d expect. And if Trevor Ariza is a black hole, then Shane Battier and Chuck Hayes are apparently the vacant vastness of space. You can pass through it, but there’s not anything there.

I’m loathe to care at all about this team’s distribution, but why not. Prices confidence is admirable, and he’s playing well. Granger should be the top, Tyler should probably not be that high, but who else, really? Uh… free Jeff Foster?

I can’t argue with this at all. This is pretty much perfect. Kaman, Baron, Rhino, Gordon, Thornton, Telfair, Ricky, Butler? You could potentially say Gordon should be higher, but then, I’ve never been sold on him as a number one option. The elephant in the room is Griffin, natch.

WEIRD. I mean, apparently, you DON”T need a nice even distribution to be a great team. I’d never argue that you need everyone taking the same amount, you clearly need one guy to be the alpha male. But the percentages here are really stunning. Kobe’s just so far above everyone else, it’s unfathomable. Notice how every other team is in the shape of a nautilus? This is more like a teardrop. How is Gasol so low? The PER’s really make this thing bizarre. Then again, you realize that they’re the best team in the league, and you know, whatever they want to do is cool. How is Odom so low on this? And this should put to rest some of the talk about Artest’s shot selection this year. He’s kept it very much under-wraps compared to the others. And finally, chill out, Jordan Farmar.

THE ANSWER MAY CAUSE LULZ. Gasol needs to be much higher on this, but he draws double-teams (which he doesn’t handle well) a lot of the time. I’d expect Conley to work his way up now that he’s found his shot. A shallow but strangely even distribution for a team with such a horrible bench. I guess they don’t realize how bad they are.

How Kobe of Wade. There’s really no one else on this team. Your probable sixth seed in the East, ladies and gentlemen!

Shows you the loss of Redd and what it means for the offense. I like the 4-6th distribution here. They need a volume shooter off the bench that’s not Ridnour.

So there’s that.

CDR is much lower than I would have expected. As is Courtney Lee, but that’s probably good with his fall-off. There’s no reason for Brook Lopez not to be the top of this, and that likely has a lot to do with their record. Check out TWill with no conscience. Other than that, you’re looking at no one wanting to take shots. The Nets would be better off with a few guys having volume-shooter approaches.

Problems that occur when your best player is a pure point guard. Bobby Brown ends up as the highest Usage guy. How this team is still in the playoff hunt is mind-boggling.

Maybe no chart is more interesting than the Knicks. Eddy Curry, Toney Douglas, Jordan Hill, Larry Hughes. Four players in the top eight are guys that have had problems with playing time for the Knicks. Is part of this an issue with players taking advantage of the D’Antoni system, and actually shooting too much? Al Harrington is the only player in the top eight who belongs where he is. Chandler’s likely too low, given his position. And Duhon, well… yeah.

I miss Kyle Weaver. Westbrook’s slight deviation from the curve is pretty indicative of the continued question of if he injects himself into the offense too much. Eric Maynor with a solid spot for a backup guard. Weird to see Thabo, Collison, and Ibaka on that end, this is such a perimeter-heavy offense.

Vince Carter sucks. That’s all I got. Oh, and Dwight Howard needs to use more possessions. SHOCKING ANALYSIS YOU CAN’T FIND ANYWHERE ELSE.

Marreese Speights needs more minutes. I know this looks bad, but by God, he’s the best option they’ve got. This team continues to ignore Speights in pursuit of Lou Williams and Jason Smith. Come on, now! Free Speights!

Oh, Alando. Rough year for Barbosa. And when you look at this chart, the Suns don’t look nearly as stacked as they do when you think about them at first glance.

For fun, take a look at Tolliver’s Usage here and his usage for GS. It’s fun. Aldridge is seventh. Take out Shav and Patty and he’s still fifth. Three guards are ahead of him. And the man doesn’t pass! How does this happen? I’m also really impressed with how Oden dialed it back before, erm, uh… his… uh… accident. /moment of silence

Andres Nocioni is earning that paycheck, allright!

Duncan reclaims his rightful place at the top of the heap, but you can tell there’s a downgrade in supporting talent. Specifically, they need shooters, and adding combo forwards has not done the trick. In closing, Matt Bonner sucks.

POPS GETS BUCKETS. Take a second and revel in the fact that the Raptors have their starting point guard, their backup point guard, Antoine Wright and DeRozan all in the back half of Usage. But Patty O’Bryant’s gonna get his! How this offense manages to function at this level is stunning. But it is a really sturdy distribution. You could argue Bargnani might need more touche….ow! Ow! The Pro-Amir zealots are attcking! Help! Help! He has no jump shot! Ow!

They really have no room for Ronnie Brewer. The only decent thing to do is trade him to Memphis for a late first round pick. Really. Also, these guys have about a zillion guys they get solid contributions from.

:(

Okay, so that’s the usage comparisons. But what do these look like, when you compare them to actual production? Previously I used point production, but this time we’re going to use PER. Now, PER’s not a clean comparison at all. For starters, PER factors in a TON more in terms of rebounding, assists, blocks, steals, the works. It’s a cornucopia. There’s not a clear relationship between the two. However, what this can tell us is if the amount of possessions a player uses is worth his on-floor production. If you’ve got a low-usage, high-PER player, that guy’s making the most of his floor time without having to gun away. comparatively, if you’ve got a high-usage, low-PER player, well, then, yeah. I might end up running adjusted plus/minus on these as well to see what they look like.

Some more things to consider.

1. These are in the same order as the above charts, which means the PER distribution shape is going to be funky. Additionally, try not to track the overall shape but the individual production points in comparison to the usage. The hollow points are where the player on the opposite side has a negative PER.

2. PER is not a perfect rating system, it doesn’t account for defense, all that matters is winning, championship rings are the true measure of greatness, blah blah blah, blah blah. If you really buy into all that leading to the idea stats can tell us nothing, you should stop reading this immediately and go do something else like watch (redacted).

3. Same as with usage, it’s not going to factor in minutes. So you’ll notice Acie Law is a friggin’ beast by these metrics, even though he gets almost no time. So we have to consider that with each comparison.

4. For the purposes of saving time, and to allow you more room to draw your own conclusions (hopefully in the comments!), we’re going with all-caps proclamations for these analyses.

AL HORFORD NEEDS MORE TOUCHES.

SOMEBODY MUZZLE SHEED.

APPARENTLY NAZR MOHAMMED IS BOTH GOOD AND CONSCIENTIOUS WITH HIS DECISION MAKING.

WHAT. A. TRAIN. WRECK.

LEBRON IS GOOD AT LIFE.

JUB-JUB VERSUS BEAUBOIS, THE DEBATE RAGES ON

JR SMITH IS NOT HELPING HIMSELF.

THAT’S A VERY EVEN DISTRIBUTION OF SUCK.

NOT A BANNER YEAR FOR MONTA.

LOWRY DOIN WORK.

SIMMER DOWN, ROOKIES

KAMAN GOOD, NOT DOMINANT.

KOBE, GIVE PAU THE DAMN BALL.

YOU CAN’T TELL, BUT MARC’S ACTUALLY REALLY GOOD ACCORDING TO THIS.

IT LOOKS LIKE THEIR PER IS WEATING A HAT.

HAKIM WARRICK IS NOT THE SAVIOR.

COREY BREWER IS NOT GREAT.

SO MANY SHOTS, SO LITTLE RESULT

THEORY: CHRIS PAUL IS THE MVP,  EXHIBIT A.

AND YOU WONDER WHY YOU DON’T PLAY.

BE AFRAID. BE VERY AFRAID.

VINCE CARTER IS THE ROOT OF ALL THAT IS WRONG WITH THIS WORLD.

MAYBE SPEIGHTS SHOULD PLAY MORE.

NOT BAD AT ALL.

MY HEAD HURTS.

SCORE ONE FOR THE MARTIN CONTINGENT

BEASTLY.

RASHO AGAINST ALL ODDS.

TIGHTEN UP ON THE ROTATION, JERRY.

THIS IS YOUR BRAIN ON DRUGS.

For more comparisons on usage and team performance, this time using Offensive Rating, you can click here for Part II.

Matt Moore

Matt Moore is a Senior NBA Blogger for CBSSports.com's Eye on Basketball blog, weekend editor of Pro Basketball Talk on NBCSports.com, and co-editor of Voice on the Floor. He lives in Kansas City due to an unbelievably complex set of circumstances and enjoys mid-90's pop rock, long walks on the beach and the novels of Tim Sandlin.