Last night I had a “great” idea for a post about Derrick Rose. But as I wrote it out this morning I was less thrilled with my “great” idea (not that it was a “bad” post, just not as “great” as I originally thought). And had I just waited, I could have simply put up a better post by re-posting the latest from Ty Willihnganz.
In Washington Wizards “Stumbling on Stupidity”, Ty introduces (with data from Hoopdata) “High Value” Scoring Possession Percentage and the Sh*t Shot Percentage. To understand what these mean, here is Ty’s post in its entirety.
Last night the Washington Wizards lost by 29 points to the Orlando Magic, and looked like a CBA team in the process. A bad CBA team.
How did it happen? Well, let’s leave aside for a moment the fact that Ted Leonsis purportedly read and endorsed David Berri’s book “Stumbling on Wins“ last spring and then signed off last summer on the formation of a roster that could be subtitled team “Team Anti-Stumbling on Wins”. (That is the worst rebounding team I have ever seen). Puzzling. But leave that aside. There’s a much simpler way to explain last night’s national cable television debacle.
Lets look for a moment at simple shot selection. It tells a lot of the tale.
In the NBA game, there are three “high value” ways to use a scoring possession: (1) at the rim; (2) behind the three point arc; and (3) at the foul line. Those three methods of usage promise the greatest return on the investment of your possessions. Thus, if a team wants to be successful over the long term, it wants to maximize those types of use. Incidentally, the NBA average “High Value” Scoring Possession Percentage used is 62.5%
Using rough numbers, the Orlando Magic’s “High Value” scoring possession percentage last night was 64.8%. That’s above average and the sign of a smart, disciplined team. Sure, they missed an unGODLY amount of foul shots, but over time that shouldn’t happen. The Magic will be successful if they put themselves in an above average number of high value situations as they did last night. (Its like playing dice. If you calculate the odds properly, and bet accordingly, you’re much more likely to succeed. Not guaranteed, but likely.)
The Wizards, on the other hand, did not play the odds properly. They did not play a smart, disciplined basketball game. Their “High Value” scoring possession percentage was a measly 52.1%. You cannot win consistently playing that recklessly (or carelessly, or however you want to describe it). Maybe you’ll get lucky on a couple of occasions, but not often.
Now let’s look at the inverse, what I call the “Shit Shot” percentage — the team’s percentage of shots taken from 16 to 23 feet away from the hoop. These are the absolute lowest value shots and should be avoided. Besides being stupid, they are also a sign of (a) weakness — you lack the strength to go to the hoop, or (b) laziness – you are settling for an available “bad” shot instead of working to get a good shot. Shit shots are available all night long. That’s because they don’t return very well on the investment.
The NBA team average “Shit Shot” percentage is 20%. One of five possessions. Last night the Magic’s “Shit Shot” percentage was 10.6% — well below the NBA average, and the sign of a hardworking, intelligent team.
The Wizards, led by John Wall’s 10 “Shit Shots” (of which he made two), had an overall “Shit Shot” percentage of a staggering 33.3%.
As Singletary would say, “Can’t win with em… can’t do it”. Those numbers are the recipe for a blowout, and that’s what happened last night.
One more observation….In the comments on Ty’s post, Chicago Tim makes this observation about the Bulls.
So against the Thunder the Bulls shot 24 of 93 field goal attempts from 16-23 feet, for a “Shit Shot” percentage of 26%. Not as bad as the Wizards, but still below average.
If only I had this idea when I wrote my Derrick Rose post!
- DJ
I knew it was on it’s way. But it doesn’t make it hurt any less. I don’t know if I can go through an entire season of this.
Leonsis is really engaged in this team and really wants to succeed. You can see it. But he has done everything wrong personel wise so far. Why would you extend Blatche before he’s done anything? Frustrating.
I’d like to see this site expand into a Database of sorts. Stats like these are more difficult to find, yet interesting and potentially telling of a teams strengths and weaknesses.
I realize there’d be legal and upkeep aspects you probably don’t want to deal with, but a well maintained database (and not just that link to WP48 you send out, which by the way I can’t see at work) would go far.
One source with: basic stats, splits, WP, WS, defensive/offensive eff, true shooting, high value/shit shot, +/-, etc. Guess thats a lot to ask.
Related to the shooting percentage and shot selection is the perhaps historic amount by which the Wizards were out-rebounded by the Magic last night. According to the box score, the Wizards were out-rebounded 53 to 25. A more than 2 to 1 advantage in rebounding in an NBA game for one team vs. another seems quite incredible. McGee and Blatche had 2 rebounds each and Wall had 1 rebound. So much for a strong young nucleus. It’s going to be a very long season in Washington.
I really enjoyed this one.
-Matt
Trust me, everyone would like an up-to-date internet accessible database that tracks these statistics. I don’t think that there are too many legal considerations as statistics are not copyright-able, much like news events.
The problem is more that there’s not one site that has all of the raw data needed to calculate those numbers. You could probably scrape multiple sites to get all of it, but doing so increases the complexity of the software you would have to write to do it. Adding a second source website more than doubles the amount of work that it takes to fix things when they break because not only are you adding a new site with a completely different system for presenting data, but you are also likely adding interactions within your software of subsets of the two separate data sets. As you add tertiary or further source sites, the interaction complexity increases a great deal.
Dealing with such complexity is fine for data companies, but it’s a bit much for a small number of volunteers (right now Andres is the only person who has really done any software development in support of the WoW network.
“and not just that link to WP48 you send out, which by the way I can’t see at work”
I assume you’re talking about Andres’s database. If you can’t see it at work, it’s because your work computer doesn’t have MS Silverlight. Ideally, we would have the resources to build a database on web standard technology. The thing you should understand though is that a web development framework like Silverlight can greatly reduce the amount of work needed to get something like that off the ground. Again it just comes back to the fact that there is neither a large community of volunteer web developers nor any money to hire developers at the disposal of the WoW network.
For the time being, if you are very interested, I suggest that you grab a version of excel (or OpenOffice, like me) and get used to ctrl+c and ctrl+v. That’s what all of the writers, including Berri, do. Alternatively, you’ll probably just have to wait for articles to come out or for some more developers to step forward and offer their time.
Sorry to be a bit of a downer, but that’s the state of things at the moment. There has been a lot of growth in a relatively short amount of time though, so I imagine that more will come in the not too distant future.
-SA
Who needs rebounds when you have 20 ft turnaround fadeaways? Oh wait…
Appreciate the response.
I’m an Excel guy, but not a software/Web guy. So, I guess I don’t understand all the complexities. I was assuming you could get a pretty sizable amount of data analysis from the box score and maybe one or two other sites (but again I don’t really know what linking these sources would entail).
Also, I do appreciate all the work you guys do, and enjoyed the article. I’m a relatively new reader and saw the sites been up for years, so I suppose I didn’t grasp that its a growing resource stemming from only a few volunteers.
I love this website for many reasons; but, the funny catch phrases like “shortage of tall people” and “Sh*t shot %” are my favorite reasosns.
By the way, did anyone catch Rambis’ reason for not playing Love late last night?
I agree, this is one of the best pieces that’s ever been posted here.
I remember reading years ago, maybe in the comment section here, that San Antonio’s overriding philosophy was to take lots of high value shots and force the other team to take a lot of sh*t shots.
And here I was thinking “shit shots” were “Warriors shots”. The Warriors had a 35.8% shit shot rate against Houston the other night. But incredibly, they hit 51.6% of those shots.
Is a shit shot < crap shoot?
(sorry, couldn't resist.)
Free Kevin Love! Free Kevin Love! Free Kevin Love!
It’s like watching a beautiful sweet woman mistreated by her boyfriend.
If I was a GM I would stand outside his bed room window at night with a bouquet of flowers and serenade him with this.
Here you go just to be evil:
Love to Portland for Rudy and Babbit
http://games.espn.go.com/nba/tradeMachine?tradeId=34mk2yh
or
Love to Lakers for Mr. Sharapova
http://games.espn.go.com/nba/tradeMachine?tradeId=2a96efd
or
Love to NOH for a trade exemption
http://games.espn.go.com/nba/tradeMachine?tradeId=23tc8rk
or
Finally my favorite:
Love for Big Baby straight up
http://games.espn.go.com/nba/tradeMachine?tradeId=2wn264w
(Hollinger has Boston losing ten games and Minn gaining 8 on this one :-) . Let’s do it!!!)
Monta Ellis has been passing the ball. He is really looking for his teammates this year and it’s helping his efficiency and the team as a whole. It’s looking pretty bright for the warriors.
Hi Arturo. Can you figure out why Hollinger says this would hurt the Celtics if even PER says Love is having a better year?
JP,
I don’t think anyone but Hollinger knows the answer to that one. I just want to take advantage of it.
Thanks arturogalletti.
Shots from 16 to 23 feet away from the hoop are the easiest to call bad but league-wide last season shots from 10-15 were very nearly as bad and still below 40% FG%. Short jumpers not considered at the rim out to 10 feet are sometimes a little better, sometimes not at player and team levels. Rather than just criticize long jumpers I tend to want to reduce all mid-range shots.
Orlando, Houston, NY have been among the best at reducing these low percentage shots.
I haven’t seen enough data on league-wide FG%s from middle distances contested vs uncontested or heavily contested vs lightly contested. That might matter some for some players. Maybe an uncontested or lightly contested mid-range shot is a good shot for some. Does Synergy give the basis for assembling a league average (whether they make it easy or not)? Whether the shot is significantly contested might not matter a lot for that many players, despite what they think about their ability to make those open chip shots.
If you can make 50-55% of open mid-rangers then those would not be bad shots. If you are truly open and if you are that good.
Look deeper my stat-scrutinizing friends.
Flip Saunders’ teams always rank among the league highest in attempts from the long two. This is a consequence of his playbook and general philosophy: work to create open shots. His version of the Hawk offense uses pindown screens on the baseline and off-ball motion to get players in position to take unguarded shots.
Often this means his players are taking the ‘wrong two’ — the long two-point shot. (Hoopdata has only a few years of shot locations, but check Flip’s Detroit years for instance). And you might expect that this means his teams are consistently posting low eFG% since they’re jacking jumpers and taking shots when there’s nobody near to foul them.
But you’d be wrong. Looking back through Flip’s history, he’s often posted high efficiency with his system.
Kevin Garnett was often noted as a jumpshooting Big. Picture Antonio McDyess in Detroit with his money jumper from the elbow. Replay Rip Hamilton running patterns on the baseline to pull up from long midrange.
Now, in their 2nd year in the system are the Wizards proficient in running the sets? No, not yet. This is a young team, occasionally an immature team, to be fair. But it’s tough to kill them merely for trying to run what their coach has designed for them.
There’s a temptation among the stat cognoscenti to sneer at coaching as something less than relevant. Talent wins games– Phil Jackson never won a ring without Kobe or MJ drawing defensive attention. But I think if you studied the science just a little bit you’d come to appreciate even more what it is you are actually watching.
Understood, numbers are perhaps somewhat less challenging, more static, easier to follow and to feel as though you have mastered them. But you’re missing out on the real fun of the game if you remain satisfied to tally the numbers after the fact.
Flip’s teams tend to improve in offensive efficiency as they learn his sets. This team has six players left over from last year, and is captained by a rookie point guard. It’s a bit cheap and easy to sneer at their showing in game one against a team who has proven dominant in defensive efficiency in recent years, and who has had little turnover on their perennial EC championship contending squads. But one game is no measure of the team’s (or any player’s) ultimate upside.
For example, check Mr Tiny-Wins-Produced John Wall vs Atlanta:
50% from the long two.
Near 60% eFG.
9 assists with a 3/1 ast/TO ratio.
I expect the team to suck somewhat (if not the epic historic rate projected by Galetti et al) but I also expect to see solid improvements over the course of the year. In the performance of John Wall especially. It’s not where you start, it’s how quickly you can fix your flaws.
I’d always felt a mid-range game was what separated the men from the boys, so to speak. Are there some who are better at it than others?
Good teams might work to get and get more open mid-rangers and often do better on them because of that or having better shooters. You want to maximize your returns there, even if you still might want to minimize the mid-range frequency.
As for Saunders in Detroit…
2007-8 6th Offensive efficiency, 15th on team eFG%
2006-7 6th Offensive efficiency, 21th on team eFG%
2005-6 4th Offensive efficiency, 7th on team eFG%
Good offenses but generally more because of low turnovers and offensive rebounding rather than good shooting and more so as time passed. The shooting deteriorated rather than improved, though this is a product of player aging and player response to coaching.
In Minnesota, some years with top 5-7 in the league on eFG% but more years of ranks of 7-13 or below and usually better ranked on overall offensive efficiency than on eFG% because of the other factors.
The message? To me it is that it is ultimately about efficient offense not just efficient shooting and that plays should give substantial attention to reducing turnovers, getting foul shots and offensive rebounds as well as making FGs. Maybe the carefully scripted offense with lots of mid-range shots reduced turnovers and made it easier for the rebounders to know when and where shots were coming from than in a less structure offense or one that relied more on 3 point hoists or interior players looking for shots instead of rebounds.
Looking at expected eFG% based on floor locations and actual to expected eFG% offensive ratios at hoopdata 8-10 teams did better than Flip’s Pistons for the years the data is available.
The leader on actual to expected eFG% offensive ratio from 06-07 thru 09-10 and probably longer? The Suns every year. That is the system that squeezed the most out of field goal shooting. What did that offense look like from a shot location perspective? High on 3 pointers and moderate to high on inside shots. Moderate to low on mid-rangers.
Of course, taking the right shots doesn’t guarantee success either. LAC was 5th in the league last season with 49.3 XeFG%, but dead last in ORatio (eFG%/XeFG%) at 0.82.
Whoops, I should correct myself. I was citing 2010-11 data for the Clips. However, 2009-10 wasn’t much different for them (near the top in XeFG and bottom in ORatio).
Speaking of the Clips, did anybody watch the game against the Spurs last night? Sean Elliott said the worst thing about Vinny Del Negro as a player was what they called “The Vinny.” Apparently, Del Negro was notorious for taking jumpshots one step inside the 3-pt line! He said they would yell at him all the time about taking “The Vinny”!
So the obvious question is… Is there any correlation between a player’s shit shot percentage and winning percentage as a coach? Lmao…
That shot drives me crazy. Does anyone try to explain the math to them?