Extra Points: A New Way to Understand The NBA's Best Scorers

http://www.grantland.com/story/_/id/9795591/kirk-goldsberry-introduces-new-way-understand-nba-best-scorers


Extra Points
A new way to understand the NBA's best scorers
By Kirk Goldsberry on October 9, 2013



LeBron James is the best player in the world. That may be obvious to anyone who has watched basketball over the past few years, but for some reason it's hard to find many statistics to support this claim. Back when Michael Jordan ruled the NBA, he would commonly lead the league in points per game; Chamberlain and Russell put up insane rebounding numbers; Kareem Abdul-Jabbar is the league's all-time leading scorer.

James recently joined those giants as one of the five NBA players to win four MVP awards. Despite that impressive feat, his dominance is not reflected in conventional basketball stats. Somehow a vast majority of the game's most oft-cited statistics obscure the greatness of the game's greatest player. James shines in terms of Win Shares and PER, but those stats only vaguely describe what makes him so good. They fail to highlight any particular element of James's performance. In fact, many of our most common metrics imply that other players are more effective or more efficient than James.

Last season, DeAndre Jordan led the NBA in field goal percentage. However, consider the following ridiculous statistical couplet:

No player scored more points close to the basket than LeBron James last season.

No player converted a higher percentage of his shots near the basket than LeBron James last year.



LeBron James Interior Force - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


Think about that. Not only did he outscore every player in the entire league within the NBA's most sacred real estate, he converted his shots at the highest rate, too. As a whole, the NBA made 56 percent of its shots in that area last season; James made a staggering 72 percent of his 637 attempts there. Despite this unreal scoring prowess, James trails inferior interior scorers in field goal percentage.

Few people would argue that Jordan is a more efficient scorer than James, but according to field goal percentage, that's exactly what he is. Jordan led the NBA in that category, shooting 64 percent, while James managed only 57 percent.1 Although it can be useful to know which players convert field goal attempts at the highest and lowest rates, that doesn't really tell us much about scoring effectiveness.

Some players, like Steve Novak, shoot lots of 3s; others, like Kevin Garnett, rely on 18-footers; DeAndre Jordan works almost exclusively in the paint. Players like Jordan almost always lead the league in field goal percentage. The problem is that while NBA players have different roles and different shooting habitats, almost all of our shooting evaluations ignore that.




LeBron James vs. DeAndre Jordan Shot Chart 2012-13 - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


Anybody who has ever played H-O-R-S-E can tell you that some shots are easier than others; a layup is easier than a free throw, which is easier than a corner 3. This basic tenet is almost entirely overlooked by our most popular shooting metrics. Similarly, NBA players are all special, and over the course of a season each player generates his own unique "constellation" of shot locations. This is also overlooked; the graphic above demonstrates that the LeBron James constellation is considerably different from the DeAndre Jordan constellation.

We can improve our shooting metrics by accounting for court space and the unique natures of players' shot constellations.

At the end of the season, James's average shot distance was 11 feet; Jordan's was 2.8 feet. Essentially, through the hazy lens of FG percentage, James is penalized for having a jump shot, while Jordan is rewarded for not having one. Despite being a very good 3-point shooter, a good midrange shooter, and the most dominant interior scorer in the game, James trailed players like DeAndre Jordan and JaVale McGee in a crucial scoring metric.

What the hell?

While it's tempting to claim that our stats are "advanced" now, think about these two basic questions:

1. Who is the best shooter in the NBA?
2. What metrics would you use to justify your answer to this question?

When it comes to shooting stats, one would think there would be a spreadsheet somewhere on the Internet that delineates "great shooters" like Stephen Curry and Kevin Durant near its top, and "bad shooters" like Monta Ellis and Josh Smith at the bottom. But there's not. We still rely on hunches and vague reputations to make our assertions about "pure shooting" skill in the NBA. Amid the chatter of an ongoing revolution in basketball statistics, the notion that we still can't effectively measure shooting ability is troubling — but it's also correctable.

The issue is, almost every NBA player's overall FG percentage will always have more to do with where he shoots than how well he shoots.

The inconvenient truth is that every NBA field goal attempt has its own level of difficulty that's determined by several factors, including the shooter's location on the court. Even though previous approaches have mostly ignored this thorny reality, thanks to relatively new forms of NBA data we can now begin to understand it.

Last year, NBA players took just about 200,000 shots. The league's collective shot chart reveals the spatial nature of the NBA's average shooting efficiency. By itself, the chart shows the stark relationship between court space and expected points per shot — that's why 3-point shots are rapidly increasing in popularity while midrange shots are diminishing.




NBA Shooting - Average Points Per Shot - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


But this chart also provides a useful baseline that we can use to evaluate individual shooting performances. By overlaying players' shot constellations, we can estimate the expected total number of points that an average NBA shooter would produce, based on where he took his shots; then we can compare a particular player's actual yield against it.

For example, last season LeBron James attempted 1,354 shots. Using that league-wide baseline as our guide, if an average NBA shooter attempted this exact same set of 1,354 shots, he would produce a yield of 1,397 total points.




LeBron James ShotScore - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


James actually yielded 1,628 points from that constellation, 231 more than expected. No player accumulated more points than expected than James. By accounting for the fundamental relationship between court space and NBA shooting averages, we can see which players scored the most and least points above expected levels in the NBA. And we can figure out which players are actually the most effective scorers in their native shooting habitats.

Along with Ashton Shortridge, a professor at Michigan State, I did this analysis for every player who took at least one shot during the 2012-13 regular season. We refer to the difference between a player's actual point yield and his expected yield as ShotScore. For good shooters this number is positive; for bad shooters it is negative.

James came out on top, and is joined in the top three by a pair of elite shooters.

Highest ShotScores in the league:2

1. LeBron James, +231
2. Kevin Durant, +204
3. Stephen Curry, +164

These are three very different players with unique scoring strengths. Each accrues point surpluses in different spaces. Unsurprisingly, most of James's gigantic surplus comes close to the basket, where he puts up those freakish numbers.




LeBron James Shot Chart - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


Kevin Durant is also quite good near the basket, but he's more active and even more deadly from the outside than James. In a normal era, Durant would likely be the most effective scorer in the league, but we're not in a normal era — we're in the LeBron era. Still, Durant is an elite shooter from virtually every spot on the floor.



Kevin Durant ShotScore - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


He is so good that the weaknesses on his shot chart are the spots where he's only slightly above NBA average. That's crazy. Curry is a better perimeter shooter than both James and Durant, but he can't match their abilities near the basket, where he's actually a below-average NBA scorer. Still, thanks to his insane jump shot, Curry accumulates points at unusually effective rates.




Stephen Curry ShotScore - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


Calculating ShotScore for outside shots only (eliminating the inside shots) reveals the great "pure shooters" in the NBA. It should come as no surprise that Curry comes out on top. Last season, the Warriors guard took 1,120 shots outside of 7.5 feet; these shots resulted in 1,247 points, or 195 more points than expected. League-wide, in terms of ShotScore for outside shots, Curry leads a top five that also includes Kevin Durant, Jose Calderon, Kyle Korver, and Dirk Nowitzki.

These are players who already enjoy great reputations as shooters, but to this point there hasn't been a metric that has certified their superior status.

Controlling for the number of total shots taken helps further refine the idea. Jose Calderon had the highest ShotScore per outside shot. Calderon led the league with a +25 ShotScore per 100 outside shots, indicating that he accumulates 25 percent more points than would an average NBA shooter for every 100 outside shots he takes. This is especially impressive considering Calderon was playing for a pair of middling teams last season.

As it turns out, of the 162 NBA players who attempted at least 300 outside shots last season, only three of them accrued more than 20 points above expected values per 100 shots: Calderon, Kyle Korver, and Steve Nash, who despite his broken-down body still managed to put together great shooting numbers. Interestingly, many of the names on the list below are not frequent shot creators, but when they do get shots they are extremely accurate relative to their peers.

Top 10 Outside Shooters Per 100 Shot Attempts

1. Jose Calderon, +25
2. Kyle Korver, +23
3. Steve Nash, +21
4. Stephen Curry, +17
5. Dirk Nowitzki, +16
6. Serge Ibaka, +16
7. Jarrett Jack, +16
8. Shane Battier, +15
9. Danny Green, +15
10. Steve Novak, +14

So does this mean that Jose Calderon is the best shooter in the NBA? No, it means that when Calderon shoots it's a beautiful thing. We all know that the NBA isn't just a catch-and-shoot league; it's also very much a create-your-own shot league. Some NBA players, like Novak, only thrive in those catch-and-shoot scenarios, while others, like Stephen Curry, Dirk Nowitzki, and Kevin Durant, constantly create their own chances off the dribble or in the post. These differences are reflected partly in the volume of shots guys take every year, and relative to players like Curry, Nowitzki, and Durant, Calderon is not a very creative shooter.

I find it interesting that Jeff Van Gundy likes to refer to Novak as the "best pure shooter" in the NBA. The numbers don't back up that claim. Suggesting that a player like Novak is a better shooter than Curry, Durant, or Nowitzki is just as foolish as saying that DeAndre Jordan is a more efficient scorer than LeBron James.

How efficient would Curry or Durant be if they just camped out on the perimeter and took catch-and-shoot jumpers a few times per night? Unfortunately for their efficiency numbers, these guys also toil inside the arc and take tons of shots off the dribble. Conversely, how bad would Steve Novak be if he tried to do this at the end of a playoff game?




NBA shooting prowess involves much more than just spotting up.

We have to consider more than just field goals made versus field goals missed. The ShotScore approach begins that process by introducing the fundamental role of court space into the equation, but it by no means ends it. I am not suggesting that this is the most important new statistic in the NBA, but I do believe it clarifies which players regularly make or miss baskets at higher rates in their native scoring zones. Emerging data sets will enable this process further. For example, the SportVU data set allows us to additionally consider the openness of a particular shot attempt; after all, a wide-open 15-foot jump shot is a lot easier than that identical shot with Larry Sanders lunging at you.

One of the hallmarks of good NBA teams like the Heat and the Spurs is their ability to regularly create wide-open looks; many great shooters in the league aren't fortunate enough to play in such a scheme, and their numbers suffer as a result. For instance, there's no question that Danny Green has made great strides as a spot-up shooter, and his performance in the 2013 NBA Finals was incredible. But how much of his emergence is due to that beautiful Popovichian orchestra down in San Antonio? He wasn't getting those looks back in Cleveland when he was released.

This is an exciting time for basketball analytics, but as is often the case, deep explorations into performance often provoke more questions than answers. It's still tempting to assert that NBA analytics are advanced now, but the truth is that we will look back at the current state of affairs the same way baseball nerds look back at the batting average and RBI era. These players give us so much; the least we can do in return is come up with an accurate way of appreciating them.




bob


.

Here's another article by the same author on the same subject.  I thought I'd just keep this topic together.  This one looks at Rondo's shot score and compares him to others.


http://www.grantland.com/blog/the-triangle/post/_/id/77702/shooting-blanks-the-players-with-the-worst-shotscore


Shooting Blanks: The Players With the Worst ShotScore
By Kirk Goldsberry on October 10, 2013 3:30 PM ET




GLENN JAMES/NBAE VIA GETTY IMAGES



Yesterday we looked at ShotScore, a new method to identify the NBA’s best scorers. You can read the full piece here, but in a nutshell, the method compares the actual point yield of an individual NBA shooter against an estimated tally of what an average NBA shooter would accrue from that exact same set of shots. This is a useful way to evaluate shooting because unlike field goal percentage, it accounts for where on the floor the shooter is most active and factors that in to the analysis. Midrange shooters are compared against the NBA’s average midrange production, etc.

Players like Stephen Curry, LeBron James, Kevin Durant, Jose Calderon, and Kyle Korver immediately rose to the top; these players consistently outperform league averages from their most active shooting zones. But, it’s also instructive to identify the players who make shots at rates much lower than league averages, the guys that regularly underperform relative to their shooting cohort. Here is the bottom five:

Lowest ShotScore, 2012-13 season
1. Monta Ellis, -159
2. Greg Monroe, -134
3. Ramon Sessions, -130
4. Russell Westbrook, -127
5. Ricky Rubio, -115

These five players are all relatively active shooters, but fail to accrue points from the field at average NBA rates in their most active zones. In other words, each player struggles in his own unique way.

Monta and Jose: The Dallas Odd Couple

You could make a good roommates sitcom about a European perfectionist, forced to work alongside a purely American free spirit who "doesn’t play by the rules." Or you could just watch the Mavs this season. It will be fascinating to see Ellis play on a team led by Dirk Nowitizki. But it might even be more interesting to watch his backcourt interactions with Jose Calderon.

Calderon and Ellis are two of the most noteworthy shooters in the league, but for very different reasons. Basically, Calderon is a great shooter, and Ellis is not. Calderon is particularly lethal in front of the basket.




Jose Calderon ShotScore - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


On the other hand, Ellis is a very creative player with a knack for getting a shot pretty much whenever he wants. This is an overlooked skill in most player evaluations. Ellis’s problems begin the moment his shots leave his fingertips. Simply put, he’s a below-average shooter with a volume problem. The combination of his high activity and his terrible efficiency has frustrated fans from Oakland to Oshkosh, and this year he brings his game to Dallas.




Monta Ellis ShotScore - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


Ellis shoots below league averages from everywhere, except for one tiny baseline area where he’s hardly active. But that actually might be the key to his improvement. If Rick Carlisle can reduce Ellis’s activity and get him to more effectively pick his spots, Ellis has a tremendous chance of improving. He has never really played in a good system with decent offensive teammates. Playing alongside savvy veteran scorers like Nowitzki, Shawn Marion, and Vince Carter should help. If that team intelligence and Calderon’s shot discretion rub off on Ellis, there’s a very good chance we will witness a drastic upswing in his efficiency.




Jose Calderon, Monta Ellis ShotScore - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


Either way, this odd couple in Dallas will make for must-watch League Pass TV this season. Signing him was a risk, but in the same way that Nowitzki and Calderon know about shooting effectiveness, that modern-day Texas tycoon, Mark Cuban, knows a thing or two about investment effectiveness. Let’s hope it doesn’t turn into a soap opera.






On Rondo, Russell, and Rubio

Even though we love to blast Rajon Rondo because "he can’t shoot," there are a few point guards more deserving of that rep, prominent floor generals who actually fail to create points at above-average rates, from both the inside and the outside. Rondo, believe it or not, can actually makes shots at rates above league averages. In fact, in the last few seasons, he has made his elbow jumper at elite rates, partly because he’s left open and partly because he is an improving shooter.

Russell Westbrook is one of the most polarizing players in the league. Although most of us love him, the jury is still out on just how effective of a teammate he is. He’s obviously a great playmaker, but he has a tendency to shoot a little too frequently, which in some cases takes away more efficient options from his teammates (one in particular). As a scorer, Westbrook could improve his efficiency by learning from one of his biggest rivals.

Back in the 2002-03 season, a young Tony Parker attempted 243 3-point shots; a decade later, during the 2012-13 season, Parker only shot 68 3s. Parker and the Spurs recognized that not only were these shots not very effective (31.3 percent for his career), they also introduced a huge opportunity cost. Every time Parker shot a 3, it meant he wasn’t attacking or playmaking. The new and improved Tony Parker doesn’t take those 3s himself, he helps create them for guys like Danny Green, Kawhi Leonard, and Matt Bonner.

There’s no doubt that defenses secretly love when Westbrook shoots from long range; he’s not very efficient, and it means he is not shredding through defenders, attacking, and dishing to open teammates. The fact that Westbrook attempted 300 3s last season is troubling — this was up from 196 the year before, plus he only made 32 percent of them. It’s really hard to justify this level of activity. Westbrook is much better suited on the basket side of the 3-point line, pulling up from the elbow, driving to the hoop, drawing fouls, and distributing the ball.




Russell Westbrook ShotScore - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


The good news is that in the same way that Peyton Manning and Tom Brady can be great without being great runners, it is possible to be a great NBA point guard without being a great shooter. In fact, Westbrook and Derrick Rose, two of the world’s best point guards, prove this on a regular basis. Still, Westbrook’s game, as great as it already is, could improve if he just stopped shooting all those 3s. Unfortunately, this season may not be the best time for him to stop. The loss of Kevin Martin has left the Thunder in dire need of replacement production from beyond the arc. We need to get Westbrook a Matt Bonner for Christmas.

Another point guard with limited perimeter passing options is Charlotte’s Ramon Sessions. He is a wannabe attack guard who has the speed and skill to maneuver through NBA defenses, but has trouble finishing at the basket.




Ramon Sessions ShotScore - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


Although he gets a lot of opportunities at the rim, Sessions converts less than half of them. He also struggles as a jump shooter, but to his credit, if his relative inactivity out there is any kind of clue, he seems aware of this limitation.

In a way, Sessions is really similar to Ricky Rubio. Like Sessions, Rubio’s troubles are most extreme closest to the basket. He made a dismal 41 percent of his shots inside 7.5 feet last season, which was tied for worst in the NBA (with Austin Rivers and Randy Foye).




Ricky Rubio ShotScore - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


Saying that Rubio has trouble finishing at the rim is kind of like saying that Toonces the cat has trouble driving on mountain roads. But Rubio’s troubles extend outward from the hoop. As it turns out, not all Spanish point guards are freakishly good shooters. When it comes to shooting efficiency, Rubio is no Calderon. Rubio's pretty lousy in those exact same elbow areas where Calderon thrives, and Rondo has greatly improved. So, to the people that continue to make those “Rondo can’t shoot” proclamations, I propose you take your sneers elsewhere, perhaps to Oklahoma, North Carolina, or Minnesota.

Speaking of troubles near the basket, the Pistons' Greg Monroe presents another interesting example. Monroe was by far the NBA’s most active shooter near the basket last year. That’s good, except that he struggled to convert his shots down there. He has never met a close-range shot he doesn’t like. This is compounded by his immature midrange game. Although Monroe’s interior numbers weren’t terrible, his slightly below-average production combined with his extreme volume resulted in him arriving at the bottom of the ShotScore list.




Greg Monroe ShotScore - Kirk Goldsberry/Grantland
KIRK GOLDSBERRY/GRANTLAND


Despite his troubles last season, Monroe remains a very strong NBA prospect. With Monroe, Andre Drummond, and Josh Smith the Pistons seem well positioned to dominate the interior for years to come. There is little doubt that Monroe will improve both close to the basket and away from it as his game matures, but as it stands, he is notable for his inefficiency.

None of these players are bad basketball players; in fact most of them are really good even compared to their NBA peers. When it comes to shooting efficiency, however, each of these players has an important issue or two to address.



bob



.

Yes, Lebron is very good at taking the ball to the hoop.

Sam

Hi,
Yes, he is.
OTOH, I wonder what it'd be if LJ is called at least in 50% cases when he "bulldozing" (aka offensive foul) his way to the hoop and/or makes 3-4 steps.

AK

Bob

I've taken a little time and looked over these shot charts they are very interesting BUT......

How many of these shots were wide open looks vs hand in the face under pressure? Or how many were forced last second buzzer beater shots before the end of a quarter or the 24 second clock, when a shot HAD to be put up?

So is a good shooter also a "clutch" shooter?

Remember when it sure seemed Bron wanted nothing to do with taking a last second shot?

Again these are interesting for sure..... worse shot to me is the one with your feet on the arc.

beat

According to these charts, most players can't hit mid-range shots.

And LeBron is benefiting from an era where offensive talents are coddled to pander to the star-studded spectacle the NBA has become. If players were allowed to be more physical on defense, LeBron would be less dominant.


KJ

Are the shot charts also accounting for shots that go in when fouled? (I am assuming shots missed when being fouled are not).

Interesting, but misses lots of the nuance of the game.

People like Monta Ellis and Greg Monroe have generally played on very poor teams - where they are often forced to shoot with time running down on the shot clock or stuck with the ball in a bad position with a court full of guys who dont want the shot.

Kendrick Perkins lead the league in FG% for a number of years in a row - so we know how numbers dont tell the whole story.

Good group of postings none the less. Thanks Bob