Shot Selection Round-Up: Part 1
Shot selection and specifically shot locations have become a larger and larger part of the basketball conversation. It’s a topic of great personal interest to me and I’ve written quite a bit about it this season. To add an easily comparably quantitative element to the conversation, I also developed Expected Points Per Shot (XPPS). This metric is based on the expected value of shots from different locations and boils the quality of a player’s shot selection down to a single number. When we talk about high-value shots were usually referring to shots at the rim, three-pointers and free throw attempts. The scale of XPPS is aligned with league averages, numbers which are constantly over and under-performed. For that reason we often compare XPPS to Actual Points Per Shot and look at the difference between the two, which is called Shot Making Difference.
I’ve built visualizations which allow you to explore, sort and filter the XPPS numbers for players, team offenses and team defenses. I know those interactive graphs can be a little overwhelming so I wanted to pull out some of the most interesting numbers from this season and go a little bit deeper with them. Today we’ll be looking at some of the shot selection numbers for individual players, with analysis of teams to follow in subsequent posts.
This first table shows the players with the ten highest and lowest-value shot selections, as measured by XPPS. I separated the players into three groups based on their USG%, to differentiate between players with different roles.
As we mentioned above players over and under-perform the expected values of their shot selection all the time, which is a big factor in evaluating whether they truly understand their offensive roles and strengths. This next table shows the same 60 players, but instead of their XPPS I’ve listed their Shot Making Difference, which is the difference between their XPPS and their Actual Points Per Shot. You can see some players who take high-value shots, but don’t necessarily make them, as well as players who make a lot of low-value shots, usually long two-pointers.
The extent to which XPPS is useful in evaluating shot selection is pretty limited if you don’t also understand the context of their skills, limitations and responsibilities within the team’s offensive structure. Here are few of those numbers, both good and bad, with the context more fully fleshed out.
LeBron James – 1.079 XPPS (8th best in the >24 USG% bracket), +0.202 Shot Making Difference – What James did this season in the scoring efficiency department this season was simply incredible, increasing his FG% from essentially every area of the floor. However, he exponentially raised the impact of those gains in accuracy by improving his shot selection as well. Last season 37.9% of James’ shot attempts were long two-pointers. This season that percentage fell to 29.7%, with big increases in both shots at the rim and three-pointers. He made shots at an incredible rate this season, but he also made an incredible effort to make sure he was taking the right shots.
Tyreke Evans – 1.119 XPPS (5th best in the 19-24 USG% bracket), -0.002 Shot Making Difference – For his first few seasons in the NBA, Evans was the poster boy for unconscionable shot selection. A sensational rookie season was met with criticism of his inconsistent outside shooting. Over the next two seasons Evans seemed determined to prove those critics wrong, spending more and more time outside the paint, and in the process, proving those critics right by missing mountains of jumpshots. This season, he made some huge changes and it showed up in his scoring efficiency. First off, he became a consistent three-point shooter, knocking down 34.2% compared to a previous career high of 29.1%. Also, for the first time in his career he attempted more three-pointers than long two-pointers. Those inefficient and inaccurate mid-range shots made up just 16.7% of his shot attempts this season, by far the lowest percentage of his career. We always find time to celebrate the players who become better shooters, but we should also find time to celebrate players, like Evans, who become better decision makers.
J.J. Hickson – 1.111 XPPS (8th best in the 19-24 USG% bracket), 0.070 Shot Making Difference – Hickson is another player, like Evans, who made dramatic improvements in offensive efficiency by making dramatic improvements in offensive decision making. Last season 51.0% of Hickson’s shot attempts came at the rim. This season that number jumped to 65.3%. By being more selective with his long two-pointers, he also became more accurate. Last season he shot 30.5% in that zone, where this season he made 47.3% with a whopping 71% of his makes being assisted on. Concentrating on what you do well can yield tremendous benefits.
Tyler Hansbrough - 1.135 XPPS (2nd best in the 19-24 USG% bracket), -0.081 Shot Making Difference - How does a player who shoots below the league average from every area of the floor end up with a TS% above the league average? Free throws. Hansbrough took 361 shots from the field this season and 263 free throws. Only Dwight Howard and Reggie Evans had a higher ratio of FTA/FGA. He’s not a great finisher or shot maker from anywhere, but he has really focused on his strengths – getting to the rim and getting to the line. That FTA/FGA ratio was a career-high, nearly 50% higher than in any of his previous seasons. This was also the first season of his career where he attempted more shots at the rim than long-two pointers.
Dirk Nowitzki, Elton Brand, Chris Kaman – 0.946 | 0.950 | 0.951 XPPS (3rd, 2nd and 4th worst in their respective USG% brackets) – There is an absolute benefit to having players, especially big men, who can step out and knock down a mid-range jumper. It’s a pressure valve for an offense and can really buoy the efficiency of a group against tough defenses. The problem is when that shot becomes the centerpiece of the offense. Nowtizki is one of the best mid-range shooters in the history of the NBA and having him take that shot on a regular basis won’t break the offense. But the Mavericks stacked their front court with mid-range shooters the entire offense suffered. Last season when Nowitzki was on the floor 21.5% of his teammates’ shots were long two-pointers. This season, alongside Brand and Kaman, 26.8% of his teammates’ shots were long two-pointers. Even making those shots at an above average rate, as Brand, Kaman and Nowitzki can do, provides less efficient scoring that a multitude of other options. The Mavericks’ offense this season was a perfect example of the lesson that, “just because you can make a shot doesn’t mean you should take a shot.”
Evan Turner – 0.973 XPPS (8th worst in the 19-24 USG% bracket), -0.17 Shot Making Difference – Making 36.5% of his three-pointers this season was a big step forward for Turner. He’s also settled into a nice, accurate groove on long two-pointers, making 42.3%. The problem, as always, is balance. This was the third season of Turner’s career, and the third in which his ratio of long two-pointers to shots at the rim was roughly 2-to-1. Those long two-pointers made up nearly half his shot attempts this season and still outnumbered his newly accurate three-point shots by more than 3-to-1. He also shot a career low 47.9% on shots at the rim this season, where the league-average was 64.7%. Turner is a respectable mid-range shooter, but that shot just isn’t efficient enough to be the foundation of a richly versatile offense game. The bottom line is that he simply can’t be a viably efficient offensive player with this shot selection.
Tayshaun Prince – 0.963 XPPS (5th worst in the <19 USG% bracket), 0.008 Shot Making Difference – At this point in his career Prince’s offensive contributions come almost exclusively as a spot-up shooter. For most players this would equate to a lot of three-point attempts, but this season he attempted four times as many long two-pointers as three-pointers. Prince’s three-point attempts per 40 minutes this season were at a career low and even declined further as he moved from Detroit to Memphis. It’s a shame because Memphis is in desperate need of floor spacing and Prince has the skills to have a Shane Battier-like effect in that department. But to make that really work he needs to move a step or two back.
Andrew Nicholson – 0.954 XPPS (4th worst in the 19-24 USG% bracket), +0.154 Shot Making Difference – Including Nicholson on this end of the list may be a little unfair. He actually had a really solid rookie season and proved himself to be a reliable perimeter threat, both spotting up and as the screener in the pick-and-roll. Although his shot-selection looks terrible, with 45% of his shot attempts coming on long two-pointers, he drastically over performed the expected value of his shots and finished the year shooting 43.8% on those long twos. Although his XPPS puts him in the bottom ten, his actual points per shot were higher than Tyler Hansbrough’s, who ranked in the top ten in XPPS. He has the potential to be a supremely better version of Brandon Bass, but if he really wants to push the bounds of his efficiency it would be worth it for him to work on stretching his range out past the three point line. Nicholson didn’t attempt a single three-pointer this season, but shot a reasonable 34.0% from 20-24ft. Besides the added value of potentially earning three points per shot attempt, adding a few feet to his range will also open some considerable space in the paint for his teammates.