Post-game Reactions

We’ve passed the peak of an NBA offseason which saw Gordon Hayward hijack July 4th, the Celtics trade the #1 overall pick, and Lonzo and Tatum sell out a Summer League game. No offseason would be quite complete without hearing about [Insert Player Name] extending his range to the 3-point line. While it would be more than a stretch to say Marcus Smart is only now adding the long ball to his game1, he could certainly use the work.

A video surfaced this offseason showing Smart knocking down 3 after 3 from Steph Curry range as he repeats, “I told you…it’s a different me.” Is this truly a different Smart or is this just another 30 second video of an NBA player hitting open shots in an empty gym? With the Celtics decision on whether to extend Smart looming, his shooting will be a vital factor in determining his future value. Aside from that 30 second video, what have we learned about Smart’s shooting ability over his first three years?

Smart has shot a paltry 29.1% from deep in his career – 33.5% in his rookie year to a career-low 25.3% in his second year before settling in the middle at 28.3% last year. That 8% drop from his rookie to sophomore year is a significant, but not unheard of, decline. Shooting percentages are some of the least consistent stats year over year, with 3P% being the most variable of the bunch. Even metrics such as BPM and WS/48, which attempt to encapsulate the entirety of a player’s impact in one number, are more consistent than 3P%. For more evidence, look no further than the newest Celtic, All-Star Gordon Hayward. Hayward’s 3P% over the years: 47%, 35%, 42%, 30%, 36%, 35%, 40%. Is he a 47% shooter, a 30% shooter or somewhere in between? So many factors besides raw shooting ability can determine the outcome of a shot, such as how close was the closest defender or how many dribbles were taken prior to shooting. This can lead to 3P% fluctuating greatly from year to year since there are not enough attempts in a single season to balance these variables, plus blind luck, out. How much data do we need to pinpoint real 3-point shooting ability if single season 3P% is not enough?

Darryl Blackport found that it takes about 750 3PAs for a player’s 3P% to reasonably stabilize. This is not great news considering Smart has attempted 845 3s in his career and is sitting at 29%. However, he does not argue that a player’s 3P% after 750 attempts will necessarily predict that player’s 3P% moving forward – players can change teams, roles, or even improve! He brings up Kyle Lowry as an example – Lowry shot 32% from 3 over his first 750 attempts and has consistently shot between 36-40% since then. And while Hayward’s 3P% has fluctuated greatly from year to year, when looking at it in 750 attempt chunks it paints a more accurate picture of his true shooting ability – 37% over his first 750 3PAs, 35% over the next 750 attempts and 40% on 374 3PAs last season. Will Smart continue to shoot at a 29% rate or can he take a Lowry type leap? Let’s look to see how other players who had similar shooting woes to start their career fared moving forward.

Of the 57 players who have shot less than 30% from 3 over their first three seasons, 52 saw their 3P% improve by at least 1% over the rest of their career2. That’s an encouraging rate of improvement, especially considering 2 of the 5 players who have not improved are Giannis Antetokounmpo and Andre Roberson, both only 4th year players last season. Smart is one of the youngest players on this list as only 16 of these players were 22 or younger in their third season, only one of which hasn’t improved their 3P% – Giannis. Even if Smart is only mentioned in the same breath as Giannis for negative reasons, I’ll take it. This list is also littered with point/combo guards similar to Smart – Larry Hughes, Terry Porter, Derek Anderson, Reggie Jackson and Russell Westbrook to name a few. Keep in mind this is a small sample to draw conclusions from and none of them had jacked quite as many 3s as Smart so their percentages may have been less stable to begin with. Still, if we assume Smart’s 3P% will improve how much of a jump can we expect?

The average 3P% increase for this group is 5.6%. If Smart makes even an average leap in terms of this group he would be a 34.7% 3-point shooter over the remainder of his career, just shy of league average. He may not make that jump in one season as we’ve seen how volatile 3P% can be. For next season my model is projecting him to shoot 32.7% from deep based on his 3P%, 3PAr and FT% from the previous three seasons. It’s important to note that an increase in 3P% is not necessarily proof that a player has improved their shooting ability. While 91% of players who shot <30% saw their 3P% improve, only 31% of players who shot >40% saw their 3P% improve. I don’t think these 40% shooters all of a sudden became worse, more that both sets of players regressed to the mean. Even if we expect Smart’s 3P% to improve based on pure variance he could certainly make some adjustments to his game to help his cause.

According to NBA.com, in the 2016-17 season only 13% of Smart’s shots came from midrange – less than Steph Curry and JR Smith – with 69% of his shots coming from the restricted area or behind the arc – more than Lillard and Beal. His overall shot distribution is quite nice but it’s the individual shot selection where Smart needs some work. On too many of his shots he finds himself off balance, often stemming from a difficult step back or attempting to draw a foul.

Having said that, 33% of his 3s last year were catch-and-shoot and while I’d like to see that number come up a bit, it’s a solid rate. The problem is he needs to make them. Smart only hit 31% of his catch-and-shoot triples, producing the lowest figure on the team for the 2nd straight season. The most troubling stat might be his 3P% when left wide open (defined as closest defender being 6+ feet away). Smart shot 27% on 128 attempts last year, 25% on 67 attempts the year before and 40% on 58 attempts his rookie year. If he is going to be a successful 3-point shooter in this league that number needs to be much closer to his rookie year than the last two years.

Marcus Smart shot charts for 2014-15, 2015-16 and 2016-17 courtesy of NBAsavant.com

This is what makes Smart such a frustrating shooter – inconsistency. He became one of 50 players in NBA history to hit seven 3s in a playoff game, but followed that up by shooting 1-9 from the field and 1-5 from 3 the next game. While people have looked to his increased FT% each season as evidence he is improving as a shooter, I’m not entirely sold. I believe what makes FT% a good predictor for 3P% is the ability to showcase a consistent shooting stroke by removing all the noise usually involved in taking a shot. While I’m obviously happy he’s improved his FT% from 65% his rookie year to 81% last year, I’d rather he consistently shot 76% each year to prove his shooting stroke is consistent and repeatable.

If he continues to shoot roughly 80% from the line then I won’t be concerned, but his 64% shooting in the playoffs last year doesn’t exactly instill confidence. There’s not much precedent for players who were this inconsistent from the free throw line in their first three years to go on to be successful 3-point shooters. Then again being consistent at the line doesn’t always translate to proficiency behind the arc, just ask Ricky Rubio.

While Smart does launch some questionable 3s it’s at least promising that he understands the value of the shot. His career 3PAr is on par with legendary 3-point shooters such as Steph Curry and JJ Redick. If we’ve learned anything about basketball in the last few years it’s that 3>2 (or that 23>23 depending on who you ask). He also understands the value of a game and doesn’t mind heaving shots at the end of a quarter. According to basketball-reference.com he had 9 heaves last season, 2nd only to heave-master Steph Curry. He also shot 95 shots with the shot clock at 4 or below, about the same number as Paul George and Bradley Beal. Not all of that is a positive takeaway – Smart will settle for long 3s at the end of the clock too often and he might like those end of quarter heaves a bit too much. Still, you won’t find him passing up a shot just to protect his percentages.

Even if Smart doesn’t take a leap in 3-point shooting I still want the Celtics to keep him around. His playmaking, defense, toughness and versatility are sometimes hard to quantify, but never go unnoticed. He doesn’t just dive for loose balls, he creates them out of thin air. He doesn’t just lock down opposing point guards, he outduels centers for rebounds. He doesn’t just heave half court buzzer beaters, he seeks them out even though we all know it’s bound to end up destroying some dude’s nachos in the 10th row. If he does become a league average 3-point shooter – whether by improving his shooting stroke, adjusting his shot selection, having a little luck bounce his way, or some combination of all three – he will be one of the best two-way guards in the league. It’s not as simple as just adding a few percentage points to his shooting splits and bumping up his points per game either. The threat of being a knockdown shooter will open up more lanes for himself and his teammates which will lead to even more winning plays.


1. Only 46 players in NBA history have attempted more 3s in their first three seasons
2. Minimum 150 3PA over first three seasons and 150 3PA over rest of career

The following two tabs change content below.

Graham Allen

Graham joined Celtics Hub at the end of the 2016-17 season. His writing focuses on the statistical side of basketball, though shockingly he enjoys watching it as well. You can follow him on twitter @grrraaahhhaaamm.
Tagged with →  
Share →
  • leitskev

    He needs modest improvement. He’s never going to be a major threat from the arc, but modest improvement opens up more space for others as defenses have to adjust to him at least a little. Also, his natural confidence means that modest improvement during the first three quarters will pay off with more moments like what we say in the one game against Cleveland.

    I am more alarmed that he can’t finish drives. And this doesn’t seem likely to improve. If you watch footage of Smart playing pickup hoops with Brown and others, this inability to finish seems to show even there. He puts himself in positions where he can’t find an arm angle, so he ends up chucking it up, and some horrific shots result. In actual games, this means his main purpose in driving is ALWAYS to draw a foul, and this annoys refs after a while. Eventually they stopped giving him the benefit. Compare this to IT, who feels the defense and shifts his arm to find the clean angle for a shot. I don’t think this kind of thing can be taught. If Smart could improve in this area it would make a huge difference in his game.