This article was written by By Stephen Shea, Ph.D and published on his blog: Basketball Analytics. You can find out more about Dr. Shea and his work in the field of Basketball Anayltics at the end of this article.
Editor’s Note #1 from Brian Williams, The Coaching Toolbox: The acronym that Dr. Shea uses for Painfully ill-advised Shot Selection “PISS” is to make the point as to how he views those shots. You might want to find a new way to abbreviate this stat if you do want to use it with your players, and the intent is not to offend any coaches who are reading.
Editor’s Note #2 from Brian Williams, The Coaching Toolbox I realize that not everyone who reads this plays with a shot clock, but I do think that there are some very good points to think about–especially the point that by taking contested dribble pull up jump shots, you score at the same rate as a player that teams foul on purpose to force him to shoot free throws.
By Stephen Shea (@SteveShea33) and Christopher Baker (@ChrisBakerAM)
NBA possessions are a valuable and limited resource. Each shot is a gamble of that resource. The goal of an NBA offense is to try and create the best bets—shots with the highest expected winnings.
Of course, defenses have the opposite agenda. They are trying to force the worst bets.
Analytics have taught us that certain shots have better odds than others. For example, catch-and-shoot jumpers are better than pull-ups, players will shoot better when open than when contested, and mid-range FGA produce far fewer points per possession than shots from 3 or at the hoop.
The analytics are clear on which shots are the most desired, but offenses can’t always get what they want. As the shot clock counts down, the likelihood of finding a good shot dwindles. A mid-range catch-and-shoot jumper from a capable shooter or a contested 3 might become a reasonable wager.
Thus, we cannot always say that a contested 3 is bad simply because an open 3 would be better. We can’t simply suggest that every open mid-range jumper should have been an attempt at the rim. Determining the quality of a particular shot just isn’t that simple.
However, there is a class of shots that is so foolish that it can be universally condemned.
Pull-up and contested 2-pointers from at least 10 ft. from the hoop and with at least 5 seconds on the shot clock are the antithesis of analytics. Pull-up, contested, and mid-range are an efficiency-decreasing super team. These shots are particularly asinine when they are taken with enough time on the shot clock for a kickout.
To avoid repeating the phrase, “pull-up, contested, 2-pointers from at least 10 ft. from the hoop and with at least 5 seconds on the shot clock” and because we have an unhealthy interest in concocted titles for the sake of acronyms, let’s call these choices “painfully ill-advised shots” or PIS.
Last season, players generated 0.767 points per shot on PIS. (All stats are courtesy of NBA.com.) That’s the equivalent of a 25.6% 3-point shot or 2 free throws from 38.4% free throw shooter. DeAndre Jordan had a FT% of 39.7 and teams intentionally sent him to the line.
At first glance it may appear as though certain players are efficient PIS takers. Tony Parker shot 50.8% on 118 PIS in 2014-15. Beno Udrih shot 49.6% on PIS in 2014-15. However, neither of these players had previous seasons that suggested their efficiency is sustainable. Parker shot 43.1% and Udrih had a PIS% of 39.3 (albeit on a small sample set) in 2013-14.
With only two seasons of the appropriate data to calculate PIS, we can’t say for certain that no players can consistently be efficient enough to make PIS a reasonable choice for the offense. However, preliminary results suggest not.
Not every player gets the same opportunity to take PIS. PIS is best understood as a percentage of a player’s shot selection. Let PIS selection or PISS be the percentage of a player’s FGA with at least 5 seconds on the shot clock that were PIS.
Most good players don’t often choose to take PIS. James Harden had a PISS of 10%, Steph Curry had a PISS of 8.5%, LeBron James had a PISS of 8.2%, Manu Ginobili had a PISS of 5.4%, and Demarre Carroll had a PISS of 2.7%.
Of course, there are the other players, players that have never seen a bad shot, or players with a seemingly uncontrollable urge to PISS.
If it were a contest, DeMar DeRozan would be the PISS champ.
DeRozan had a PISS of 31.6%. Nearly a third of his 850 shots with at least 5 seconds on the shot clock were PIS.
DeRozan shot an atrocious 33.5% on PIS. That equates to 0.67 points per shot. In all of his 2014-15 FGA, DeRozan scored 0.85 points per shot. If he simply removed those 269 PIS (kicked out instead of forcing a bad shot), he would have raised his efficiency to 0.92 points per shot on the season.
DeRozan’s team would have appreciated the kick outs. His team produced 1.02 points per shot. If his team kept that efficiency on kick outs that replaced DeRozan’s 269 PIS, the team would have scored another 93 points in DeRozan’s 60 games.
DeRozan’s PISS of 31.6% in 2014-15 was up from 26.2% the previous season. As the influence of analytics in the NBA grows, we might expect players’ PISS to decrease. In time, it probably will. In fact, DeRozan might be one of the last of a dying breed. Players that depend on PIS as significant portions of their offense were the norm in the 80s and 90s. In 5-10 years, they will likely be extinct.
However, suggesting that analytics are only pushing PISS down would be ignoring half the game. Analytically savvy teams are studying their opponents’ offensive weaknesses. They are recognizing when players like DeRozan will too easily take a PIS. Then, they are forcing that player to take a PIS. If you don’t believe me, listen to Shane Battier talk specifically about his defensive strategy to force Kobe to take a PIS. Click here to listen to Shane Battier on how analytics made him better (three and a half minutes).
In an interview, Battier revealed, “after studying and going through the school of analytics, I knew exactly to a tee who Kobe Bryant was. And I knew as a defender trying to stop him, Kobe’s worst-case scenario and my best-case scenario was to make him shoot a pull up jumper going to his left hand…”
So, DeRozan’s increase in PISS might be a result of both him not realizing the inefficiency of PIS and opposing defenses exploiting his lack of that understanding.
DeRozan is the PISS champ, but he’s not the only player taking too high of a percentage of these shots.
PISS appears to be more an intrinsic quality of the individual than a product of a team’s offensive system. Jarrett Jack, Shaun Livingston and Mo Williams have all changed teams over the last two seasons and seen their PISS follow them.
This doesn’t mean that teams can’t employ strategies to decrease PISS. Of course, teams can try to avoid acquiring players with a high PISS. Beyond that, teams can stress the importance of avoiding these shots.
Teams can support players by providing alternatives to a PIS. For example, the team can regularly station quality shooters on the perimeter and in the corners. This provides good options for a kick out on penetration. Good shooters on the perimeter will also keep help defenders out of the lane and increase the likelihood that the penetrating player will have a path all the way to the hoop.
Houston, Philadelphia and Atlanta had the lowest team PISS in 2014-15. That’s not surprising given that these teams are all recognized as being analytically savvy. Philadelphia’s position shows that teams do not need a tremendous amount of talent to have a low PISS. It’s an organizational decision based on an awareness of analytics.
For teams, the applications of this analysis are direct.
- Teams should be reluctant to acquire players with high PISS. This is especially true when that individual is weak in other areas (such as defense).
- Teams should look to make defensive adjustments that force players to take a PIS. This would mean running a player like Kobe Bryant off of the 3-point line while having the help defense under the hoop that will persuade Bryant to pull up in mid-range.
- Finally, teams should implement offensive systems that reduce PISS. This means stressing the team’s disinterest in this shot, but also playing the personnel and designing plays that regularly provide better opportunities.
NBA possessions are a valuable and limited resource. Players taking contested, mid-range, pull-up 2-pointers with at least 5 seconds on the shot clock are pissing it away.
About the Author, Stephen Shea
Stephen Shea is an associate professor of mathematics at Saint Anselm College in Manchester, NH. He earned a Ph.D. in mathematics from Wesleyan University in Middletown, CT, and a B.A. in mathematics from The College of the Holy Cross in Worcester, MA. His mathematical expertise and publication record is in the areas of probability, statistics, dynamical systems, and combinatorics. For years, he has been applying his abilities in these areas to study professional and amateur sports. Stephen is a managing partner of Advanced Metrics, LLC, a consulting company that provides analytics solutions to basketball and hockey organizations. At Saint Anselm College, he runs a course on sports analytics. His sport writing has been featured in the Journal of Quantitative Analysis in Sports, Psych Journal, the Expert Series at WinthropIntelligence.com, and the Stat Geek Idol Competition for TeamRankings.com. In 2013, Stephen coauthored the book, Basketball Analytics: Objective and Efficient Strategies for Understanding How Teams Win, and co-created the accompanying blog BasketballAnalyticsBook.com. In 2014, he authored Basketball Analytics: Spatial Tracking.
