Nylon Calculus: Why Chris Paul and James Harden are so tough in the penalty

A few days ago, I wrote about how long players play with the penalty/bonus to their advantage or disadvantage. That was a carryover from team-related statistics split up by how many fouls teams had left to give.

Splitting player (and team) minutes by accounting for how many fouls their team and opponent have to give is important, simply because it showed that not all 48 minutes are equal. On average, about one-fourth of each game involves teams in the penalty, also called the bonus, after committing either four team fouls in one quarter or one in the final two minutes. When that happened last season, teams scored about 5 points per 100 possessions more efficiently when their opponents run out of fouls to give. When one team had fouls to give and the other didn’t, the former team had a net rating of +3.9 per 100 possessions.

The main reason for the bump in efficiency and/or net rating is because free throw rates nearly double in that situation. That’s not surprising when the reward from getting a team to defend in the penalty is two free throws for every team foul. There are minor changes across most statistics, though, likely because of end-of-quarter basketball is more about possession-hoarding. Below is a look at how several statistics changed last season once opponents had no fouls to give:

However, those were team averages. While they’re the same across the board for players, there was a wider range of results here. By splitting scoring efficiency for players by whether or not their opponents had fouls left to give, we’ll see a number of things.

For example, some players clearly benefit once they can get to the free throw line on every team foul. Bouncing off that, we can use these splits to, among other things, find the best candidates on multi-star teams to play with bench-heavy units in a staggered substitution pattern. If a starter shows little change in efficiency when in and out of the penalty/bonus, it would be wise to sub him out before the end of first and third quarters, and start him in second and fourth quarters. We can also see how ridiculous some free throw rates become once the Hack-a-Shaq situation is in effect and whether or not the rule changes last season had any effect.

Let’s start with a player example for how to read the splits. One of my favorite players in these results was Chris Paul. Below is a look at his career shooting splits in and out of the penalty/bonus over the years. The full dashboard includes field goal, three-point, and free throw percentages, too:

Paul has benefitted from playing in the penalty/bonus more than the average player. While he’s typically been a better shooter in the beginning and middle of quarters, he’s rarely ever been to the free throw line during those phases of the game. That makes some sense. Paul’s drawn fouls not from sheer aggression around the paint like the Russell Westbrook types, but more from exploiting perimeter defenders when they’re out of position. Some might call these fouls junky, others crafty.

The rip-through foul is one example, a foul that went from a shooting one to an inbound from the sideline several years ago. But once opponents have no fouls to give, those types of fouls lead to an automatic two free throws, and that’s when Paul’s free throw rate has taken off. While fellow teammate James Harden might get to the line more frequently overall, Paul’s free throw rate in the penalty/bonus has been higher than Harden’s since the latter player entered the league.

Below is a comp in free throw rates since 2013, when Harden first played for the Houston Rockets.

The huge increases in Paul’s free throw rates in the penalty/bonus have led to nice jumps in true shooting percentage, often going from slightly above-average in efficiency when opponents have fouls to give to one of the most efficient scorers in the league once every opposing team foul committed leads to two free throws. Paul’s also been less turnover prone when opponents have run out of fouls to give, nearly trimming his turnover rate in half last season.

The chart of Paul and Harden’s free throw rates also suggests that if Houston plans on having at least one of them on the floor at all times, Harden should be the player to sub out first and lead bench-heavy units. Not only does Harden have the higher free throw rate when his teams are out of the penalty, his overall efficiency appears less sensitive to his team’s foul situation.

Below is a comparison of Paul and Harden’s true shooting, effective field goal shooting, free throw rate, and turnover rate in and out of the penalty since 2013:

Over a decent sample size, Harden appears more likely to remain an efficient player, even if he plays less often when the Rockets have the penalty to their advantage. But it makes sense to have this sub pattern even without these splits. Harden has the ability to carry an insane amount of the offense, which is fitting for the beginning of second and fourth quarters. Paul’s defense would also be wasted against second-string point guards.

In an ideal NBA where finishing games matters more than starting them, the Chris Pauls of the league who are on the floor a ton, but clearly benefit from the penalty/bonus would sit out the first three to four minutes of each quarter. Using this subbing pattern, Paul would likely play about 33 to 40 percent of his minutes with opposing teams having no fouls left to give, compared to the 28 percent I calculated from last season, and become more efficient than he already is by turning a clever foul or two into more free throws instead of inbound plays.

Splitting scoring statistics by foul situation also proved that the Hack-A-Shaq tweaks made last season worked. The main targets of the Hack-A-Shaq were DeAndre Jordan, Andre Drummond, and Dwight Howard, all seeing significant increases in their free throw rates in 2016 when opposing teams had no fouls left to give, and mostly because of the rise of intentional fouling that season. But in 2017, they all saw decreases back to where they were in 2015 or 2014.

While these players and others are still vulnerable to intentional fouling, their free throw rates are back at acceptable levels. They’re back to when intentional fouling was in the playbook for opponents, but not a subject to discuss on what felt like every national televised broadcast when those players were on the floor.

On the opposite of the intentional fouling candidates are the type of players who often sees little change from playing in the penalty/bonus. This is typically the spot-up shooter. That’s not surprising given how few fouls most of them draw, though some who are exceptional free throw shooters become go-to options when late-game fouling happens.

There are a variety of other ways to use scoring statistics split by foul situation. Maybe a player’s true shooting percentage, or how frequently he gets to the line without the penalty/bonus in their favor is a better way to evaluate their scoring efficiency or foul-drawing, respectively. As with other statistics, though, sample size matters. Some players can go a whole season scoring remarkable less efficiently with the penalty in their favor because the sample sizes can be pretty small. Players could also suffer from being a scoring option at the end-of-quarters, helping their turnover rate in inefficient isolation plays, but harming their shooting statistics.

Also, something possibly awkward in the dashboard I published was that because I was unable to get penalty time outside of last season, I didn’t include a column for player minutes. If that’s a problem, hit me up on Twitter and I’ll work total minutes into the dashboard. For now, it did not seem necessary mixing in total minutes with stats that only happen in certain phases of each game. Instead, I made the typical columns for shots, trips to the free throw line, and turnovers, but also ones adding up all of them titled as total offense possessions. The dashboard below is sorted by total offense possessions, but there’s also shot possessions which add just shots and free throws.

Hopefully this all made some sense. Below is a dashboard with scoring statistics split by whether or not opponents had fouls left to give. If you prefer more width, check out this dashboard:

Stay tuned for more penalty time analysis to come.