Intro to Ben Taylor's Passing Metrics
In the ever present pursuit to breakdown the game of basketball, Ben Taylor, creator of Thinking Basketball (both a book he authored, and a Youtube channel in which he delves into deep analysis of basketball), has created several metrics to evaluate the offensive creation of an individual player; Box Creation, Offensive Load, and Passer Rating. These stats go deeper than what Ben exclaims as “Rondo Assists'', which he defines as a basic pass to a skilled scorer. It instead looks at how many opportunities an offensive hub produces (Box Creation), how involved they are in the offense (Offensive Load), and the rate in which they make quality passes (Passer Rating).
"Assists aren’t a great measure of offensive shot creation. They suffer from the problem of the “Rondo Assist,” when credit is given for passing to a good isolation scorer, or hitting teammates freed by a screen. These vanilla passes don’t tell us whether a player broke down the defense and created an open opportunity for a teammate by forcing help." - Ben Taylor
Box Creation is shot creation, determining how many open shots are generated by a player drawing multiple defenders to him. "an estimate for the number of open shots created for teammates (per 100 possessions)".
Offensive load measures the percentage in which a player is directly involved in scoring attempts and turnovers. Similar to usage rating, it tells us how centric a player is to a team's offense.
Passer Rating is an evaluation of a player's passing ability. Using the Box Score, Ben estimates the passing ability based on the percentage of good looks created, relative to the offensive load of the player
Potential Criticisms
Box Creation does not take into account defenses responding to multiple players within an action. It also values lay-ups higher than it does mid-range jumpers.
Offensive load is a pretty sound concept, and is the best quantifiable measure for how much a player contributes to their offense. One takeaway could see this as a metric not indicative of winning, but this goes into the deeper topic of whether or not it is good for one player to be responsible for the majority of offensive decisions. Comparing it with time of possession could better determine how quickly the offensive decisions are made.
Passer Rating tends to overrate taller conservative guards, as the higher point of view leads to easier open looks over defenses. It also devalues high volume scorers, since more shot attempts means less passes. Finally, it does not distinguish what kind of passes a player excels at, instead it compresses quality passes from pick and roll, transition, post, and others.
Beyond the Metrics
In a video breaking down Lebron James' passing, Ben listed four attributes of James' passing to exemplify him as one of the greatest passers of all time. He hopes this will guide fans into analyzing players on their own, without the need of numbers. The attributes are as follows:
1. Directionality: A threat to pass in every direction at all times.
2. Speed: Quick release and high velocity deliveries.
3. Anticipation: Throws before a player is open.
4. Manipulation: Head movement, no-look, body language, shot fake.
Calling back to the old-school philosophy of the "eye test", these qualities can be observed at any level of the game, and a truly gifted passer will excel in all four. It is important to commend the pure showmanship of electric playmaking. Masterful playmaking requires a high degree of basketball IQ, hand-eye-coordination, spatial awareness, anticipation, and confidence.
I would also like to add the wrinkle of height into Ben’s musing on passing. Taller players offer the unique advantage of having a higher vantage point than their peers. Nikola Jokic (who he’s also done extensive content on) is considered to be the best passer in the league if not the best passer of all time. While an incredible soft touch and super-computer like processing ability makes him an apex playmaker I would argue his height combined with skills, and IQ that take him over the edge. The pure geometry of the court is simply different in his eyes, and he can whip balls into passing angles smaller playmakers can’t see.
It goes without saying passing is a complicated and nuanced skill, which is why it needs statistical and visual analyzation to contextualize its impact.
Box Creation tracks opportunities created in the presence of a drawn defense. In the formula Ben uses assists, shooting, and scoring volume via usage, which is defined as: points (per 100) plus turnovers (per 100). Points are used instead of attempts to better reflect a successful offense. This combined with a high assist total approximates creation. To better quantify efficient 3pt shooting with high volume, Ben created 3pt proficiency. Better illustrating an effective playmaker to a roleplayer, as a lethal shooter would draw more defensive pressure leading to more opportunities created. The formula is as follows: Creation = Ast*0.1843+(Pts+TOV)*0.0969-2.3021*(3pt proficiency)+0.0582*(Ast*(Pts+TOV)*3pt proficiency)-1.1942 3pt proficiency = (2/(1+EXP(-3PA))-1)*3P% Offensive Load
Offensive Load determines an individual player's contribution to a single possession. This is an evolution of usage rate which covers percentage of possessions where a player registers a scoring attempt or turnover. This does not take into account shot creation, which is where Box Creation fills the gap. Therefore Offensive Load is broken down to: true shot attempts, a creation estimate, turnovers and non-creation assists. Using per 100 data, the final formula is:
Offensive Load: (Assists-(0.38*Box Creation))*0.75)+FGA+FTA*0.44+Box Creation+Turnovers Passer Rating
Passer Rating attempts to measure the overall passing ability of a player. This is broken down through a high rate of assists relative to offensive load, percentage of lay-up assists, non-3pt creation to load ratio, height, and turnovers.
The formula is as follows: Passer Rating = 1/(1+e^(-1*LOG(2.5)*((4.79+6.48*Layup%+0.19*Load-76.71*Ast:Load-0.07*Height-0.08*TOV^2-2.71*(non-3 creation:ast)+1.19*(Height*Ast:Load))-5.6)))*11.3
Classifying Turnovers
Turnovers like assists can have different values that better illustrate how detrimental said turnover was to the team. Using offensive load we can further clarify these types of turnovers.
Scoring attempt: Prevents the individual attempt which varies on the respective player
Idle pass: An errant pass early in the shot clock, this wastes a whole possession.
High risk pass: Higher degree pass usually with contested arms in the passing lane. This essentially trades in missed shots for turnovers, but in turn yields more production, particularly a high amount of lay-ups.
Using this we can see the distribution of these turnovers can differentiate someone as more of an ISO scorer (higher score attempts and idle passes), and a playmaker (Smaller of the former variables and more high risk passes). Despite a higher turnover ratio, a playmaker creates more quality looks for teammates, which in turn cuts down on turnover ratio across the team.
In Summary
Box Creation estimates shots created for teammates, encapsulating on and off-ball creation, but not all creation is equal.
Offensive Load expands upon Usage Rating by adding in a playmaking factor.
Passer Rating estimates passing ability through rate of successful attempts and volume of attempts.
Different assists and turnovers have varying impacts. A higher amount of turnovers might stem from more quality offense, while in turn a high turnover ratio might indicate a more conservative offense.
In closing, I would urge everyone to delve deeper into the mind of Ben Taylor, as his ability to analyze and make high quality basketball content is astounding.
Source
This document is based on the video Beyond Assists