Efficiency Isn’t Enough: Contract Language, True Shooting %, and Athlete Empowerment

Are You an NBA Player Who Missed A True Shooting % Performance Bonus? You May Be Entitled to Compensation!

The new age basketball adage of “3 > 2” has been a paramount, albeit reductive, hallmark of this new era of basketball. The death of the dead ball era of the early 2000s and the rise of volume three-point shooters like James Harden and Steph Curry to superstardom has brought forth a data driven era of analytics basketball. True shooting percentage (“TS%”) was one of the earliest “advanced” stats, with the goal of trying to account for the difference in value of different shots to provide a more accurate depiction of a player’s efficiency. However, the evolving analytics of modern basketball have slowly aged TS%, and it is only one of many examples of popular metrics becoming aged.

The reason aging metrics matter beyond a basketball perspective is that the calculation of these metrics plays a surprisingly big role off the court. Bonuses and conditions are a big part of NBA contracts. Players often have incentives for hitting statistical milestones, awards and achievements, and even injury and physical condition bonuses. Calculation of the salary cap is further complicated by either counting towards it as a likely bonus or not as an unlikely bonus. In terms of TS%, it has been commonly used as a bonus metric; for example, Cameron Johnson last season had a half a million bonus contingent on having above 60% TS%, Chris Paul had bonuses tied to Offensive Rating and TS%, while Moses Moody missed a bonus from TS% by shooting 57% instead of 60%. While half a million dollars may seem trivial relative to these player’s multimillion dollar contracts, the increased restrictions of the new CBA makes each dollar a critical squeeze.

The issue with TS% is in its calculation. The original formula is as follows (definitions for abbreviations are at the bottom of this article):

The logic is that three-pointers are worth 150% more than two-pointers, hence the 1.5, and one-point free throws are half of a field goal, hence the 0.5. The issue here is the coefficient of 0.44 within the denominator. Originally, that value was derived from the estimate of free throws in games that actually take up a possession; as and-1 free throws and technical free throws do not count as a possession, they are removed from the model. 

Shifting changes in the game have affected that variable; while the overall data from 2000-2025 confirms that variable, since 2019 that percentage of free throws that occupy a possession has dropped to only 38%. This is driven by an increase in both two-point and three-point and-1s, though slightly balanced out by an overall reduction in technical free throws. Pre 2019, the league averaged 2.23 two-point and-1s and 0.06 three-point and-1s per 100 possessions, which has jumped to 2.32 and 0.1, respectively, since (See here for pre 2019 and post 2019). This outweighed the slight drop in technical free throws per game from 0.61 to 0.5.  

While this change may seem minor, it does meaningfully affect the TS% of players. While players are still measured consistently to each other, contract bonuses may be built on previous expectations from expected values of past eras. If the goal of the bonuses is to incentivize players to perform better, then it stands to reason the metrics should accurately measure what it is trying to depict. While NBA teams are cognizant of this fact and likely incorporate a multitude of stats beyond TS% into their incentivization and calculus, it is still a minor disadvantage to current players and worth discussing.

Given that statistics now play a key role in contracts, player empowerment, and negotiation, I produced a calculation of how an updated TS% calculus would affect player metrics. Consider here that naturally, by reducing the free throw coefficient, the denominator of the formula decreases, thus all players will see a slight bump. However, certain archetypes will see a particular increase; for players and agents invested in using TS% as a performance metric or bonus metric, it is good to know where your archetype falls. The updated formula simply replaces the .44 with the correct .38 creating the below formula

Applying the new formula to players who have played since 2019 (where the new .38 value was derived from), and on average, players would see a 0.7% bump to their TS% with some seeing as high as 2%. Given all players received a bump, we also produce a measure that shows the size of their bump relative to league average. The graph above shows the distribution of relative increase against the final New TS%. We labelled the top 10 and bottom 10 players in terms of change. Exact change in value can be seen further below in table 1 and table 2.

While there is a wide range of player archetypes here, there does exist a common trend from the formula shift. Players who draw a high amount of shooting fouls and do not take a high amount of three-pointers got the largest benefits. As a result, rim running bags in the mould of Dwight Howard and Rudy Gobert saw the largest bump. Additionally, shifty crafty on ball players that can manipulate angles and draw fouls such as Jimmy Butler, Trae Young, and the reigning MVP Shai Gilgeous- Alexander also benefited. Unfortunately for many role players that work off ball as spot up shooters they will not enjoy as much benefit, as much of the lowest bumps were from players of that archetype. For the aforementioned Moses Moody, he would have missed out on his 60% mark even with the bump. 

Overall, the current TS% scheme claims to properly evaluate field goal and free throw attempts relative to their mathematical value but in truth it undervalues free throws which then slightly inflates the value of a three-pointer. This creates a non-trivial shift for most players, which can have significant impact on their individual bonuses and incentives and a team’s overall salary structure.

To be perfectly clear, a change in how TS% is calculated may not be a huge shift overall in terms of negotiations at a macro scale. While 1~2% change to hit performance metrics could have non-trivial contract and salary cap implications in specific cases, teams aren’t oblivious to issues in subjective metrics and likely already account for them in some ways; this can be using these metrics in reference to other players of similar archetype, or having a relative to league average benchmark. The shift we found here in TS% was non-trivial, but relatively un-damaging given the range of changes is not huge; the 1^st quartile is at 0.005%, while the 3^rd is at 0.01. However, TS% is only one of many examples of modern sports' advanced statistics that have less than objective valuations with shifting validity used as performance incentive benchmarks; for even more subjective examples, one can turn to popular football metrics in ESPN’s Quarterback rating (QBR) or Pro Football Focus’s PFF score. For negotiating parties taking these metrics at face value, it is of paramount importance to understand both what they claim to measure and how this measurement is done to best advocate for their interests. 

Table 1: Players With Highest TS% Increase Under New Calculus.

Name	TS% Increase	New TS%	TS%
Dwight Howard	2.70	67.50	64.79
Rudy Gobert	2.61	71.31	68.70
Jimmy Butler III	2.29	63.57	61.29
Joel Embiid	2.24	64.22	61.98
Mason Plumlee	2.17	66.02	63.85
Moussa Diabaté	2.04	63.02	60.98
Steven Adams	1.90	58.86	56.96
Dwight Powell	1.87	73.27	71.41
Danilo Gallinari	1.86	61.30	59.44
Drew Eubanks	1.81	65.29	63.49
Trae Young	1.80	59.81	58.00
Shai Gilgeous-Alexander	1.80	63.19	61.39
Mitchell Robinson	1.78	69.13	67.35
Jarrett Allen	1.76	69.27	67.52
Willy Hernangomez	1.66	60.25	58.60

Table 2: Players With Lowest TS% Increase Under New Calculus.

Name	TS% Increase	New TS%	TS%
Bryn Forbes	0.06	59.08	59.01
Julian Champagnie	0.08	58.83	58.75
Langston Galloway	0.11	58.26	58.16
Dean Wade	0.11	57.50	57.39
Danny Green	0.15	57.00	56.84
Terrence Ross	0.16	54.10	53.93
Doug McDermott	0.17	61.40	61.23
Thaddeus Young	0.19	56.05	55.87
Sam Hauser	0.19	62.53	62.34
Santi Aldama	0.20	57.24	57.05
T.J. Warren	0.20	59.89	59.69
T.J. McConnell	0.20	57.05	56.85

PTS: points scored

FTA: free throw attempts

FGA: field goal attempt

2PM: two point field goals made 

3PM: three point field goals made