Author: Cameron Van, J.D.

Contributing Author:
Cameron Van, J.D.
University of California, Davis School of Law, Davis CA

400 Mrak Hall Drive, Davis, CA 95616
Phone Number: (650) 740-2235

Cameron Van is a recent UC Davis School of Law Graduate with a focus on the intersection of business and the law.


This article offers the NCAA a reputable, repeatable, and reasonable formula for a student-athlete revenue scheme that will ensure its competitive edge in an ever-encroaching market. The NCAA uses amateurism to restrict artificially the compensation of student athletes’ compensation to “cost of tuition,” at best. It is precisely this reason that more athletes are finding alternative ways to capitalize on their talents. As a result, this amateurism scheme is not Pareto Efficient. Pareto efficiency is reached when a situation cannot be modified in a way that would have one party better off without making another party worse off. Notably, Pareto efficiency does not imply equality, equity, or fairness, rather simply that there could be no economic changes that would better off the overall system. Here, this article explores a rare occurrence where the system can be made both more efficient and equal by increasing the supply of revenue generators – the athletes. This article will build upon Stocz formula for deriving a student-athlete’s salary, as well as give examples of what such a salary would look like for said athletes.

Key Words: NCAA, student-athlete, compensation, tuition, salary


The NCAA reports that due to the championship television marketing rights, Division I Men’s Basketball alone brings in $867.5 million annually (30). This Championship, fondly known as “March Madness,” is a tournament comprised of 68 teams with play spanning over 21 days. March Madness brings a promise of the national spotlight, social media trending, and a draft pick by any of the 30 National Basketball Association’s teams for the athlete. However, players have a better chance of leveraging social media to become the next big influencer, than being drafted with the NCAA reporting Men’s Basketball players have a 1.2% chance to make it to the big-leagues (6). So, if there is no promise of future pay, and their efforts, at least in part, help generate $867.5 million in revenue not to mention, attract students to attend their respective college, it is only natural that a player would want to receive some form of compensation.

Athletes have felt so strongly about this idea that some have taken the NCAA to court– the most famous being that of Ed O’Bannon v. NCAA. As it stands, the ruling in O’Bannon means that by giving full cost of attendance, the NCAA cured their antitrust harm caused by its otherwise unlawful amateurism rules (21). Since the remedy is already commonplace, the 9th circuit’s ruling does not impose any change upon the NCAA rules. More cases have budded in this area after O’Bannon; however, none has mandated student athletes must be paid.

There is a clear movement among athletes, whether inside or out of the NCAA, to push for a more equitable sharing of the generated revenues. Governor Gavin Newsom signed the “Fair pay to Play Act”, a California statute, in September of 2019 on Lebron James’s television show. The Act allows NCAA athletes to obtain endorsements and sponsorships while still maintaining athletic eligibility and is indicative of the, albeit slow, movement towards NCAA athlete compensation. More specifically, The Act forbids California Colleges from retaliating against athletes whom gain compensation for their use of their name, image, and likeness (NILs) (7). It also allows athletes to retain agents and other representatives to navigate potential commercial opportunities that The Act affords. While The Act does not require schools to pay their athletes, it opens the door for athletes to retain and receive compensation for the use of their NILs by creating a statutory right for college athletes. The Act will not go into effect until January 1, 2023.

The NCAA should recognize the “handwriting on the wall” and act accordingly, or the league will not only lose players to international teams, but to NBA development “G League” teams in the United States. The NBA understands the value of these athletes and is making moves to entice these players with its new NBA G-League Pathway Program. This program affords select “elite prospects” to play in the NBA G League even though they are not yet eligible for the NBA, since they are not one year removed from high school (18). At its creation, it has allowed G League teams to offer $125,000 “Select Contracts” for the five month 2019-20 season. However, despite this new initiative, top prospects like Lamelo Ball still opted to play internationally in leagues such as the Australian National Basketball League (4). The NBA responded by increasing the starting salary to more than $500,000 a year, which, as of April 28th 2020, has proved enough to entice multiple 5-star recruits including Jalen Green, Isaiah Todd, and Daishen Nix (39). Notably, Nix decommitted from UCLA in favor of joining the G-League. This movement away from NCAA in favor of the G league will only strengthen with time if the NCAA continues to fail to modernize as the NBA has done.

To the NCAA’s credit, just days after these student athletes made their move to the G League; the NCAA’s Board of Governors hinted that they would support a proposal to allow athletes to retain their NILs (30). This change would mean athletes could receive payments from third-party endorsements, social media influencing, their own products or business activities, personal promotions, and notably no caps on earning. However, this proposal comes with key restrictions, namely that schools and their boosters, cannot help facilitate endorsements or be otherwise involved. Moreover, the new rules, if proposed, would not be drafted until October, voted on in January 2020, and likely enacted the following school year. Nevertheless, the NCAA has recognized the movement and is trying to act accordingly, but it is not enough to stay the movement.

As a result, the NCAA should move forward, so as  not to cause “a patchwork of different laws from different states [that] will make unattainable the goal of providing a fair and level playing field for 1,100 campuses and nearly half a million student-athletes nationwide” (33). The NCAA should move towards adopting both a more generous NIL rule that mirrors The Act, as well as a wage compensation for student athletes.

This paper proposes a student athlete compensation scheme that takes into account the student athlete’s contribution to the sport, as well as the sport’s contribution to the school’s overall revenue inflow.  Part I provides an overview of the NCAA. Part II discusses three different compensation models for NCAA athletes, along with their relative advantages and disadvantages.  The current amateur model limits compensation to tuition only, then two alternative compensation models: the Olympic model and the revenue-sharing model are presented. The argument for revenue sharing.  Implementation of either of the two alternative compensation models would necessitate a workable athlete payment formula and Part III offers an argument for the revenue sharing scheme. Part IV presents the Stocz payment formula. Part V presents a data analysis of analysis of NBA statistics and feed it into the Stocz model to calculate sample NCAA athlete payments.  Part VI offers an application to sports.

NCAA Overview

The NCAA lauds itself as a member-led organization that is “dedicated to the well-being and lifelong success of college athletes” (37). The NCAA is a non-for-profit organization that oversees 1,090 colleges and universities in every state, amounting to almost half a million college athletes and 19,886 teams. While the NCAA is widely regarded to be the dominant college sports organization, there are other similar organizations such as the National Association of Intercollegiate athletics (26). The NCAA asserts that the distinction between collegiate and professional athletics is the for-profit nature, whereas collegiate athletics focus on the educational benefit of the athlete, student morale, campus public relations, institutional profile, fundraising, and student physical fitness (30). The NCAA is divided into three divisions, I, II, and III, with 350, 310, and 438 institutions respectively.

Compensation Schemes

There has been much scholarship on the both the merits of and best methodology for NCAA student athlete compensation. Joshua Senne adeptly laid out the basic framework and relevant issues regarding paying student athletes (30).

Current System (Cost of Attendance): Amateurism

In “The Business of Sports” Kenneth Shropshire dispels the widely held incorrect belief that the NCAA’s foundation of amateurism stems from the ancient Greek’s Olympics (31). Shropshire explains that it is but a “myth” that the ancient Greeks participated in sports for glory, rather than for receiving compensation for competing or winning. Moreover, Shropshire shows that there is no mention of amateurism in Greek sources or any other ancient sources in which the current amateurism system can be tied (31). Instead, Shropshire argues that this myth was propagated by certain scholars who would directly benefit from an amateurism system.

The first published mention of the word “amateur” was in an 1866 publication from the Amateur Athletic Club of England, though the term was likely used before (31). Shropshire argues that the prevailing belief of the time was that those whom competed in sports for money were of questionable character. Further, the working class was not allowed to compete in sports because they had developed muscles, which were deemed an unfair advantage due to their workman-like lifestyle. Should an amateur athlete lose to a laborer, said athlete would lose their status in society. As a result, this belief manifested in a “mechanics clause” which held that if someone used muscles as part of their employment, and then they had an unfair competitive advantage (31).

Taylor Branch in her Atlantic article “The Shame of College Sports”, reveals the true origins of the NCAA’s amateurism rules (2). According to Branch, the head of the NCAA Walter Byers “crafted the term student-athlete” and “embedded in all NCAA rules and interpretations” the term in a concerted attempt to defend against being held liable for workers compensation benefits to athletes injured during competition (2).

The term student-athlete was deliberately ambiguous. College players were not students at play (which might understate their athletic obligations), nor were they just athletes in college (which might imply they were professionals). That they were high-performance athletes meant they could be forgiven for not meeting the academic standards of their peers; that they were students meant they did not have to be compensated, ever, for anything more than the cost of their studies. Student-athlete became the NCAA’s signature term, repeated constantly in and out of courtrooms.

Branch outlines how the NCAA has won a plethora of liability cases using “student-athlete” as its main defense, most notably, in an account of Texan Christian University’s (TCU) Kent Waldrep in a game against Alabama Crimson Tide in 1974. The TCU running back completed the “Red Right 28” sweep towards the sideline. Waldrep was hit hard by a litany of defenders, and later awoke in the hospital. Waldrep survived but lost all movement and feeling below his neck. TCU stopped paying his medical bills after 9 months, so his family turned to charity.

Waldrep sued for workers’ compensation, which the appeals court ultimately rejected in June of 2000 (37). The court ruled that Waldrep was not an employee because he did not pay taxes on his financial aid that he could have maintained irrespective of whether he stayed on the team. Branch argues that this case only emboldened the NCAA to use their “student-athlete” designation as both a liability shield and a “noble ideal” (2). Simply put, the designation, Branch argues, was originally crafted to keep those carried off the field from bringing their schools into the courtroom. 

Olympic Model

Alex Moyer outlines the merits of the Olympic Model applied to the NCAA in his George Washington Law Review article (16). Moyer argues that this model would guarantee student-athletes cost of attendance, scholarships, and free access to the commercial market by allowing endorsement deals and other benefits. Moyer argues that under the Olympic model, athletes should not be paid for their participation, since doing so would “harm much of the differentiation between college and professional sports that contributes to its success.” Next, schools should mandatorily cover full cost of tuition, which Moyer argues would gradually increase the price of student-athlete “labor” and cure the current rules’ anticompetitive effects. He argues that this regime would excise the “shortfall” that keeps students impoverished (16).

Notably, Moyer argues that the NCAA should allow institutions to pay beyond scholarships based on licensing revenues. He places three conditionals upon these payments 1) the payments need to be an immediate stipend and not a deferred trust, 2) the minimum cap for additional payments should be raised from $5,000 to $10,000, and 3) that the schools should be able to offer whatever they see fit to different recruits, rather than having to provide the same payout level (16). This ensures, through price competition, that the student is being compensated for their specific value.

Lastly, Moyer indicates that the NCAA should allow student-athletes to profit from the use of their NILs. He argues that this would cure anti-trust issues, citing Bylaw as a specific antitrust issue, and should not hurt NCAA’s popularity. Moyer notes that this scheme would tend towards benefiting star, and mostly male, athletes. However, he argues that this is what the free-market demands, as this would simply reflect, “their current contributions to the profitable system, and the new rule allows those players to also enjoy the benefits of their value” (16). Moyer calls the Olympic model an efficient system of fair governance and just compensation. However, the Olympic System does not share any of the revenue their institutions make via the athletes’ efforts, making it an inequitable system just as the current one. It also does not share workers’ compensation, which is essential for collision sports e.g. football.

Revenue Sharing Model

There has been much scholarship on student athlete compensation, but not much consensus. Some scholars like Thomas R. Hurst argue that NCAA should allow DI institutions to afford small sums of money, which he calls “laundry money”, ranging from $30-$50 per month in addition to their scholarships (10). He pushes further for a revenue sharing scheme based on team success, advising, “The university should pay an equal amount to each female student-athlete in all sports,” to avoid Title IX issues. Hurst argues, “Any money paid to student-athletes is derived directly from revenue.” He further suggests that the athletes should only receive 10-15% of the revenue as an artificial “salary cap”, and that each university should have the choice whether to opt in or not into this system. He further advises that in order for the NCAA to continue having its tax-exempt status in light of the institutions sharing revenue with the students, “it is likely that the NCAA could finance a lobbying group to preclude the IRS from categorizing athletic revenue as “unrelated business income” (10). While Hurst outlines how the NCAA could avoid legal issues including Title IX, worker’s compensation, vicarious liability, taxation, and Anti-Trust, he falls short of giving the universities a way of determining how to pay each player. In the next section, this paper will outline the argument for a revenue sharing scheme with a plug-in-able formula.

Argument for Revenue Sharing

This paper will provide a methodology to institute a revenue sharing scheme within the NCAA. Given the growing momentum for student compensation, the change towards revenue sharing seems closer than ever. Moreover, the NCAA stands to increase Pareto efficiency by enacting such a scheme. Pareto efficiency, or Pareto optimality, is an economic situation where resources cannot be moved in order to make a person better off without making at least one other person worse off. Pareto efficiency only suggests optimality of recourse allocation and says nothing about a system’s equity, or fairness (23).

The right revenue sharing scheme affords the NCAA a rare moment where it cannot only promote Pareto efficiency, but also equity, bolstering the popular opinion on which it thrives. The NCAA’s current rules artificially depress the supply of revenue generators by forbidding athletes from compensation of any kind other than cost of attendance. Despite these constraints, the NCAA still brings more than a billion dollars per year in revenue (27). The market for student-athlete endorsements is apparent, yet remains suppressed by the current rules. Moreover, this billion-dollar industry is directly due to student athlete’s yearly contributions on the field. And it is precisely this reason that institutions will spend lavishly on facilities in an attempt to woo the next year’s top talent, like Clemson Football’s complex complete with laser tag, sand volleyball, a movie theatre, bowling lanes, and a miniature golf course (12). Instead, institutions could use their money more effectively by sharing their revenue with the student athletes, both incentivizing them as stakeholders of their shared enterprise, as well as swaying athletes back to the NCAA from other professional leagues domestic and international. Finally, such a scheme is likely to improve public perception, which in an industry that monetizes viewership; this would prove invaluable for the long-term success of the NCAA. While a revenue sharing scheme does entail sharing some of the revenue pie with the student athletes, if done right, the entire pie stands to grow, bettering everyone in the end with true Pareto efficiency and equity.


Stocz Payment Formula

In order to derive an adequate formula, this paper will draw heavily from Mike Stocz’s 2019 article published in the Journal of Higher Education Athletics & Innovation. Stocz’s article lays out a usable compensation model for all NCAA athletes, not just DI or even Power Five Conference schools. Stocz’s article, inspired by the Ed O’Bannon case, proposes a model that is based on 1) a number of metrics to discern a player’s individual compensation, and 2) a 14% “loading factor” applied to said compensation total, which Stocz identified as the exact percentage of revenue used to cover the student-athletes cost of attendance (31). In his formula, 14% of the revenue generated by a University would go towards covering a student athlete’s college expense (CE). CE includes tuition, books, room and board, and other costs like laundry. While Stocz advocated for deleting the cost of attendance payment and simply paying outright athletes 14% of the overall revenue, comparable organizations like the NBA, NFL, and MLB pay their players 50% (17), up to 48.8% (29), and about 50% (3) respectively. In light of the comparable league’s revenue sharing plans, this article recommends increasing that payment to an even 20%. Players have a large economic impact on not only their institutions, but also their institution’s sponsors. As a result, they should see a larger share that reflects their billion-dollar impact.

Available revenue per team

This new 20% of the total revenue generated by each individual sport within a school will be dispersed through a new account, rather than towards scholarships and other CE expenses, which Stocz calls a Student-Athlete Fund [SAF]. The total revenue [TRSn] is captured by combining ticket sales, sponsors, and television revenue. The following equation represents the SAF generation, with different sports represented by S1-3.

SAF = S((0.2(TRS1) + 0.2(TRS2) + 0.2(TRS3)… 0.2(TRSn))  

Next, the SAF disbursement will be modeled after the MLB (1), where each team within their institution will receive a disbursement based upon their success. Here, teams would receive a Percentage of Total Revenue [PTR], of the SAF based upon the total revenue they accrue. The total percentage of a team’s claim PTR will be capped at 33% to ensure that the revenue is being disbursed to other teams similar to the MLB revenue sharing rules. Furthermore, the PTR is derived by adding the TR and revenue not related to specific teams [RNRST like donations, symbolized by SRV, and ultimately divided by the grand total of revenue [GTR]. The equation of the individual team’s percentage of total revenue is as follows.


Next, we calculate the non-adjusted totals (NAT) which represent a benchmark for each team’s eventual student-athlete fund for their particular team. A team’s NAT is calculated by multiplying their individual PTR by the overall SAF.


After calculating the NAT, we subtract the difference between the NAT and the adjusted total [AT]. This is to determine the amount of funds that will be redistributed to the teams below the 33% threshold. These funds going to the sub 33% teams is symbolized by GA.

GA = NAT1 – AT1
AT = (PTR33%)*(SAF)  

The GA is added to all teams below the 33% threshold for a new added total [ADT]. The total amount of the redistributed funds is then determined in part by the number of athletes [T1A] on a given team. Therefore, the total number of athletes on teams with a PTR below the 33% threshold is the BTA. Next GA is divided by the obtained BTA in ordered to derive the multiple player additive [MPA], which is next multiply the result by the number of athletes on the team. This total will be added to each team, and is symbolized by ADM. For teams that are above the 33% threshold, they are not eligible for an adjustment as their formula is simply PTR<33% team = ADT33%.

BTA = S(T1A 33%>PTR)+( T2A 33%>PTR) … + ( TnA 33%>PTR)

Player Salary Cap formula

Stocz derives a formula for the individual athlete’s pay. However, for this Stocz only derives a player maximum salary, effectively instituting a per team salary cap, while also recognizing that not all players are equal in value to a team, and the market demands excellence for top dollar. While arguably artificial and anti-free market, this methodology represents most professional leagues with both team and player salary caps. This player salary cap [PCA] is explicitly the maximum a player can earn in a specific team. The sum of a team’s athletes is symbolized by TA.

PCA1= ADT1/(T1A);  PCA2= ADT2/(T2A) ; … PCAMen’sbasketball=

Stocz has these funds only available for the athletes who played in the current season of their sport, with funds being disbursed directly after the season. He also wants to give 14% of each athlete’s salary cap [PCA] to the athletes outright, while the remaining 86% will be put in a trust that will be rewarded after the athlete completes their bachelor, or graduate, degree. This paper differs slightly, suggesting that the sum of the cost of attendance [CE] is disbursed before the school year, at minimum, so these student-athletes can pay tuition, which is due before classes, and any seasons are completed. The remainder will be put in a trust, so even if the athlete does leave early to play in their sport’s professional league, and even those who do not, are incentivized to graduate. Table 1 displays the model applied to the Duke University based on their revenue and student athlete participation data from the U.S. Department of Education’s Equity in Athletics Data Analysis (EADA).

Table 1.

Duke University Team Revenue PTR ADT PCA
Men’s Basketball $ 35,489,891.00 32% $ 6,579,772.56 $ 469,983.75
Woman’s Basketball $ 3,821,554.00 4% $ 779,669.09 $ 55,690.65
Football $ 37,844,852.00 34%   $ 63,539.09
Baseball $ 1,797,242.00 2% $ 415,052.08 $ 10,922.42
Men’s fencing $ 111,832.00 0.5% $ 102,785.76 $ 4,282.74
Women’s Fencing $ 618,765.00 1% $ 194,863.95 $ 9,279.24
Field Hockey $ 1,588,297.00 2% $ 372,435.19 $ 17,735.01
Men’s Golf $ 1,764,573.00 2% $ 401,395.23 $ 50,174.40
Women’s Golf $ 845,454.00 1% $ 232,801.42 $ 33,257.35
Men’s LAX $ 2,043,512.00 2% $ 462,458.84 $ 9,839.55
Women’s LAX $ 1,657,266.00 2% $ 387,624.79 $ 12,504.03
Women’s Rowing $ 2,423,745.00 3% $ 532,099.09 $ 11,321.26
Men’s Soccer $ 1,422,891.00 2% $ 343,931.31 $ 12,283.26
Women’s Soccer $ 2,042,379.00 2% $ 455,856.77 $ 20,720.76
Women’s Softball $ 1,587,447.00 2% $ 372,023.73 $ 18,601.19
Men’s swimming and diving $ 81,051.00 0.4% $ 96,892.39 $ 4,212.71
Women’s Swimming and diving $ 1,844,500.00 2% $ 420,893.85 $ 15,588.66
Men’s Tennis $ 638,367.00 1% $ 195,640.48 $ 19,564.05
Women’s Tennis $ 1,167,968.00 1% $ 292,126.13 $ 36,515.77
Men’s Track and Field and X-country $ 1,228,415.00 1% $ 316,753.56 $ 5,192.68
Women’s Track and Field and X-country $ 2,294,247.00 2% $ 511,962.29 $ 8,392.82
Volleyball $ 1,562,909.00 2% $ 366,506.43 $ 22,906.65
Wrestling $ 95,901.00 0.5% $ 98,844.84 $ 4,942.24
Total $ 103,973,058.00   Student Athletic Fund $ 20,794,611.60
Non-specific sports rev $ 9,564,704.00   Grant Total Rev $ 113,537,762.00

Duke was chosen as a model because it has a history of success when it comes to basketball in particular, making it a prime candidate for analysis (24). It is important to note that the revenue available was of the 2018 fiscal year, meaning that any fiscal data that may reflect the aforementioned “Zion” effect is unavailable. Therefore, it is likely that Duke’s numbers would be greatly inflated in such a historic year. Nevertheless, the basketball players are unsurprisingly the top earners, receiving $469,983.75 for their efforts. Again, this is only if the athletes were to receive 20 percent of the overall available sports revenue. For perspective, on average Duke’s head coaches received $888,921 for men’s teams and $217,429 for women’s teams respectively (35).

Deriving an Athlete’s individual Payout

In order to derive the Player Compensation Coefficient [PCE], Stocz derived the below formula.

PCE = SV + TM + IS

For determining an athlete’s individual payment from each sport team, Stocz uses three factors Sport Visibility (SV, 33%), Team Success (TM, 34%), and Individual Statistics (IS, 33%). Each factor would need to be individualized per sport, and this article suggests what it would look like for a basketball team. SV is perhaps the most difficult to capture adequately. SV could include ticket sales, number of sellouts, nationally televised games, social media mentions and trendings, and an estimated reach of an individual team compared to other teams within a said sport. This could encourage teams to grow their own social media presence, adding potential viewership, bringing in more participants into the market, ultimately increasing the overall sum of funds. However, it is worth noting that trending on social media does not always lead to revenue and is not always a good thing in general, so the social media component needs to be weighed well by the institutional officials. Next, TM is derived objective accomplishment factors such as, winning record or overall record, conference tournament appearances and championships, NCAA tournaments and championships, and number of All-Americans and number of Academic All-Americans.

Lastly, Stocz derives IS by weighing different individual player statistics, which naturally would vary by sport. Instead of simply choosing randomly which statistics are included and which are not, this article suggests using the statistically significant individual statistics [SSIS].


If the PCE coefficient is equal to 1, then that athlete will be receiving the maximum salary that a player in their specific sport can attain given the cap. The final result will be multiplied by each year’s player cap (PCA) in order to finally determine exactly how much the NCAA will be compensating specific athletes given their efforts. Stocz left the door open for either the NCAA, or individual institutions to determine each sport’s PCE within a given institution. This article suggests that the NCAA keep its responsibilities to a minimum by allowing the institutions to derive their own PCE, this will both divert legal responsibility from NCAA to the institutions, and afford aspiring sports statisticians’ gainful employment within an area of interest with their institution. That being said, if the institution is to decide, when looking at the statistics, a school may inherently value assists over scoring outright, promoting a more team first ethos. If so, institutions should be forthright about which individual statistics they input in their salary derivation. This way, athletes will know how their salary is being derived before, during, and after their season. This is why it is important that athletes also obtain the right to retain an agent so that the inner workings of the salary cap can be explained by an unbiased third party. The total athlete compensation is symbolized by AC.

AC1 = (PCE1)*(PCA1); AC2 = (PCE2)*(PCA2); … ACn = (PCEn)*(PCAn)  

Using the NBA as a Model to Derive PCE

In an effort to show which individual statistics an institution should weigh when deriving an athlete’s individual value, this paper uses the NBA as a proxy. The regression below was obtained via Stata, by amassing the salary of 523 players in the 2018-19 NBA season, as well as their conventional statistics, advanced analytics, and other variables amounting in 76 in total. These variables were regressed on Salary, omitting variables that are highly correlated like offensive rebounds and defensive rebounds, instead inputting total rebounds. However, schools are welcomed to add bonuses for specific performances, like averaging over 10 rebounds a game for centers, or 10 assists for a point guard. Nevertheless, for basketball specifically, in order to determine which values are statistically significant, and since so many of these variables are highly correlated, teams should to stick to either strictly using advanced analytics, or conventional statistics. While some values are important depending upon the scheme, this paper suggests opting for using the more conventional statistics when having to choose between the two. It is true that the advanced analytics like assist percentage (an estimate of the percentage of available rebounds a player grabbed while playing) may paint a better picture of a player’s success given the circumstance, conventional statistics are more easily determined since “potential assist opportunity” is a vague and “fudge-able” statistic.

Table 2. Salary/Individual Salary Cap Regressed Against Significant Statistics

Games -0.00389*** (0.000749)
Minutes Played -0.0000989 (.0000560)
3pt Percentage 0.0659 (0.0670)
Personal Foul 0.0185 (0.0231)
Points 0.000416*** (0.0000965)
Player Efficiency Rating -0.00901* (0.00409)
Usage Percentage 0.00761** (0.00294)
Win Shares 0.000362 (0.0158)
Box Plus Minus 0.0164** (0.00519)
Value Over Replacement Player -0.0391 (0.0254)
Veteran° 0.0433* (0.0198)
Twitter Followers°° 1.81e-09** (6.76e-10)
Total Rebounds 0.000303* (0.000126)
Assists 0.000774*** (0.000215)
Steals 0.000303* (0.000126)
Blocks 0.000908 (0.000497)
Turnovers -0.00171** (0.000636)
Point Guard -0.0118 (0.0243)
Small Forward .0561* (.0230)
Center°°° 0.0347 (0.0328)
True Shooting percentage 0.00788 (0.133)
rookie°°°° -0.0573* (0.0230)
Pace -0.00494 (0.00281)
Contract L 0.0309*** (0.00812)
Playoffs_2019 0.00183 (0.0165)
Playoffs_2018 -0.0468* (0.0214)
Playoffs_2017 0.00225 (0.0269)
Playoffs_2016 0.0538* (0.0231)
Constant 0.593* (0.248)
Observations 523
R2 0.596

Note: Standard errors in parentheses
p < 0.05, ** p < 0.01, *** p < 0.001 ° Experience >6 years
°° As of March 18, 2020
°°° shooting guard and Power Forward omitted for collinearity
°°°° Experience >1year

Table 2 shows a regression the statistically significant variables regressed upon salary and the log of salary (to shore off the tails). This linear regression established the 28 variables that could statistically significantly predict a player’s salary, F(28, 494) = 26.06, p = 0.0000, explaining 59.63% of the variability of salary with an R2 value for SALARY of .5963. Due to the highly subjective nature of assessing an athletic talent coupled together with a team’s specific need within the ever-moving market of the NBA, we argue here that the R2 value is acceptably high, even though the R2 value is not necessarily indicative the model’s goodness of fit. However, in an attempt to avoid any heteroscedasticity issues, a robust model was taken to ensure a trustworthy model, since robust standard errors tend to be more trustworthy (9).

Ramsey Reset Test
H0:  model has no omitted variables
F(57, 436) =      1.20
Prob > F =    0.1641

Moreover, the Ramsey RESET test (25), which tests whether there is a significant non-linear relationship, a potential indicator of omitted variable bias, gives a f test result of 0.1641. Under a 95% significance test, we fail to reject the null hypothesis. This means that there is no evidence of functional form misspecification and that the linear regression model sufficiently explains the relationship between the dependent (Salary) and independent variables. The equation of the regression line of the statistically significant variables to the 5% level reads: 

Player Salary/Player Salary Cap = .593 –.004(Games) + .0004(Total Points)
– .009(Player Efficiency Rating)  + .008(Usage Percentage) + .016(Box Plus
Minus) + .043(Veteran) + 1.81e-09(Twitter Followers) + .0003(Total
Rebounds) + .0008(Total Assists) + .002(Total Steals)– .002(Total Turnovers)
+ .031(Contract Length) – .058(rookie) – Pace(.005) – .047(Playoffs_2018) +
.054(Playoffs_2016) + e

This equation estimates an NBA player’s PCE, which could be used by the NCAA as the closest analogue for their basketball teams. The equation is not perfect in light of NBA contracts have complexities like Supermax eligibility, bird rights, Designated Veteran Player Extensions, and many more caveats that are inapplicable to the NCAA. Nevertheless, here we divided Player Salary by Salary Cap, which if 1, means that player is reaching their individual salary cap. The individual salary cap (ISC) variable was derived by putting players into three general categories based upon their experience: 0-6 years (25% of cap), 7-9 years (30% of cap), and 10+ (35% of cap) (20).

The regression shows that players start at .593 of their PCE. This is desirable because it means that at their base athletes are getting more than half of their salary cap, insuring they get something. Sixteen variables have positive coefficients, all of which ones would expect, like points, assists, and box plus minus.

Next, interestingly games played is negatively correlated with an NBA athlete’s salary. This is likely because many marquee players this past year were hurt or on minutes restrictions, like John Wall for example, who played 32 games in in the season due to a torn Achilles, yet still receiving a $42 million dollar per year contract the following year (11). As a result, this paper suggests actually flipping the sign, and rewarding athletes to play more games and minutes. After all, their presence on the field is an indicator of value to the team.

There is also a negative coefficient attached to the 2018 playoff variable, which may suggest that some teams with low salaries made it to the playoffs, indicating young teams or teams that, despite low spending, made it to the playoffs. This is akin to a smaller, lesser-known team making it deep in the NCAA tournament, like Steph Curry’s Davidson (15). Irrespective, players should be rewarded for their playoff performances and an aggregate of the past four years of playoff performance, if applicable, should be included as a net positive in any revenue sharing scheme. Rookie and Turnovers predictably have negative coefficients, as these statistics are undesirable. However, Player Efficiency Rating, statistics that one may predict to be positively correlated with salary, have negative coefficients. This may be because the highest paid players are generally the best, and the best players are the most likely to be put in one on one, “go get the team a bucket” type situations. As such, they are more likely to be put in positions to slow down the pace, and take tough to make, high-risk high reward shots with the game on the line. While the best players tend to hit those shots, being forced to take shots like that likely drive down their PER relative to their role-player teammates. Finally, while contract length is not directly applicable, players with more eligibility would be more valuable in the long term to their teams, so contract length here would be between 1 through 4.

Zion Williamson PCE = .593  – .004*(33) + .0004*(746)  –  .009*(40.8)  +
.008*(28.6) +  .016*(7.2) + .00084 + .0003*(293) + .0008*(68) + .002*(70) +
.001*(59) -.002*(78) + .031(4) – .058 – .047 + .054 + e

Applying this equation to Zion Williamson’s college stats (40), the number one overall pick in the 2019 NBA draft, we arrive at .989, plus an error term. This equation is not perfect as Zion’s Pace and PER were not available statistics, nevertheless .989 coupled with the error term should amount to about 1, which maps well to Zion’s real-world status as one of the most hyped, and marketable, college players ever (32). Therefore, Williamson’s Salary would be $469,983.75, just barely below Jalen Green’s reported $500,000 he will receive in the G-League in the 2020-2021 season (22). Overall, this is a clear, calculable, and circularly formula that adequately quantifies value such that administrations may trust it to derive and justify a NCAA player’s salary.


This article offers the NCAA a reputable, repeatable, and reasonable student-athlete revenue scheme that will ensure it remains competitive in an ever-encroaching market. By subscribing to this revenue scheme, the NCAA might not only reach the high performing student-athletes who may once again consider the NCAA as the right pathway for their athletic, as well as academic future, but also may inspire the very students walking their campuses to enter the market, since they could actually be paid for their efforts. This phenomenon would ensure its overall competitiveness in the future. Finally, fair pay for the athletes may turn the public’s opinion of the NCAA for the better, could increase both the supply and demand for NCAA sports–an impact that may just prove for all those involved, priceless.

Limitations and Further Research

This article aimed to build upon Stocz revenue compensation formula by both providing examples from various institutions with a basketball lens, and by providing further motivation as to why the NCAA should act sooner rather than later. However, different reputable databases have slightly different values when it comes to revenue. This underscores the need for transparency and universality amongst college revenue reporting. This paper used the revenue data from the EADA as that is a U.S. Department of Education run database and arguably stands to have more information. Future research should be conducted as to the veracity of colleges’ revenue sharing in general by third-party institutions. Furthermore, further research into the legal ramifications pertaining equal pay and Title VII in a compensation scheme such as this could be conducted by future scholars. Moreover, because of the nature of social media, the twitter follower numbers are likely to be different from the time writing this to the time it is read. Moreover, some athletes do and will not have a twitter, opting for Instagram, Facebook, Twitch, or no social media. While this article opted to use twitter followers as that is more readily ascertainable, future research should be conducted, perhaps using a social media listening tool, to ascertain which social media platform best explains an athlete’s salary. Finally, while most of an NBA players salary may be explained by individual statistics, at least some their salaries may also be influenced by external factors such as being the only all-star free agent for that year which would afford them a high likelihood at getting near a top salary, as well as a player taking a pay cut in order to make salary cap space for another high caliber player. Further study may aim to quantify these variables in a tangible way, if possible. Nevertheless, this article simply uses the NBA as a comparator to the NCAA, but not every NBA wrinkle will be workable for the NCAA.


  1. B-R Bullpen. (2019). Retrieved May 26, 2020, from
  2. Branch, S. (2019, August 30). The Shame of College Sports. Retrieved May 25, 2020, from
  3. Brown, M. (2019, January 11). MLB Spent Less On Player Salaries Despite Record Revenues In 2018. Retrieved May 26, 2020, from
  4. Conway, T. (2019, June 17). LaMelo Ball Signs 2-Year Contract with Illawarra Hawks of Australian NBL. Retrieved May 25, 2020, from
  5. Curtis, C. (2019, February 21). Nike’s stock dropped after Zion Williamson’s shoe exploded. Retrieved May 26, 2020, from
  6. Estimated probability of competing in professional … (n.d.). Retrieved May 25, 2020, from
  7. Fair Pay to Play Act (2019) Retrieved May 25, 2020, from
  8. Fenton, C. (2019). Https:// Retrieved May 26, 2020, from
  9. Hagquist, C., Stenbeck, M. (1998). Goodness of Fit in Regression Analysis – R 2 and G 2 Reconsidered. Quality & Quantity 32, 229–245
  10. Hurst: Payment of Student-Athletes: Legal & (and) Practical Obstacles. Published by Villanova University Charles Widger School of Law Digital Repository, 2000 Retrieved May 26, 2020 at
  11. John Wall Stats | (2020). Retrieved May 26, 2020, from
  12. Kalland, R. (2015). Clemson’s new football complex will have laser tag and a … Retrieved May 26, 2020, from
  13. Lauletta, T. (2019). LeBron James hosted California Gov. Gavin Newsom as he … Retrieved May 25, 2020, from
  14. Li, G. (1985). Robust regression. In Exploring Data Tables, Trends, and Shapes, ed. D. C. Hoaglin, F. Mosteller, and J. W. Tukey,  Wiley. ISBN: 978-0-470-04005-8
  15. Lopresti, M. (2018, March 12). March Madness: A look back at Davidson, Stephen Curry’s run in 2008. Retrieved May 26, 2020, from
  16. Moyer, A. (2015). Throwing Out the Playbook: Replacing the NCAA’s Anticompetitive Amateurism Regime with the Olympic Model. The George Washington Law Review. Retrieved May 26, 2020 at
  17. NBA Collective Bargaining Agreement. (n.d.). Retrieved May 26, 2020, from
  18. NBA G League Introduces Professional Path For Elite Basketball Prospects. (2018, October 18). Retrieved May 25, 2020, from
  19. NCAA MANUAL, supra note 8, Bylaw, at 71. Retrieved May 26, 2020 at
  20. NBA Salary Cap FAQ. Retrieved May 25, 2020 at Archived from the original on July 5, 2012.
  21. O’Bannon v. Nat’l Collegiate Athletic Ass’n, 802 F.3d 1049, 1052 (9th Cir. 2015)
  22. O’Donnell, R. (2020, April 16). Jalen Green is the perfect G League prospect to take down the NCAA. Retrieved May 26, 2020, from
  23. Pareto Efficiency Definition – Investopedia. (n.d.). Retrieved May 26, 2020, from
  24. Position U: Which college basketball programs produce the best players at each position? (2020, April 01). Retrieved May 26, 2020, from
  25. Ramsey, J. B. (1974). “Classical model selection through specification error tests”. In Zarembka, Paul (ed.). Frontiers in Econometrics. New York: Academic Press. pp. 13–47ISBN 0-12-776150-0..
  26. Rosner, S. R., & Shropshire, K. L. (2011). The Business of Sport, 2nd ed. Sudbury, MA: Jones & Bartlett.
  27. Rovell, D. (2018). NCAA tops $1 billion in revenue for first time – Retrieved May 26, 2020, from
  28. Sallee, B., & Silverstein, A. (2020, April 29). NCAA takes big step toward allowing name, image and likeness compensation for athletes. Retrieved May 25, 2020, from
  29. Schad, T. (2020, March 15). Here are the six biggest changes in the NFL’s new collective bargaining agreement. Retrieved May 26, 2020, from
  30. Senne, J. (2016, December 23). A Review of the NCAA’s Business Model, Amateurism, and Paying the Players. Retrieved May 25, 2020, from
  31. Shropshire, K. L. (2011). Legislation for the Glory of Sport: Amateurism and compensation. In S. Rosner & K. L. Shropshire, The business of sports (pp. 648-652). Sudbury, MA: Jones & Bartlett.
  32. Smith, M., & March 11, 2. (2019). The Business of Zion. Retrieved May 26, 2020, from
  33. Stacey Osburn. (2019, December 06). NCAA statement on Gov. Newsom signing SB 206. Retrieved May 25, 2020, from
  34. Stocz, M., Schlereth, N., Crum, D., Maestas, A., & Barnes, J. (2019). Student-Athlete Compensation: An Alternative Compensation Model for All Athletes Competing in NCAA Athletics. Journal of Higher Education Athletics & Innovation, 1 (5), 82-101. Retrieved May 26, 2020 from Retrieved from
  35. Vattikonda, N. (2019, April 10). Duke Athletics’ annual equity report sheds light on spending, coaching for men’s and women’s teams. Retrieved May 26, 2020, from
  36. WaldrepV. Texas Employers Insurance Association. 21. S.w.3d. 692. (tex. App. 2000)
  37. What is the NCAA? Retrieved May 25, 2020, from
  38. Where Does the Money Go? | – The Official Site of NCAA … (n.d.). Retrieved May 25, 2020, from
  39. Zagoria, A. (2020, April 26). NBA G League President Shareef Abdur-Rahim Targeting More Players To Join Jalen Green, Isaiah Todd On New G League Pathway. Retrieved May 25, 2020, from
  40. Zion Williamson College Stats: College Basketball at Sports. (2020). Retrieved May 26, 2020, from
Print Friendly, PDF & Email