How high will the NBA's Salary Cap rise due to the new TV deal?

There has been a large amount of speculation for the rising Salary Cap due to the upcoming National TV deal. Because of this, I thought it would be beneficial to put my own predictions forward and discuss how I came to them as well as what might potentially be troublesome with the current NBA projections. There may be more information in this article than the typical reader may care to digest, so because of this let me give you the results now and you can feel free to skip the gory details:

Season	Predicted Salary Cap	Predicted Luxury Tax	Predicted 0-6 Max	Predicted 7-9 Max	Predicted 10+ Max	Implied Predicted BRI	Implied Predicted Benefits
2015--16	$66,400,611	$80,858,154	$15,528,613	$18,634,336	$21,740,059	$4,945,567,500	$220,628,556
2016--17	$84,722,058	$102,864,544	$19,835,860	$23,803,032	$27,770,204	$6,206,095,875	$234,945,569
2017--18	$89,480,273	$108,588,351	$20,953,848	$25,144,618	$29,335,387	$6,536,400,669	$239,977,463
2018--19	$94,387,580	$114,491,258	$22,106,885	$26,528,262	$30,949,638	$6,876,970,702	$245,129,297
2019--20	$99,451,455	$120,582,242	$23,296,728	$27,956,073	$32,615,419	$7,228,319,237	$250,406,379
2020--21	$104,679,750	$126,870,730	$24,525,224	$29,430,269	$34,335,314	$7,590,985,199	$255,814,270
2021--22	$110,080,710	$133,366,623	$25,794,312	$30,953,174	$36,112,036	$7,965,534,459	$261,358,807
2022--23	$115,662,992	$140,080,313	$27,106,026	$32,527,232	$37,948,437	$8,352,561,182	$267,046,109
2023--24	$121,435,686	$147,022,714	$28,462,505	$34,155,007	$39,847,508	$8,752,689,241	$272,882,594
2024--25	$127,408,336	$154,205,286	$29,865,993	$35,839,192	$41,812,390	$9,166,573,703	$278,874,996

(Maximum Salaries are based off of 42.14% of Projected BRI instead of the 44.74% that determines the Salary Cap.)

You may notice that these projections are less than what have been reported elsewhere. So let me take you through the steps for arriving at these numbers.

Current National TV Deal (covering 2008-09 to 2015-16)

During the 2007 offseason, the NBA announced that they had reached an agreement to extend their television partnership with ABC/ESPN and TNT which was due to expire at the conclusion of the 2007-08 season. At the time, the total value was reported to be around $7.44 billion over eight seasons, which turns out to be an average of $930 million a season. It is important to note that this does not mean that each year the NBA receives $930 million a season. The payout actually increases over time, although the only snippet of information on a specific year's amount comes from a Larry Coon ESPN Insider article:

According to league sources, the new agreement escalates over time, starting at $2.1 billion in 2016-17 and climbing to $3.1 billion in 2024-25. The current agreement provides $1.03 billion in 2015-16, meaning the network TV revenue will jump by slightly less than $1.1 billion in 2016-17.

With a little bit of knowledge of the Collective Bargaining Agreement, some arithmetic, and numbers of Basketball Related Income from Larry Coon, we can piece together a pretty good idea of what the current National TV Deal has paid out yearly. To do this, we need to take a step back and understand how the Salary Cap is set each year.

Calculating the Salary Cap

I do not enjoy using mathematical equations, but it is the easiest way for me to convey how we can better understand how the Salary Cap, and its projections, is calculated. With that said, \(SC_{t}\) is the Salary Cap in year \(t\), which is the NBA Season of interest:

$$SC_{t} = \frac{ \text{Projected BRI}_{t}*0.4474 - \text{Projected Benefits}_{t} } { \# \text{ of teams (30)}}$$

BRI is Basketball Related Income, which is calculated and reported at the end of year audit in July. Projected Benefits are also calculated with the audit and they effectively represent the pension plan for retired NBA players in some form or fashion. We can see that the Salary Cap grows because either Projected BRI increases and/or Projected Benefits decreases.

BRI has increased every year since 2002-03 when a recession negatively affected revenues for teams. Along this time, the Salary Cap has decreased twice: in 2002--03 when BRI decreased and in 2009--10 when Projected Benefits increased at a faster rate than Projected BRI.

Because the focus of this article is on how TV Revenues affects the Salary Cap through BRI, we need to dissect BRI a bit more. BRI is calculated as follows:

Basketball Related Income (BRI) essentially includes any income related to basketball operations received by the NBA, NBA Properties1, NBA Media Ventures, or any other subsidiaries. It also includes income from businesses in which the league, a league entity or a team has an ownership stake of at least 50%. BRI includes:

Regular season gate receipts, minus taxes and certain charges including those related to arena financing; Broadcast rights; Exhibition game proceeds; Playoff gate receipts; The value of all complimentary tickets, minus "excluded complimentary tickets" (1.6 million tickets in 2011-12, increasing by 50,000 each season thereafter); Novelty, program and concession sales (at the arena and in team-identified stores within proximity of an NBA arena); Parking; Proceeds from team sponsorships; Proceeds from team promotions; Arena club revenues; Proceeds from summer camps; Proceeds from non-NBA basketball tournaments; Proceeds from mascot and dance team appearances; Proceeds from beverage sale rights; 40% of proceeds from arena signage; 40% of proceeds from luxury suites; 50% of proceeds from arena naming rights; 50% of the proceeds from team practice facility naming rights; Proceeds from other premium seat licenses; Proceeds received by NBA Properties, including international television, sponsorships, revenues from NBA Entertainment, the All-Star Game, and other NBA special events.

Some of the things specifically not included in BRI are proceeds from the grant of expansion teams, fines, all forms of revenue sharing, interest income, and the sale of assets.

BRI is effectively the majority of all revenues generated through the NBA and its associated entities. But we actually need Projected BRI. This is comprised of two components, the known revenues for the upcoming season (TV) and a guess at what the unknown revenues will be (Non-TV):

\begin{align} \text{Projected BRI}_{t} &= {TV}_{t} + 1.045 * \text{Non-TV} {BRI}_{t-1} \\ &= {TV}_{t} + 1.045*( {BRI}_{t-1} - {TV}_{t-1} ) \end{align}

The NBA TV deal has typically been negotiated once every 7 or 8 years. This helps in keeping the projections of the Salary Cap fairly stable, but the non-TV portion of the Salary Cap is less stable. This portion of BRI will fluctuate based upon the performance of Large Market teams (the better they do, the more revenue generated) and influx of newer Arenas (newer Arenas generate more revenue). Because of this, the NBA simply takes the previous season's BRI (less TV revenue) and increases it by 4.5%. This 4.5% was carefully chosen in the CBA negotiations and coincides with the maximum raises for most contracts, so the NBA uses this amount to try and be consistent from year to year with respect to player contracts and the Salary Cap.

Share of BRI as TV Deal

With the nuts and bolts of the Salary Cap covered, it is helpful to look at how BRI has increased over time and how the TV share of BRI changed as well. As I mentioned before, it is difficult to determine exactly what the yearly amount for the National TV contract is, so I've taken my best guess at what the actual TV Revenues from the most recent contract were paid out. Simply put, I know the total TV contract amount ($7.44 billion over 8 years) and the amount for the 2015--16 season from Larry Coon ($1.03 billion). Starting from the 2015-16 Season, I worked backwards and assumed that the first payout in 2008--09 was for $825 million increased each year by $30 million each year (this produces a total of $7.44 billion over 8 years, but the last year payout is $5 million higher than reported by Larry Coon).¹

Details of the yearly amounts for the TV deal have been discussed by both Zach Lowe and Larry Coon. The amount has been stated to be $24 billion over 9 years, which makes for an annual average of $2.67 billion. But we do not care about the average amount, we care about what the payout is for each year. Zach Lowe mentions that an NBA memo mentioned that the payout starts at $2.1 billion in the first season and peaks at $3.1 billion in the final season. I use a similar methodology from the previous NBA TV Contract. I know the total TV amount, the first season's value, and the final season's value. Using this information, I chose to assume the that each year increases by $125,000,000 (total compensation is $23.4 billion, less than the reported $24 billion) which is a similar method to what Dan Feldman did for ProBasketballTalk. This doesn't matter too much, because we are only focusing on the "jump" for the 2016-17 Salary Cap.

(TV Revenue as share of BRI listed above bars.)

Non-TV share of BRI for the 2001--02 Season is estimated to have been $1,967,000,000 and increased to $3,547,000,000 for the 2013--14 Season. This is an average annual growth rate of approximately 5%, which is the amount I decided to have Non-TV share of BRI increase for the 2014--15 Season and beyond. I have termed this to be Predicted Non-TV Income so as not to confuse anyone with Projected Non-TV Income.

If we only care about the "jump" in the BRI due to the National TV deal, then this is a straight-forward calculation. The previous National TV deal increased by an assumed average of $30 million each year. In 2015--16, the value was $1.03 billion and so we can assume the TV deal for 2016--17 would have been $1.06 billion if the terms for the current TV deal were extended. Instead, the new TV deal jumps to $2.1 billion in 2016--17 for an increase of $1.04 billion that we can attribute to the new TV deal. Borrowing from some details I explain later, I would have estimated that the Salary Cap for 2016--17 would have been slightly below $70 million. From the first table above, you can see that I have predicted the Salary Cap with the new TV deal to be about $85 million in 2016--17. So I would value the impact of the new TV deal as an increase of approximately $15 million in the Salary Cap. That's a lot.

Projected Benefits

Information on the benefits paid to players is available from Larry Coon's CBA FAQ site from 2001--02 through 2013--14, although to piece all of the values it takes a little bit of algebra. I spare you the gory details of how I determined the values that are not explicitly stated in the CBA FAQ and present them below.

While Benefits are estimated to be $72,143,519 in 2001--02 and $207,400,000 in 2013--14, it would not be appropriate to use the growth rate throughout this period to predict Benefits for the years 2014--15 through 2020--2021. This is because the definition and breadth of Benefits for NBA Players has expanded throughout 2001--02 to 2013--14 due to changes in the CBA.

From 2011-12 and beyond, the major change in Player Benefits is that 1% of BRI each year is used to fund a new benefits pool for the players. This makes the Projected Benefits a slightly different beast, which I describe as follows:

$$\text{Projected Benefits}_{t} = 0.01*\text{Projected BRI}_{t} + 1.045 * \text{Residual Benefits}_{t-1}$$

The Residual Benefits essentially include Pensions, player 401(k), supplemental medical benefits, education trust, insurance policies, playoff pool, employer portion of payroll taxes, anti-drug program, and the salary of minimum salaried players in excess of the 2-year veteran minimum. These totaled $140,050,000 in 2011-12, $161,970,000 in 2012--13, and $162,180,000 in 2013--14. I used my judgement to ignore 2011-12 benefits because of the strange nature of having played a 66 game regular season and instead focus on 2012--13 and 2013-14 to predict forward at 1% each year.

Predicting the Salary Cap

Current projections for the 2015-16 Salary Cap from the League Offices are around $67.1 million next season. The same projections put the 2016-17 Season at $89 million and $108 million in 2017-18. Frankly, those projections give nonsensical results. To understand this, we need to again turn to some math to figure out what the NBA projections imply for Projected BRI and Projected Benefits. To do this, all we need is how the NBA calculates the Luxury Tax in order to back out these values. Conveniently, the Luxury Tax is calculated in a similar fashion to the Salary Cap except with a different multiplier associated with the Projected BRI:

$$LT_{t} = \frac{ \text{Projected BRI}_{t}*0.5351 - \text{Projected Benefits}_{t} } { \# \text{ of teams (30)}}$$

At this point, you should have flashbacks to High School math class. We have a system of two equations (SC and LT) with two unknowns (Projected BRI and Projected Benefits). We can solve for Projected BRI by setting the Projected Benefits of each equation equal to each other to end up with:

$$\text{Projected BRI}_{t} = \frac{30}{0.0877} * (LT_{t} - SC_{t}) $$

Once we have Projected BRI, you can solve for Projected Benefits by inserting this into either the Salary Cap or Luxury Tax equation. This is the methodology that I have followed to create the first table. However, the NBA has their own projections that we can back out their implied projections of BRI and Benefits. Here they are:

(Predicted is from the first table.)

Season	Projected Salary Cap	Projected Luxury Tax	Implied Projected BRI	Implied Projected Benefits
2015--16	$67,100,000	$81,600,000	$4,960,091,220	$206,144,812
2016--17	$89,000,000	$108,000,000	$6,499,429,875	$237,844,926
2017--18	$108,000,000	$127,000,000	$6,499,429,875	-$332,155,074
2018--19	$100,000,000	$121,000,000	$7,183,580,388	$213,933,865
2019--20	$102,000,000	$124,000,000	$7,525,655,644	$306,978,335
2020--21	$107,000,000	$130,000,000	$7,867,730,901	$310,022,805

Well, this is strange. Apparently Projected Benefits in 2017--18 are negative. Why might that be?

There are three explanations that I can think of: 1) the Salary Cap is calculated slightly different than I have explained, 2) the NBA does not predict future Salary Cap and Luxury Tax using the components in Projected BRI and Benefits, or 3) human error.

For 1), I have taken liberties with how the NBA calculates the Salary Cap and Luxury Tax. This stems from the CBA guaranteeing the players receive at least 49% of BRI and at most 51% of BRI in a given Season. In the event that the players' Salaries and Benefits are less than 49% of BRI for any given Season, then the NBA will pay the difference to the NBAPA. The NBPA will decide how to distribute these funds to the players, but then this amount (divided by 30) is added to the following year's Salary Cap and Luxury Tax. So it is possible that the NBA is projecting a shortfall in Salaries and Benefits for the 2016--17 Season.

If this is the case, then this is dubious because the Salary Cap is designed to increase/decrease in order to match Projected BRI for each season. Projecting a shortfall would seem to imply that NBA teams will drastically spend less than any rise in the Salary Cap. In order to get an Implied Projected Benefits of around $240 million, then the shortfall in BRI would need to be $540 million (or $18 million per team). This is an implausibly large amount. With Projected BRI of $6.5 billion, the guaranteed share (51%) would be $3.315 billion. A $540 million shortfall would mean that Salaries and Benefits for 2016--17 would need to equal $2.775 billion with a Salary Cap projected at $89 million. For reference, Salaries and Benefits equaled $2.34 billion for the 2013--14 Season that had a Salary Cap of $58.7 million. So to get this result, one would need to assume that Salaries will only increase at a rate around 6% per year even though the Salary Cap is projected to increase by over 40% across that same time.

It is highly likely that 2) is at play. The NBA undoubtedly has more information than I do with respect to team revenue streams. For instance, there are Local TV Deals for each team that the NBA would know the terms to. If this is the case, the NBA may have more accurate projections than me. However, even if this were the case we still get the strange issue of having Projected BRI the same for 2016--17 and 2017--18 as well as the funky result of negative Projected Benefits. So while this may be at play, it does not answer why these projections are so different.

And finally, 3) is something that is always at play. Might there have been a spreadsheet error from the NBA in calculating projections? This could be a simple answer to why Projected Benefits are negative for 2017--18 or why Projected BRI is the same in both 2016--17. Might there have been a dissemination error from the official NBA projections to media members? That might have happened. Hell, there is even a chance that I have made a misteak mistake which is why I have provided a .csv file of the data I used below. As a note, I do not think I have made a mistake and I do not assume (or think) that the NBA has made a mistake.

Concluding Remarks

I'll conclude this article by mentioning that I think there is pause for concern with the current NBA Projections. I am not here to say that the NBA Projections are definitely wrong, but I would caution against accepting their projections to be what will occur. I would also caution against accepting my projections as some sort of truth. Making forecasts of random variables (Salary Cap, Non-TV BRI, Salaries and Benefits, etc.) is extremely difficult. There is a fair amount of uncertainty in the future that can easily derail a forecast. You should only use a forecast with the understanding that there is a fair amount of variation in outcome. And  these predictions only apply to the current CBA. If that changes (likely), then a new set of predictions would need to be generated.

---

In case you are interested in the raw data, here it is: Prediction Data

^{1. One can also determine that the NBA TV is not flat by recognizing that you can express the TV contract as:}

$$TV_{t} = \frac{ 30*{SC}_{t} + \text{Projected Benefits}_{t} } { 0.4474 } + 1.045*{TV}_{t-1} - 1.045*{BRI}_{t-1} $$

^{Then inserting hypothetical values of the TV contract based on the known values of the Salary Cap, Benefits, and BRI for a particular reason. You will find that only increasing values of the TV contract will make sense. ↩}