When the news broke that the NBA and NBPA were submitting themselves to an independent mediator, we collectively perked our ears up because such a move seems to be an admission that the negotiations are too broken to move forward without outside assistance. NBA.com's David Aldridge wrote an in-depth report on the developments, and one of the things offered as almost a throwaway was this information he extracted from NBPA head Billy Hunter:
Hunter also said the union would be willing to discuss reductions in roster size from the current maximum of 15 down to 12 players, or allowing "split" contracts that would allow teams to send players down to the NBA Developmental League during the season if they thought the player needed work he wasn't getting in the NBA.
"We've already indicated that we'd have discussions about the size of the roster," Hunter said.
Perhaps the media has already been over-saturated with angles that can be taken with this lockout and adding another wrinkle plunges us deeper into the labyrinthine Fire Swamp that we are beginning to be afraid we'll never escape. Even so, this little bit of information is interesting to me because it seems that it would 1) be a major concession of the players to simply dismiss potentially 20% of their work force; 2) it could potentially save a lot of money; and 3) barely anyone seems to care. So at the risk of completely mis-reading #3 (which could potentially be, "nobody cares because it isn't important"), it seems like a worthwhile exercise to dig in a bit and examine the kind of impact that roster reduction could have.
(As an aside, it does bother me slightly that Hunter made the comment in such a glib, off-handed manner. If something like roster reduction is on the table which could involve dozens of players suddenly finding themselves out of a job, I think it should have been a little more front and center.)
In order to see what kind of money could be in play if roster spots were eliminated, we can do a bit of break-even analysis. This break-even point exercise is probably more academic than practical, and involves a number of assumptions that are most certainly wrong, but I think it at least gets us thinking about how rearranging the deck can change the parameters of the debate.
I examine the potential effects of roster reduction using two methods:
- Roster reduction in averages - the examination of what happens when the average NBA salary is used as the basis of roster reduction.
LIFO roster reduction - Borrowed from the LIFO accounting method, the examination of what happens when the lowest marginal NBA salaries are removed from the overall salary structure.
We start with a few assumptions that are undoubtedly wrong but it just makes it easier to work through the math:
Number of teams: 30
Number of current players per team: 15
Total number of NBA Players: 450
Given a roster change from 15 to 12, total roster spots eliminated: 90
1. Roster Reduction in Averages
This method of break-even analysis is probably the easiest as well as least likely based on the assumption that we're using the average salary of NBA players to figure things out. What is the average salary of an NBA player? Surprisingly, even that answer isn't as easy as you might think.
In math class, we learned that to get an average, you just divide one cumulative number by the number of things that comprise it. Or, for analysis sake, if we take last year's BRI of $3.817 billion, take the 57% players' cut and divide it by 450 players, you get an average salary of about $4.8 million per player.
You also might recall that when league officials cite the average salary of players, it is a little bit higher. The reason why is because the league uses a specific calculation which can be found at Larry Coon's FAQ. For the purposes of this analysis, I'm going to go with a straight average number because at the end of the day, it doesn't really matter where we start.
Based on last year's final BRI and applying the range of player percentages that have been discussed, we start with this chart:
If we start at last year's BRI but instead apply the 53% player split, as they have publicly demanded as their absolute floor, you can see that the average salary would decrease from about $4.8 million to $4.5 million. You can see that if you reverse the split and give the players 47% as the owners would like, it would cost them about $600,000, or 13% of their average salaries.
If we reduce the roster size by one player, this chart reflects the result in our averages:
You can see here that by eliminating just one player from each team's roster (30 in total), it increases the personal stake each player has in the league's revenue considerably. Now, the players only need to take a 50% cut to get almost the exact same average salary as if they took 53% with a full roster.
To calculate the precise break-even point: 53% * 14/15 = 49.47%
At the next level of roster reduction where two players are eliminated, which removes 60 spots league-wide:
By eliminating two spots, the percentage that players would need to take the same amount of money now dips below even what the owners are looking for. The precise break-even point here is 45.9%.
Lastly, we take the 12 man roster talking point which Hunter mentioned, which would eliminate 90 roster spots league-wide:
Under this final scenario, the players see their percentage fall all the way down to 43%, which was one of the percentages floated early on by the owners, and even then they would still manage to come out ahead. The break-even point here is: 42.4%.
The net effect of this reduction in roster spots is considerable and it underscores the fact that the talent pool in the NBA is small but receives a kingly equitable proportion simply because of the amount of money that is made. The reduction of but 6.7% of the league's roster has a dramatic effect on each player's slice of the pie. By the time we wind down to Hunter's casual comment, if the players were to actually acquiesce to a roster reduction, they would still make more on average than they did in the previous season, even at a 46% stake in BRI.
2. LIFO Roster Reduction
Let us put aside all of these charts and calculations because we know they're not actually based on reality. We know that the league is not made up of 450 players who each make $5 million dollars; rather, there are a handful of "max contract" players, a hundred or so players still on their rookie contracts, and then a bevy of players who fall somewhere in between.
Instead, let us consider a more realistic method which I shall refer to as the LIFO method. LIFO is a method of cost accounting that means, "Last In, First Out." Or in other words, the last men on the totem pole are the first ones to get the ax (please go easy on me accounting geeks; just grant me the use of this pneumonic for a spell to make my point). In other words, if the Thunder had to reduce their roster from 15 players to 12, it is likely that the kind of guy who would get cut is a player like Byron Mullens ($1.3 million), not Russell Westbrook ($5.1 million). Using an average salary reduction likely overstates the impact of roster reduction and potential savings. A more useful way to consider it is to look at the cheapest players and see what the savings floor might be.
If we apply some of the statistics from the great ShamSports' site and do some back of the envelop math calculations, we can figure out how much money would be saved if we simply took off the bottom 90 players.
- 30 spots removed: Savings = $6.1 million
- 60 spots removed: Savings = $20.1 million
- 90 spots removed: Savings = $46.7 million
Clearly, these numbers are far removed from our average calculations, which would have yielded us with $150, $300, and $450 million in savings, respectively. It does however help create a lower boundary, as compared to the average salary calculation's upper boundary. This is because the truth is, the cheapest players aren't likely to be eliminated either, since many of those players are playing on rookie salaries which grossly underpay them (for example, Serge Ibaka
made a paltry $1.8 million last season because he's still on his rookie contract).
I am not sure how realistic it is that the NBPA is actually putting roster spots on the chopping block of CBA negotiations, but if we are, then here are a few tentative conclusions:
- If the NBPA is really considering conceding roster spots, I think this should be a big negotiating point in the CBA talks. If Hunter can go to the bargaining table and tell teams that he's willing to make a concession that could save the league somewhere between $50 million and $450 million per year AND the players would be willing to take a smaller cut of the pie in the process, they should be able to fight for much larger concessions on the NBA's side. The elimination of a hard cap, for example.
- While the former calculation probably overstates the effect of eliminated roster spots, you could also argue that the latter calculation understates it. I think that the elimination of the bottom rung players would not just clear out cheap salaries, but could also add some downward pressure on the NBA's middle class, thereby dragging down the middle players' salaries as well.
- Along with salary savings, there would undoubtedly be ancillary savings as well. With fewer players, teams would spend less on things like hotels, travel expense, per diem expense, health insurance, and all the other benefits that get tacked onto employing high net worth individuals.
- While the money for the remaining 12 players goes up proportionately, it also makes each one of those roster spots even more valuable. Therefore, you could see teams even more willing to avoid strapping themselves with ill-advised signings, a reduction in contract terms, and even the removal of guarantees all together.
A great many thanks to two of the smartest guys in the room who inhabit the virtual space at 8 Points, 9 Seconds who helped me develop this analysis.