The Second Apron and the Summer of 2026: How the Payroll Rewrote the NBA Order

On 1 July 2026, the Milwaukee Bucks placed Damian Lillard on waivers. The two years remaining on his contract, 112.9 million USD in total, were split into five equal instalments and spread across the payroll until the summer of 2030. Four days later, the Milwaukee Bucks signed Myles Turner to a four-year, 107 million USD deal.

Read as a news item, this is a roster swap by a team changing direction: a 34-year-old guard who had just torn his Achilles leaves, a 29-year-old centre arrives. Read as a spreadsheet, it is a subtraction. The Milwaukee Bucks sat above the second apron in 2026-25, a threshold of 207.8 million USD for the 2026-26 season. A team at that tier is barred from aggregating salaries in trades, barred from sending cash, has a distant first-round pick frozen, and loses the right to sign waived players above the mid-level exception. To get Myles Turner, they had to drop below the line first.

They dropped below it by erasing a name from the roster.

Three numbers frame the case: 112.9 million USD stretched, 107 million USD newly committed, 22.58 million USD of dead money per year. None of those three numbers says whether Milwaukee is stronger or weaker. They say something else: in the NBA summer of 2026, the contract became a tactical instrument, and anyone writing about transfers without reading the clause structure is merely transcribing press releases.

The 207.8 million USD line

The 2026 CBA built two walls instead of one. For 2026-26, the salary cap is 154.647 million USD, the tax line is 187.895 million USD, the first apron is 195.945 million USD, the second apron is 207.825 million USD. Four numbers, and only the fourth one genuinely changes behaviour.

Cross the first apron and a team loses the taxpayer mid-level exception, faces tighter salary matching in trades, and cannot receive a player via sign-and-trade. Cross the second apron and the penalty list gets longer: no aggregating multiple salaries for one player, no sending cash in any deal, no trading a first-round pick seven years out, that pick frozen at the bottom, and if a team sits in the second apron in three of five seasons, its first-rounder is pushed to slot 30 regardless of record.

That is the break from the previous era. The old luxury tax was an invoice; rich teams paid it and carried on. The second apron is an invoice plus a list of prohibited actions. It does not punish you for spending. It punishes you for having spent and still needing to fix the roster through trades.

The summer of 2026 was the first summer the system had enough data to force action. Two seasons earlier, teams still believed they could linger above the line for a year and step down. By July 2026, the cost of lingering was fully visible: a frozen first-rounder, stripped trade rights, and a queue of players in the 15 to 25 million USD band who could not be bundled together for anyone.

I have tracked the transfer market since the summer of 2026, when I opened a 30-row spreadsheet for the window and logged fee, salary, clause and announcement date for every deal. That spreadsheet had no apron column. The summer 2026 spreadsheet does, and it decides every other row.

Summer 2026: three escape routes

The Boston Celtics are the clearest case. On 12 May 2026, Jayson Tatum ruptured his right Achilles tendon in Game 4 of the Eastern Conference semi-finals against the New York Knicks. The team lost its cornerstone long term, and at the same moment a group led by Bill Chisholm completed a 6.1 billion USD purchase of the franchise.

New owners do not buy a team in order to pay tax. Within ten days at the end of June 2026, the Boston Celtics sent Jrue Holiday to the Portland Trail Blazers for Anfernee Simons and two second-round picks, then moved Kristaps Porziņģis to the Atlanta Hawks in a three-team deal involving the Brooklyn Nets, receiving Georges Niang. Holiday had three years and 104.4 million USD remaining. Porziņģis was a 30.7 million USD expiring deal.

The savings figure reported in the American press was roughly 180 million USD, combining salary and tax. I cannot verify that to the dollar, so I use it only as a marker: Boston did not sell players because they were bad. They sold because a team carrying three large contracts and a recovering superstar cannot simultaneously keep its roster and keep the right to rebuild it.

The Denver Nuggets chose a lighter dose. On 30 June 2026 they sent Michael Porter Jr. plus a 2032 first-round pick to the Brooklyn Nets for Cam Johnson. The salary gap between the two is around 15 million USD a season, and 15 million USD at the second apron tier is not small change: it is the difference between having the mid-level exception and not.

The Phoenix Suns took the heaviest dose. On 22 June 2026 they sent Kevin Durant to the Houston Rockets, receiving Jalen Green, Dillon Brooks, the 10th pick in the 2026 Draft which they used on Khaman Maluach, and five second-round picks. In July 2026 they negotiated a buyout on the remainder of Bradley Beal's contract, waived him and stretched close to 97 million USD across five years, roughly 19.4 million USD per season. Beal then signed with the Los Angeles Clippers for two years and 11 million USD.

Three teams, three prescriptions, one diagnosis. None of those three lost because they lacked money. They lost because the money was locked into a structure that would not let them repair.

Oklahoma City and the inverted model

The Oklahoma City Thunder won the 2026 NBA Finals, beating the Indiana Pacers 4-3, with Game 7 finishing 103-91 on 22 June 2026. Shai Gilgeous-Alexander took both regular-season MVP and Finals MVP. They won with a roster built almost entirely through the Draft, on a payroll below the tax line during the championship season.

Ten days after the title, their spreadsheet changed. On 1 July 2026, Shai Gilgeous-Alexander signed a four-year, 285 million USD extension. Jalen Williams signed a five-year rookie maximum extension worth up to 287 million USD. Chet Holmgren signed for five years and 250 million USD. Total new commitments passed 820 million USD.

That is the trap of the homegrown model. You pay rookie-scale money for four years, you win a title, and then four rookie contracts mature across two consecutive summers. The Oklahoma City Thunder hold more Draft picks than any other team in the league, and that is why they can live with this invoice longer than anyone else. But they cannot live with it forever. By 2027-28 those three extensions alone will consume most of the salary cap.

I trust data more than people — because people know how to lie, while data only knows how to be wrong. Here the data is saying something nobody wants to hear: the best-built team in the league will be the first forced to break itself apart, precisely because it was built correctly.

Dončić, the supermax, and a decision not made on the floor

On 1 February 2026, Luka Dončić moved to the Los Angeles Lakers, with Anthony Davis, Max Christie and a 2029 first-round pick going to the Dallas Mavericks. That trade was read almost entirely as a dispute between a star and a front office.

The spreadsheet reads differently. Dončić was eligible for a supermax with the Dallas Mavericks, a five-year deal worth about 345 million USD. Being traded removed that eligibility. On 2 August 2026 he signed with the Los Angeles Lakers for three years and 165 million USD, with a player option for 2028-29.

The gap between the two numbers is about 180 million USD. No team on earth makes a decision about a 25-year-old purely on the basis of 180 million USD. But no team makes that decision without looking at the number either.

At the second apron tier, a supermax is no longer a reward for a star. It is a commitment consuming close to 35 percent of the cap, plus locking out the ability to aggregate salaries to repair the roster for four more seasons. The Dallas Mavericks did not trade Dončić because he was not good enough. They weighed a different question: keep him at 345 million USD inside a system that does not permit error, or swap for a 32-year-old on a shorter deal plus a pick.

The answer may still be wrong. But it is the answer of a team that read the contract before it read the rumour, and that is the point worth recording: from the summer of 2026, the biggest trades in the NBA are no longer decided by who is better. They are decided by who is more expensive inside a system that permits only one team to be expensive.

The official NBA story is tidy: the second apron stops rich teams from buying championships, forces everyone to build through the Draft, and returns the league to balance. The 2026-26 season opened with eight different champions across fifteen years, and no team has defended a title since the Golden State Warriors in 2026-18. It sounds reasonable.

But the apron does not redistribute money. It only restricts how money is spent. And inside a system that restricts spending, the most valuable asset is not cash — it is Draft picks and player development. Those two things concentrate in a very small group of franchises, and no clause in the CBA spreads them out.

The Oklahoma City Thunder accumulated picks by selling off pieces and losing for three seasons. The Brooklyn Nets did the same by absorbing bad contracts for picks. The Utah Jazz do it systematically. Those three are not richer than anyone. They are merely more patient, and patience inside a system that penalises roster repair through trades is the highest-yielding asset in the league.

The second blind spot sits in the middle tier. A player averaging 14 points and 7 rebounds, good in the playoffs, is worth 20 million USD a season. He used to be a flexible piece: a team could bundle two or three contracts to acquire him. Now a second-apron team cannot aggregate, and a team below the line does not want another 20 million USD. The mid-tier player has become the least tradable asset in the league, and no CBA clause protects him.

The third blind spot: the apron does not touch the buyout market. A team below the first apron can still sign a former All-Star at the minimum in February. The balancing mechanism leaves open exactly the door the strongest teams need.

Data does not interrupt the narrative — it tells a different story, and it is rarely wrong. The story here is that the second apron has not flattened the league; it has merely changed the currency. Cash is taxed. Draft picks are not.

The part not in the spreadsheet

On 27 April 2026, Damian Lillard tore his left Achilles tendon in Game 4 of the first round against the Indiana Pacers. He was 34, had two years left on his contract, and was entering the longest rehabilitation of a career.

Sixty-five days later he was placed on waivers by a team that needed to shave 22.58 million USD a year for the next five years. There was no farewell press conference. No thank-you note on the team website. Just one line in the transaction log.

In July 2026 he signed with the Portland Trail Blazers, the franchise he played eleven seasons for. It is a good ending in a way the spreadsheet neither predicted nor needed to predict.

I write this section for a professional reason. When you cut off an argument with a table of numbers, you tend to believe the table is the whole story. It is not. A 34-year-old erased from a roster while learning to walk again is a human event, and if my analysis cannot say that, my analysis is missing a column.

That does not change the conclusion about the mechanism. The Milwaukee Bucks still had to do what they did, or accept losing the right to repair the roster for four years. But it changes how I write that conclusion.

Deadline

In July 2026, when the 2026 CBA took effect, I wrote that the second apron would force at least three teams to sell a cornerstone before the 2026-26 season opened. The actual number was four: the Boston Celtics twice, the Phoenix Suns once, the Milwaukee Bucks once. Right on direction, wrong on scale. Recorded for the file, no excuses attached.

Three forecasts for the next stretch, each with an expiry date.

First: before 1 March 2026, at least one team sitting above the second apron will trade a player earning 25 million USD or more, and the primary motive will be escaping the threshold rather than upgrading the roster. Probability: 70 percent. Alternative scenario: if no such trade happens, it means teams have found a way to route around the cost through rookie-scale exceptions, and I will have to rewrite my mechanism section.

Second: the Oklahoma City Thunder will not keep the trio of Shai Gilgeous-Alexander, Jalen Williams and Chet Holmgren on a payroll below the second apron after the 2027-28 season. Probability: 80 percent. This is the easy forecast, because the arithmetic has nearly written itself.

Third: before 1 July 2026, at least two more teams will use the waive-and-stretch provision on a contract of 50 million USD or more. Probability: 60 percent. That provision is the only tool left to a team that has locked itself in.

A spreadsheet does not lie — only the person too lazy to read it fools himself. The summer of 2026 handed me one more column, and that column is saying the next transfer window will no longer be decided by who wants whom. It will be decided by who still has the right to want.

The greatest story in basketball lives in the data columns nobody reads. The apron column is the newest, and it has just finished writing its first chapter.