The hard cap

Teams restructure contracts for three reasons, and only one of them is desperation. The first is compliance: every club must be under the cap when the league year opens in March, and converting salary into signing bonus is the fastest legal room there is. The second is funding: the room does not have to cover a shortfall — it can pay for someone else’s free agent, which is why contenders convert in the same week they sign. The third is reward-and-retain: wiring a veteran his money years early, at no change to the headline, as a gesture that binds both sides. All three run on the same machine from the contract’s anatomy: take Paragraph 5 salary the player was owed anyway, pay it today as a bonus, and let proration spread the cap charge across the years that remain.

Players almost never refuse, and it is worth being precise about why. Cash now beats the same cash later — but more than that, money already paid is the most guaranteed thing in football. A salary is a promise with conditions attached; even a “guaranteed” one has vesting dates and offsets. A conversion bonus clears the day it is signed, beyond the reach of any injury, release or change of heart, and every converted dollar also raises the price of every future exit — deepening the club’s commitment without a word of new language. The restructure is the rare transaction both sides want, for opposite reasons, which is exactly why the league does so much of it.

Strip the vocabulary away and a restructure is a loan. The principal is the converted amount; the repayments are the equal slices future caps must absorb; the interest rate is zero. What makes the loan better than free is the collateral’s behavior: the cap has compounded at roughly 7% a year across its whole life, $34.61M in 1994 to $301.2M in 2026, with exactly one decline ever. A charge deferred is therefore a charge shrunk, in the only unit that matters — share of the ceiling. The anchor deal’s $12M proration slice is 4% of the 2026 cap; the same $12M against a cap three growth-years on is nearer 3.3%. Repaying 2026 dollars with 2029 cap space is the entire trade, and the league’s calendar offers it every March.

This is why conversion is not a desperado’s tool but standard treasury practice: a club that never restructures anything is declining an interest-free loan from a lender whose money inflates away. The difference between a disciplined cap sheet and the Saints’ sheet below is not whether to borrow — nearly every sophisticated front office does some of it — but how much, for how long, and against which players. Borrow against a 26-year-old quarterback and the future caps that repay it will barely notice; the arithmetic is genuinely on your side.

The bill kills for one reason: the loan is unsecured against decline. Every projection in the trade assumes the player is still worth his roster spot when the installments land, and when he is not — injury, age, a locker-room ending no model prices — the machinery runs in reverse. The charges arrive anyway, attached to zero production, often accelerated into a single season by the exit itself. Nine of the twelve largest dead-money charges in history date from 2021 or later, precisely the era in which caps grew fastest and the borrowing looked cheapest. Both truths hold at once: the discount-rate argument is correct, and the record book below is its graveyard.

The simulator’s Extend verb applies a real convention worth unpacking. The press will call that extension “$42M a season”, and here is the arithmetic: a $30M bonus plus $121M of rewritten salary is $151M of consideration, minus the $67M he was already owed in Years 3–4 — leaving $84M of new money over the 2 added seasons, $42M a year. The club’s ledger sees something tamer: $229M over 6 seasons is $38.17M. Both numbers are true; they answer different questions — the agent is pricing what the new signature bought, the ledger is pricing the whole marriage. Every extension headline you have ever read uses the first convention.

The cap sees a third thing entirely, and it is the extension paradox: the moment $84M of new money lands, Year 3’s charge falls from $42M to $20M, because a large salary was replaced by a small one plus a bonus prorating into seasons that do not exist yet — two of them void years bought purely as proration runway, with the final $6M parked in a season no one will play. The anchor deal signed with none of this machinery; one extension installs all of it. And a drafting constraint waits at the far end: no contract reaching past the CBA’s 2030 expiry may raise its non-bonus salary by more than 30% of its final-league-year figure per season — the flat late-year schedules of long extensions are that rule’s fingerprint, and the 2030 horizon explains why it exists.

There is one lever that genuinely shrinks the bill rather than moving it, and it is the least used: a straight reduction of Paragraph 5 salary. It requires the player’s signature — a contract is a contract — so the club’s leverage is always the release it could execute instead, and the usual dressing is a chance to earn the cut back through incentives. Unlike a restructure, the relief is real and immediate: nothing prorates, nothing waits in a void year, nothing accelerates later, because money never promised is money the cap never has to book. It is rare for the obvious reason — a veteran asked to fund his own team’s room can usually collect the same discount on the open market, with a change of scenery included.

The June 1 rule splits every exit bill by a single date: released before it, all the remaining proration and guarantees land on this year’s cap; released after it, this year keeps only its own slice and the rest defers to next season. The designation lets a club take a March cut and elect the June treatment anyway — two per club per league year, terminations only — but the fine print is a trap for the impatient: the full scheduled cap hit stays on the books until June 2, so the room the move creates is unusable while the actual free-agent market opens, shops and closes in March. Kyler Murray’s 2026 release is the mechanism live: of a $54.7M bill, $47.5M stayed on the 2026 cap regardless, and the ≈$5.1M of relief arrived only on June 2, months after anyone worth signing with it was gone.

Trades get no designation at all — the CBA grants it to terminations only — which produced 2026’s defining bit of calendar theater: Cleveland and Los Angeles agreed the Myles Garrett trade in the spring and processed it after 4:01 p.m. ET on June 1, letting the ordinary post-June-1 rule split $41.09M into $15.53M and $25.56M. A week’s patience in the A.J. Brown trade was worth $27.16M of 2026 cap the same month. Nothing was gamed; the split is simply what the rule pays anyone willing to wait, and June 2026 was the month waiting became standard operating procedure.

Dead money is not waste, and it is not a fine. It is the amortized cost of decisions already made: cash that left the building in some earlier March, which the cap — by the conservation law the simulator will not let you break — was always going to book somewhere. The exit only chooses when. Read a cap sheet accordingly: a club carrying a large dead charge mistimed a bet it deliberately made, while a club carrying none may simply have refused the cheapest loan in sports; neither number is virtue by itself. The money was spent on the player at the moment it was promised to him — the cap just hadn’t finished writing it down. What happens to the roster spot the bill sits on — tags, tenders, waivers, the market — is free agency.