NFT collections · Floor credit, fungible shares, scheduled settlement

Every ape becomes a company

A floor loan escrows the NFT and pays the holder cash. Fractionalize what's left and the piece becomes a tiny capital structure: one bond, 100e18 shares, and a scheduled exit. This is the derivatives layer raw fractionalization tried to be — with the floor paid for, the settlement dated, and the holdouts solvable.

The capital structure

One bond, one float, one deadline

senior bondThe floor lender's claim — $15k already paid to the holder, locked until the structure resolves. The "guaranteed floor" is a purchase that already happened, pending reversal. One whole instrument, and it trades (see the floor market below).
the sharesEverything above the debt, cut into 100e18 fungible shares worth (value − debt) ÷ supply. This is the tradable top: liquid, levered exposure to the piece — a $40k ape with $15k drawn is a $25k float, and a 50% floor rip is an 80% share move.
the deadlineThe tenor schedules a settlement event — the missing piece that killed raw fractionalization. Shares here can always become cash: at term, the structure settles or consciously extends (section 06). Nobody can be trapped.
the one ruleDebt freezes at fractionalization. Raising the principal would transfer value from shareholders to the borrower's pocket — so rolls extend time at the same principal, and any principal increase pays out pro-rata to shareholders, never to the borrower. One mechanical rule; no governance needed.
02 · Can the floor walk away?

The floor is guaranteed exactly when someone has paid for it —
by borrowing against it, or by renting it.

A bid on the collection book lives in one of three states, and only one of them can leave:

restingAn unfilled bid is free to leave — it's order-book depth, real and honestly quoted, but promised to no one. Pulling it can never touch an existing loan; it only thins what future borrowers can draw.
filledThe moment a bid fills as a loan, it cannot be withdrawn — structurally. The lender's cash is in the borrower's wallet; there is nothing left to un-bid. Borrowing against the floor is what freezes the floor.
reservedThe gap between those two is a product: pay a standing fee to time-lock a specific bid at your piece without borrowing — the holder buying a put, completing the book's option structure (lenders were always the writers).
Only unpaid floors can walk away — which is honest in a way "guaranteed floor" marketing usually isn't. Every firm floor on this book has a name, a price, and a fee attached.
The floor market

The bond layer trades too

buyoutA new lender who believes at $18k can buy out the $15k claim at a negotiated premium — the original lender books a profit on a bond, and the floor under the structure steps up. Believers compete for the right to be the floor.
shares repriceThe debt doesn't change when the floor bid improves — but the settlement reserve rises, so the shares reprice upward off the floor market. The bond book and the share float price each other continuously.
no leakThe borrower can't harvest a higher floor by upsizing the loan (the one rule) — a stepped-up floor benefits the people holding the equity, which is exactly who it should benefit.
the seat paysBeing the top committed bid is a paid position, four ways: the entry fee when the bid fills as a loan (the put premium), a buyout premium when a higher believer displaces you, dilution carry if the structure enters standby, and assignment upside — the ape at your chosen discount if it all ends at the floor. Resting bids earn none of this; commitment is what gets paid.
The issuer's seat

The house is incentivized to hold its own floor

A collection's issuer fits as the floor sponsor — and the design converts money they already spend (marketing, buybacks, royalty enforcement) into positions that pay:

sponsor bidsThe issuer seeds USDC into the collection's ladder — junior, time-locked, visible on-chain. The floor is the collection's stock price; today it's defended with vibes and sweeps. "Floor backed by $2M of locked bids" is a verifiable fact, not a promise. Marketing budget becomes balance-sheet depth.
assignment = buybackWhen the issuer's bid fills, they acquire their own NFT at their own chosen discount — a standing buyback program, automated and pre-priced, executing precisely in the drawdowns when buybacks matter. The put-writer's "risk" is the outcome the issuer would pay for anyway.
royalties, resurrectedMarketplace optionality killed royalty enforcement. The share float and bond transfers are registry-mediated by necessity — so an issuer royalty slice on every share trade, bond buyout, and settlement is structurally enforceable again. The collection's credit activity becomes the recurring revenue its JPEG trades stopped being.
rent-a-floorThe issuer can subsidize reserved bids under holders' pieces — "every ape carries a paid floor" as a loyalty benefit — converting marketing spend into holder protection with a price tag. And holders who can borrow don't panic-sell: sell pressure becomes borrow demand.
the guardsDisclose the sponsor's share of the ladder (a floor that's 90% issuer bids is a different fact than an organic one), and self-borrow caps apply to treasury-held pieces. Issuer bids are self-punishing against manipulation anyway — fake enthusiasm buys real apes with real dollars.
The holder's options

Cash now, upside kept, exit scheduled

Three ways to keep your ape — always:
1 · repayPay back the principal any time before expiry — no interest, no accrual (the fee was paid upfront) — and the ape leaves escrow.
2 · rollPay a fresh fee at current terms and the deadline moves. Available every term, indefinitely while the position is whole or you hold ≥50% of the float. (Sell the majority and rolls stand only until the majority forces settlement — but way 3 still wins you the ape.)
3 · win the auctionIf you fractionalized, reassemble through the settlement auction — where your own shares net against your bid. Holding 80% means truly paying only the debt plus the other 20% at fair value. No holdout can stop you.
The piece is only ever lost by choice: not repaying, not rolling, and not bidding is a decision — and it's the rational one exactly when the cash you already hold is worth more than the ape.
borrowTake the floor in cash — $15k today, the piece in escrow, the redemption right yours.
fractionalizeMint the float and sell as much of the top as you like — sell 60%, keep 40%: you've banked $15k + $15k and still own 40% of every dollar above the debt. Partial fractionalization is partial keeping.
how the sale runsPrimary distribution is an executor Dutch auction — a decaying ask with your reserve as the floor, filled by whoever bids first; no demand at your reserve and the sale honestly fails (the float is a venue for demand, not a source of it). Afterward the shares trade on the native share book — registry-mediated bids and asks (which is what makes the transfer fee collectible), with a public entrypoint aggregators can route into.
repay → keepBefore shares exist: repay the principal (no accrual — the fee was upfront) and walk out with the ape. With shares outstanding: reassembly happens through the settlement auction, where your own shares net against your bid — owning 40% means funding only the debt plus the other 60%.
walk → paidLet the term lapse. You keep the day-one cash, the share-sale proceeds, and any retained shares — walking forfeits only the exclusive redemption right. The structure outlives you (next section).
Expiry

Settle, or stand by — the market decides

The borrower walked, the term lapsed — but the shares still trade above the floor. Killing live equity with a fire-sale would expropriate shareholders. So expiry forks on what the market says the equity is worth:

the auctionAnyone, anytime post-expiry, can trigger settlement: an auction with the debt as reserve, seeded by the standing floor bids. Waterfall: bond made whole first, everything above pro-rata to shareholders. A bidder's own shares net against their bid — the cheapest reassembly for whoever owns the most float.
standbyIf no one triggers and shares trade above floor, the ape simply stays in escrow, fractionalized — and the carry switches to dilution: each epoch, the structure mints a risk-priced sliver of new shares to the floor lender as payment for waiting. The lender monetizes by selling into the float at the floating price — the market pays them; no liquidation does.
convergenceStandby ends three ways: shares sink to the floor (equity dead by market consensus) → title to the lender, shares retire; the lender's diluted stake crosses a threshold → clean-up buyout of the dust; or someone triggers the auction because the piece is worth reassembling. Every path is chosen by prices, not by a cliff.
who funds the rollA roll never needs the sitting lender's consent — it's a refinancing cleared by the book: the fee goes to whoever funds the next term, the incumbent exits at par plus accrued coupon (their scheduled, tenor-bounded exit), and a replacement takes the seat at current terms. No taker at the outstanding principal = the roll honestly fails — the market just said this floor no longer supports this debt, and the borrower rolls smaller, repays, or settles. Credit nobody will fund shouldn't be extendable; here that's enforced by capital, not committee.
the majority rightA borrower holding less than half the float no longer controls the deadline alone: rolls stand by default, but holders of >50% of the shares may force the settlement auction at any term boundary. Not a ballot — a threshold-gated trigger; silence is consent, and shares being fungible means an activist can simply buy the majority and force realization. Corollary: keep half to keep the wheel — retain ≥50% and the roll is yours by arithmetic.
who pays for timeExtension rights follow economics. A borrower who sold his whole stake has no reason to roll — and doesn't need to: standby is the automatic extension, with dilution charging every shareholder pro-rata for the lender's patience, no coordination required. (Any shareholder may pay an explicit roll fee instead — dilution just solves the free-rider problem for them.)
the honest lineStandby is a doctrinal exception: the floor lender's principal isn't tenor-bounded while it runs. It's compensated (dilution carry), exitable (the bond trades — someone else can take the floor seat), and priced at entry — lenders in share-mode markets quote knowing standby exists.
No cliff, no fire-sale, no expropriation: equity dies only by market consensus — or pays, in shares, for its own patience.
Play the structure

One ape, your decisions

You own a $40k ape and just borrowed $15k against it (100-day term — the dashed marker). Press play: the floor moves on its own, and the story pauses at each fork to ask what you want to do.

day 0 · loan open
ape floor valueyour debt ($15k)best committed bidterm expiry
Ape value
$40.0k
Best bid
$15k
Your cash
$15.0k
You hold
the ape (in escrow)
d 0day 0 — loan opens: $15k lands in your wallet, the ape goes into escrow. Term: 100 days.
Why shares work here

Raw fractions died of three diseases. All three are cured.

no floorRaw fractions could drift to zero on pure illiquidity — nothing stood under them. These shares sit on a floor someone already paid for: the loan cash is in the holder's pocket, the bond claim locked against the piece.
no exitRaw fractions had no terminal event — value realization required a buyer for the whole asset who never had a mechanism to appear. Here the tenor schedules the settlement, and post-expiry anyone can trigger the auction at any time.
holdoutsRaw fractions never reassembled — one stubborn holder trapped the asset forever. The settlement auction is forced reassembly: the waterfall pays every shareholder mechanically, and no one gets a veto.
Raw shares failed. Shares on a floored, term-bound structure are a different instrument — equity in a company with a bond, a book, and a scheduled liquidity event, not confetti from a shredded JPEG.
Yield & the field

What the structure pays, and what came before it

velocityTransfer fees on the share float — fungible shares mean real volume, and every trade pays lenders and debt down. Registry-mediated transfers make the fee collectible.
bond churnFloor buyouts pay transfer fees on the bond layer; believers competing for the floor seat is itself a revenue line.
collar premiumsThe escrow can write covered calls on the piece under a pre-authorized policy — premium is realized cash, split between debt paydown and the float. Floor below, upside sold above: an NFT collar.
standby carryPost-expiry dilution — the floor lender's payment for patience — is yield in shares, monetized on the float.
vs the fieldOracle-floor lending produced the 2022 auction-into-panic cascades; title-and-waterfall settlement has no cascade surface. Bid-based lending (credit where due) proved capital-as-price for NFTs — our deltas are the passive door (curated collection ladders), fixed terms with pending windows, and the share float + collar income on top.
honestyCollection-wide bids are quotes on the worst acceptable piece — adverse selection is structural, priced by bidding below floor, disclosed rather than solved. And the float's liquidity is trader-supplied: series standardization helps, but market-building is real work.
Open parameters

The mechanism is done. These numbers aren't.

Everything above is settled design. What remains is calibration — logged here so nothing is lost:

reserved-bid feePricing the bought put — the standing fee for time-locking a bid under a specific piece (risk-engine output).
floor-vol inputEstimating collection volatility from thin, lumpy sales data — the σ that prices ladders and entry fees.
collar policyCovered-call authorization once shares exist: whose consent, which strikes, what premium split.
series & tiersStandardized position series (poolable floats) and trait-tier sub-books — later, at the cost of fragmenting liquidity.
Illustrative mechanism walkthrough — not audited economics, not an offer. Companion to the Overview, Pricing, Risk, and Yield pages.