Auction-Settled Credit Rate (ASCR)
The rate at which over-collateralised credit clears, discovered by competitive auction and settled on-chain. This page is the complete, permanent definition: what ASCR measures, how it is computed, what it does not claim, and what happens if a published rate ever changes.
What ASCR is
ASCR-<ASSET> at a given tenor is the annualised rate at which loans denominated in the quote asset, collateralised by <ASSET>, settled on the Gavel Protocol for that maturity.
It is discovered, not set. Borrowers describe the loan they want; lenders compete in an open auction, bidding the rate down; the auction closes and the loan settles on-chain. Each settlement is one observation: a real rate, at a real maturity, that two parties agreed to. No committee sets it, no formula computes it from pool utilisation, and no price oracle sits anywhere in the loan's lifecycle.
The family is asset-parameterised. Each instance measures one collateral asset, and says nothing about any other.
The published series
| Instance | Collateral / quote | Status |
|---|---|---|
| ASCR-BTC | WBTC / USDC | Awaiting first settled auctions |
| ASCR-ETH | WETH / USDC | Not yet accepted as collateral |
Published tenors: 7d, 14d, 30d, 60d, 90d, 180d, 365d, 730d. A tenor with no settled auction returns null and is not interpolated.
How the rate is derived
Settled auctions are binned into standard tenors. A log-quadratic is fitted through the bin midpoints:
rate = β₀ + β₁·ln(d) + β₂·ln(d)²
where d is the loan duration in days. Log-duration space captures normal, inverted and humped term structures naturally, without switching models or imposing a shape in advance. The fit publishes its own R² and RMSE with every response, so a consumer can judge how much to trust it rather than taking the number on faith.
The canonical curve is volume-weighted: a larger loan carries proportionally more weight in its bin, matching the convention of every published benchmark rate — SOFR, T-bill yields — rather than treating a token trade and an institutional loan as equal observations.
What goes in
Settled loan auctions on the Gavel Protocol, read from Arbitrum One. Nothing else. The protocol is permissionless and open-source, so every observation behind every published rate is independently verifiable on-chain by anyone — without asking us, and without trusting this page. That is the property that makes ASCR checkable rather than merely published.
What happens if a published rate changes
The question anyone citing a benchmark has to be able to answer in advance. This is the policy, and it is part of the benchmark's definition rather than an operational note.
This URL does not move. https://www.thegavel.io/methodology/ascr is a permanent public interface. A benchmark that breaks its own citations is not a benchmark.
What ASCR does not claim
Stated plainly, because the people worth being cited by are the people who check. None of this is a reason not to use the rate; it is what using it correctly requires knowing.
- Own-account seeding. The observations behind the current rate are predominantly own-account: Aletheia Analytics participates on its own account to seed observable auction data while the market is young. Until external participation dominates the fit, the rate is a published data product — not yet an independent market assessment. The live own-account share is shown alongside the curve, and this limitation retires itself once external participation crosses the independence threshold (and reappears if it falls back).
- Sample depth. The rate rests on the auctions that have settled. Early in a market that is few, and a fit through few points carries wide uncertainty. The current sample count is published alongside every rate rather than buried.
- Coverage gaps. Tenors the market has not discovered return null. They are never interpolated across, and the curve is never extended past the longest settled loan.
- Not a risk-free rate. ASCR is the rate at which over-collateralised credit cleared, including whatever the market prices for smart-contract risk, collateral volatility, and the borrower's option to walk away. It is not comparable to a sovereign yield without adjusting for those.
- Single venue. ASCR measures one market — the Gavel Protocol. It is not a composite across venues, and does not claim to represent the wider crypto credit market.
- Collateral-specific. Each instance measures credit against one collateral asset. ASCR-BTC says nothing about credit against any other asset.
The stable endpoint
ASCR is served from a documented, stable endpoint, and to AI agents over the Model Context Protocol. The free tier requires no key.
GET https://api.thegavel.io/v1/yield-curve?pair=WBTC/USDC
How to cite it
ASCR is free to cite, quote, chart and republish, with attribution and a link to this page. There is no licence to negotiate and no permission to request.
ASCR-BTC 30-day, [rate]% APR (2026-07-16). Auction-Settled Credit Rate (ASCR), Aletheia Analytics SASU. https://www.thegavel.io/methodology/ascr
Who publishes it
ASCR is computed and published by Aletheia Analytics SASU, which operates this application and its data product. Aletheia does not operate, control or guarantee the Gavel Protocol, which is permissionless and open-source — it reads the protocol's settled record and publishes a derived rate from it. Questions about the methodology, and reports of a suspected error, go to contact@thegavel.io.
Informational purposes only — not financial advice, not a recommendation, and not an offer. ASCR is a record of rates that settled, not a quote you can transact at.