Independent marks with a documented method.
Independent valuation inputs with a stated methodology and point-in-time history, the two things a risk function needs from a pricing source.
Block Scholes3 min read
- Point-in-time since
- 2020
- Tokens covered
- 150+
- Options venues
- 10+
- Listings catalogued
- 1.8M+
/01 The problem
What a defensible mark must answer
Valuing a book asks what a position is worth, where the number came from, whether it would have been the same yesterday, and whether anyone could have moved it.
A raw venue feed is weak on all three: it is one venue’s view, its history can be revised, and its method is whatever the venue happened to print. Such a mark is difficult to defend.
/02 The fair price
A fair price, built from the whole market.
A single venue provides one exchange’s view of the market, including its outages and erroneous prints. A Block Scholes mark is built from the wider market. Every venue in the composite contributes, and each is weighted by how far it should be trusted.
Each source is weighted on four measures: how much of the market’s traded value it carries, how much of the strike range it quotes, how much of its data survives cleaning, and how well it fits the calibration across every listed expiry. A venue that goes thin, stale or wide loses weight automatically, before it can move the mark.
The result is a fair price: the level the market as a whole supports, calibrated onto one continuous surface. Each mark can be explained and therefore defended.
/03 The method
From venue quotes to a mark, in three stages
Filter. Every quote is screened before calibration: one-sided, stale and arbitrageable prices are removed.
Repair. The remaining quotes are checked for static arbitrage across strikes and expiries. A violation small enough to repair is repaired; one too large to repair is discarded.
Weight. Each source is weighted by confidence, and the weights sum to one. The result is fitted across every listed expiry onto one continuous surface, so any strike or tenor has a calibrated value.
/04 Point-in-time
History back to 2020, as it was known
Every series is kept point-in-time back to 2020: a valuation run today for any date since then reads the surfaces, curves and marks as they stood on that date, with no lookahead.
This supports revaluation, model validation and limit testing: a past month-end can be re-run against that month-end’s surfaces, the reason a mark moved can be shown precisely, and a margin model can be tested through past stress on the inputs that existed at the time.
Delisted instruments stay in the record, and history is served on the same schema as the live endpoints, so a historical query and a live one read the same shape. Corrections are rare, deliberate and made openly.
/05 Mark-price engine
A mark-price engine as well as a feed
Some clients also supply their own orderbook and receive a mark price in return: their book evaluated against our index, delivered as the reference their venue runs on. The service is the same for on-chain and centralised exchanges.
Settlement pricing is delivered the same way, as a TWAP over the settlement window or as a single point-in-time read where that is what settles the contract. On-chain, signed through the oracle, it becomes a mark a smart contract can settle against with the same confidence a risk team places in it.
Questions
What teams ask before they build.
How far back does the history go?
To 2020, point-in-time: a historical query returns what was known at that moment, with no lookahead.Is the valuation methodology documented?
Yes. Quotes are filtered, checked and repaired for static arbitrage, then weighted by confidence across venues. The stages are described on the Volatility page.Can marks be delivered on-chain?
Yes, through the oracle, as well as over REST, WebSocket and MCP.
Read next
The data and tools behind this.
- /01
Data & Analytics · Volatility
Implied and realised volatility, skew, term structure and calibrated surfaces across strike and maturity.
- /02
Data & Analytics · Historical Data
Point-in-time history for backtesting and model validation, to three years of depth.
- /03
Infrastructure · Oracle
Signed pricing and volatility data delivered through push or pull infrastructure.