Scoring Validator

A probability, not a promise

A Scoring Validator (SV) publishes a probability of default for covered assets and keeps those predictions current. A prediction of 5% does not promise repayment; it says that default is expected to be uncommon across comparable observations. The protocol evaluates a collection of predictions rather than treating each default as misconduct.

The operator also declares a methodology commitment and reference. This makes methodology changes observable without making the contract an auditor of the underlying model.

From predictions to outcomes

1. The SV publishes and updates predictions for covered assets.
2. A cohort closes, fixing the predictions used for that cohort.
3. The configured forward horizon and payment allowance elapse.
4. The contract evaluates eligible scheduled-payment outcomes.

The implemented outcome test reads whether relevant payments were resolved on time. It does not rely solely on when a keeper happened to mark a payment missed. Assets with no scheduled payment in the horizon are excluded from the eligible sample.

The reference design uses a one-year forward horizon. The deployment's configured horizon, publication cadence, and evaluation deadlines govern actual operation; a cohort is not ready merely because its publication window has closed.

What calibration measures

The test uses a Brier score: the average squared difference between a predicted probability and the observed binary outcome. Confident predictions that turn out wrong contribute more error.

The protocol compares observed error with the expected error implied by the predictions, allowing for sample size. Lower error is better, but perfect calibration does not require zero error. A minimum eligible sample is required; a skipped small sample is not a successful calibration result.

Evaluation can proceed in bounded chunks of on-chain computation, so a large cohort does not have to fit into a single transaction.

Operational responsibilities

Follow the Brier example to calculate an observed score, expected score, standard error, and breach threshold. The parameter reference records the verified horizon and minimum sample. Keep predictions current, declare methodology changes through the supported process, and track when cohorts become evaluable. Overdue evaluation can restrict rewards and new volume; it is different from a completed evaluation with too few eligible observations.

A calibration breach enters the penalty process. Missed publication cadence and unannounced methodology changes have their own enforcement paths.

Limits of the signal

Calibration depends on the protocol's settlement records and defined horizon. It does not independently assess every legal, liquidity, valuation, or operational risk of an instrument. The current exit model also does not retain an SV's collateral until every future cohort has been evaluated; a late evaluation is not a guarantee that collateral remains available to debit.

Prepare scoring operations

Follow Become a Business Validator for admission, role collateral, and monitoring preparation. A scoring role does not grant token-management authority or commit an insurer to cover the asset.