A scoreboard for your own mind

The Forecaster's Ledger is an in-development method for recording predictions with stated confidence and reviewing resolved outcomes. The current calibration example uses declared seed data only; it is not a rule for future estimates or evidence of performance.

The engine

Predictions

Calibration by domain

Using it

The documented method records a prediction, stated confidence, and resolved outcome. Any displayed score or calibration output describes the entered data; it is not a prescribed adjustment or guarantee of future calibration.

Your first ten predictions

  1. State something about the future in one falsifiable sentence.
  2. Set your honest confidence, 0–100, and tag its domain and a resolve-by date.
  3. When an outcome is recorded, review the Brier field. The retained schema contains a Brier formula, but its live behavior has not been independently verified.
  4. If domain-level calibration outputs are present, treat them as descriptions of entered data, not verdicts on judgment or evidence of future performance.
  5. Treat the displayed ratio as a description of the declared seeded example, not as a prescribed adjustment.

The retained schema contains formula fields and declared seeded examples. Reminder behavior, live rows, and current view configuration remain unverified; no paid-AI step is required by the documented method.

What it is, and what it is not

An in-development method for reviewing scored predictions. It does not make predictions for you, establish general reliability, or guarantee an outcome.