RECEIPTS FIRSTRF / PUBLIC LEDGER

The Rules

Why the results cannot be quietly manipulated — first in plain English, then in the exact terms the scoring code uses. The rules are set before the result, and they apply to every claim equally.
CHAIN OF CUSTODY: created → timestamped → published → resolution window closes → scored → never edited.

01 / Logged before the move

Every claim is written down before its resolution window. There are no retroactive entries and no edits after commit.

Each thesis enters the ledger with a creation timestamp and a fixed horizon (1 week / 1 month / 3 months / 6 months). The claim's resolution window starts at its commit time. What you see in the ledger is the original wording; resolved claims are never reworded, re-dated, or removed.

02 / Scored against the market

If the whole market rose, that does not count as skill.

Directional (long/short) claims are scored on the market-adjusted return: realized − β·SPY over the same window, where β comes from ~120 trading days of daily returns before the claim. A LONG claim is a HIT when the adjusted move clears +2%, a MISS beyond −2%, and PARTIAL in the dead zone between.

03 / Range-bound means quieter than normal

Calling "sideways" on a stock that always moves 10% is not a prediction. The bar depends on the stock.

Range-bound claims are scored against the stock's own volatility: the band is 0.674·σ·√t (the median expected move), so a skill-less "it'll stay flat" caller converges to ~50% — the same coin-flip baseline as directional claims.

04 / No trigger, no credit

Conditional claims only count when the stated condition actually occurred. An unfired branch is void — not quietly added to the win column.

Catalyst scenarios are written as branches ("if X, then up; if Y, then down") and every branch is logged. When the event resolves, only the branch whose condition fired is scored; the others are stamped VOID and excluded from every statistic — in both directions. Dead branches are not wins, and they are not misses either.

05 / Misses stay published

The scorecard is the entire record — the denominator is public.

Hits, misses, partials, voids and pending claims are all counted and all visible in the ledger. Nothing resolved is ever unpublished.

06 / Uncertainty stays visible

The uncertainty is part of the result. False precision is still false.

Headline accuracy always carries a 95% Wilson confidence interval. Each claim states an explicit probability (P of scoring a HIT under these rules), and those probabilities are graded with a Brier score against the 0.25 no-edge baseline — overconfidence is penalized quadratically.

07 / The machine reads its own report card

The system is shown its own scored record — including what it got wrong — before every new claim it makes.

Every research cycle feeds the resolved scorecard back into the model: directional accuracy by setup, calibration of its stated probabilities, and its systematic biases. The misses are not just published; they are the training signal.

What "early record" means

A small sample is suggestive, not proof. We will not dress it up as one. Until the ledger is hundreds of decided claims deep, the honest statement is the interval, not the point estimate — and we print the interval everywhere.

Not investment advice. Informational research only. Research is released on a deliberate delay; every published item carries its original "as of" date. Built with AI. Judged by reality.