Skip to crypto evidence
Crypto Signal AuthorityCalls / Revisions / Execution

Crypto signal outcome ledger

Separate closed wins, closed losses and unresolved calls before reading a win-rate headline.

Editorial revision: 8 October 2026. Source limits appear below.

Public-document research and educational exercises. No paid-room tracking, exchange connection or audited customer return is claimed.

The denominator is a policy choice

The closed-trade rate uses wins divided by wins plus losses. The all-resolved-event rate also includes cancelled or expired calls in its denominator and must not be advertised as the same metric. Unresolved calls are shown separately. These counts must be mutually exclusive; a cancellation after a genuine fill belongs in the execution ledger, not the never-filled bucket.

What this count cannot measure

Twenty tiny wins and one large loss can still lose money. This worksheet does not calculate expectancy, leverage, drawdown, costs or whether omitted calls exist. Its invented defaults explain denominators only. A complete publisher record must define scratches, partial exits and duplicated messages before entering counts.

Sources and scope

The original ledger retains its original convention

This four-count worksheet uses wins and losses as its closed denominator and includes cancelled events in a separate resolved-event share. Its invented defaults remain twelve wins, eight losses, five unresolved and three cancelled events. The new eight-count lens does not replace that rule: it adds explicit gross flats, expired entries, listed unclassified calls and a declared population. Read the named denominator before comparing the outputs. No hidden conversion maps open positions into wins, and no account return is computed by either count worksheet.

Reconcile the row policy before entering a count

A persistent idea may have several messages and targets. Identify whether the supplied count describes ideas or events, and retain the same treatment for unfavorable calls. A cancellation after a partial fill is not a never-filled event. An omitted call needs a coverage request, not an assigned outcome. The original tool cannot detect absent source rows, duplicate ideas or inconsistent classification from four totals alone. The workbook makes those limits inspectable through invented records, while actual channel population and exchange executions remain unavailable.

Empty counts and invalid edits have different meanings

All-zero counts can describe an empty teaching batch but have no defined win percentage. Negative or fractional counts are invalid declarations and must clear the old output and disable export. Reset returns to the authored values, not a saved account record. The downloaded JSON is a conditional arithmetic result with its entered counts, not an authenticated provider audit. Do not paste credentials or private records. Software tests of these cases check the implementation only; they add no real trading evidence or strategy-validation samples.

A purchase decision needs additional objects

Request selected offer, lawful original history, receiver timing, quantities, costs and account valuation if those claims matter. A count may be perfectly reproducible while still omit the buyer's actual execution and capital path. A gross winner can lose after fees or financing, and many tiny winners can coexist with a large loss. Keep those questions beside the fraction. The site adds no live exchange connection, paid-room experiment or automatic monitor. Missing authentic evidence remains a specific next request rather than a fabricated score.

Complete call evidence and count policies