AION Edge Audit Reports

The problem

Anyone can claim a track record. Almost nobody publishes the losses.

The standard practice in this industry is to show the winners, quietly drop the rest, and describe the remainder as a strategy. Screenshots prove nothing, backtests are trivially overfitted after the fact, and by the time a claim can be checked the audience has moved on.

Which leaves anyone evaluating a signal provider with no way to answer the only question that matters: when this was wrong, how wrong was it, and how often?

How it solves it

Every signal is recorded when it fires and scored against what actually happened afterwards. The full archive is published, not a selection from it.

Losses are in the table. They are not footnoted, aggregated into a favourable average, or removed once the period closes. The record is written as events occur rather than assembled later, so it cannot be tidied retrospectively.

This exists so you can be sceptical efficiently. You should not have to take a performance claim on trust, and with this you do not have to.

How to use it

Read the table

Each row is one signal with its outcome. Sort it, and look at the failures first — they are more informative than the successes.

Archive

A rolling 70-day window is retained and queryable.

Related

Veritas keeps its own separate daily scorecard at validation receipts.

Reading it honestly

• Past outcomes do not predict future ones. The archive tells you what happened, not what will.
• A period of good results can be luck. Judge the sample, not the streak.