Every closed trade since going live, in R-multiples, against the envelope fixed on 19 July 2026 — before the first trade closed. Closes publish after a 48-hour delay; open positions are a count only.
The goalposts were drawn first. Hover or tap the line for any trade's position in the envelope.
The bands are the prediction. On 19 July 2026, before the first trade closed, we ran the ruleset over its own outcome distribution and fixed the resulting percentiles — median, quartiles and 5–95th tails — as the yardstick. They have not been redrawn, re-fitted or re-scaled since.
A close below the 5th percentile is the pre-committed trigger for a full review. That trigger has fired once in this book’s life — the review was held on 19 August 2026 and recorded in the journal. No parameter was changed.
…
Matched against exchange fills before entering the record — currently …. Fees and funding included in the R figure.
Closes publish 48 hours after the fact; open positions show as a count only. This is a record, not a service.
One R is the loss taken if the initial stop is hit. A +6R trade is six times the risk it was given, on any capital base.
Reconciliation. Every trade on the record today is matched to exchange fills. If one ever cannot be, it is booked from engine marks and flagged as an estimate — never silently blended in.
The delay. 48 hours is not about obscuring anything — it is about making the record useless to copy, so the incentive stays on whether the numbers hold up over time.
R-multiples. Reporting in R makes the record independent of account size and impossible to flatter by quietly changing leverage. Dollar figures mostly communicate how much capital someone has.
Not investment advice. Past performance does not predict future results. We publish outcomes, not recommendations.