Every trading track record you’ve ever seen asks you to trust its author about one thing: that the rules weren’t adjusted along the way. Strategies quietly swapped after a bad month, windows chosen after the fact, losers left out of the tally. You can’t audit any of it, so the record is worth exactly as much as the author’s word.
For the final act of this project, we built a track record where the author’s word doesn’t matter.
Out-of-sample by construction #
On 2026-08-31 we signed a protocol — a public document in the repository — that froze everything an author could later fudge:
- The universe: exactly 9 of our 54 configurations, admitted by five pre-registered criteria
applied to the forward books (the selection beat a random-set control — claim
LIVE-ALLOWLIST-FROZEN). Not one slot was hand-picked. - The parameters: the deployed set, pinned by cryptographic hash into the signed document. The replay tooling refuses to run if the universe file differs by one byte from the signed hash.
- The rules: one contract always, the engine’s stated fill conventions, mechanical kill rules, and a “never-list” (no manual overrides, no window cherry-picking, no pausing to wait out a drawdown) — violations void the record’s claims from that point, by the protocol’s own text.
flowchart LR
P["✍️ signed protocol
2026-08-31, public in the repo"] --> U["THE UNIVERSE
9 of 54 configurations,
admitted by five pre-registered
criteria (beat a random-set control)"]
P --> PAR["THE PARAMETERS
pinned by cryptographic hash —
replay refuses to run on a
one-byte difference"]
P --> RU["THE RULES
1 contract · stated fills ·
mechanical kills · never-list
(violations void the record)"]
Because the parameters and universe were published before any future data existed, every recorded window is out-of-sample by construction. When new data arrives, it is first audited against the previous delivery for retroactive changes (a repaint check), then replayed with the frozen set. The result becomes a new claim in the machine-verified ledger — a losing window under exactly the same rules and prominence as a winning one. The protocol says so in writing: a negative outcome is a publishable result of the protocol, not a failure of it. No verdict is allowed before a pre-registered power threshold; interim windows are descriptive only.
Changes? Dated, attributed, append-only amendments — the signature claim’s falsifier literally counts them and checks their wording. There is no quiet edit.
flowchart TB
D["📦 new market data arrives"] --> A["repaint audit against the previous
delivery — any retroactive changes?"]
A --> H{"universe file hash ==
the signed protocol's hash?"}
H -->|"differs by one byte"| X["⛔ the replay tooling
refuses to run"]
H -->|"match"| RP["replay the 9 frozen configurations
under the frozen rules"]
RP --> CL["result becomes a new claim
in the machine-verified ledger"]
CL --> PUB["published — a losing window under
the same rules and prominence
as a winning one"]
PUB --> D
Where this honestly places us #
We wrote a positioning document (docs/POSITIONING.md in the repo) that grades this project rung
by rung against academic practice and trading-firm practice, each cell linked to its evidence. The
honest summary: on verification governance — pre-registration, powered negatives, machine-
replayable claims, a tamper-evident record — we operate at a standard we have not found in any
public trading research codebase. On proven live edge — the thing trading firms actually have —
our record is young by construction and says so; its first windows accumulate under the frozen
protocol, and per-slot verdicts need the sample sizes the protocol pre-computed. We claim the
method, and we let the record earn the rest at the only speed honesty allows.
The project, wrapped #
That closes the arc this series opened at the end-to-end overview: the famous strategy that failed 225 of 225 pre-registered configurations (paper: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=7428398); the one place in the economic calendar that pays; our own strategies kept to 17.6% by an honest forward window; the machinery that made all those sentences trustworthy (paper: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=7428478); and now a track record that continues on autopilot — auditable by anyone, cheatable by no one.
The whole thing is open under AGPL-3.0, DOI-archived (10.5281/zenodo.21473312), verifiable offline without any market data, and runnable end-to-end on your own data feed: github.com/mulhamfetna/trading-strategy-finder.
Built at BeInMedia (Nmo AI), Kuwait • Dubai • Doha, by Mulham Fetna, with thanks to Abd Ulfatah Esper for project leadership and support.
