AlphaAssay puts every signal on trial — out of sample, against chance, against overfitting — and returns a signed verdict nobody can edit and anyone can verify. Proof, not promises.
| signal family | deflated sharpe | out-of-sample | verdict |
|---|---|---|---|
| momo-btc-4h | 0.41 | −38.2% | ✕ fail |
| meanrev-es-1d | 0.87 | −12.9% | ✕ fail |
| carry-fx-1w | 1.31 | +9.4% | ✓ pass |
Submit a signal or backtest — get a verdict with the reasons attached.
FROM $0.50 / CALL 02Timestamp the hypothesis before the data exists. Cannot be backfilled.
$0.10 + $0.25 / EVAL 03The full autopsy. Know where the edge died.
$2–5 / REPORT 04Export the signed verdict — credibility that travels.
$29–49 / CERTA no without a finding is an annoyance. A no with a finding is a shortcut — it saves the most expensive thing you own: weeks of searching in a dead direction.
AlphaAssay{
"verdict": "fail",
"died_at": "gate_2_family_deflation",
"failure_codes": [
"edge_collapses_at_lag1",
"family_budget_exhausted"
],
"placebo_percentile": 61, // beats 61% of 500 random twins
"survival_map": {
"delay_1bar": "dead",
"cost_stress": "survives",
"jackknife": "survives",
"regime_split": "dead",
"param_wiggle": "survives"
},
"search_budget_left": 4
}
We store only a one-way fingerprint and the cause of death — never the strategy itself.
Our verdicts can devalue a signal, never crown one. There is no list of winners to raid.
No exchange access, no execution, no custody. There is nothing here that could trade against you.