Methodology
How we grade
Generation, vetoes and settlement, as the code runs them.
Rules checked against the code on 22 Sep 2026
Research mode
No edge after costs is established. The signals stay public and the measurement keeps running.
| Counter start | 22 Sep 2026, 22:00 UTC |
|---|---|
| Entered positions | 0 |
| Target / stop, matured | 0 / 0 · — (n=0) |
| Net after fees | 0.00% over n=0 |
| Break-even | 0 |
| Closed by BTC veto | 0 |
Updated 23 Sep 2026, 04:49 UTC
The hack is transparency
The hack is transparency. Every signal page shows its DNA: the engines that fired, their agreement, the combo, the Evidence Score, the BTC context, any veto and the result net of fees. It is not a claim of edge.
Generation
| Markets | USDT pairs on Binance spot and USDT-M perpetual futures |
|---|---|
| Timeframe | 4-hour candles |
| Sources | 23 engines: 11 Smart Money Concepts, 8 indicators, 4 math models |
| Publication floor | Evidence Score 75 / 100 or higher |
| Entry window | 16 hours from publication |
| Exit plan | Stored with every signal: stop to entry at +0.5R, half closed at +1R when that level is nearer than the target, rest to the target |
Vetoes
| Before publication | 9 quality gates: crash guardrail, BTC dominance, adaptive source gate, directional risk, neutral-short, CUSUM, confirmation, concurrency cap, combo gate |
|---|---|
| Refused setups | Logged with the gate that refused them. Never published, never in the ledger |
| BTC veto, after publication | BTC trend turns and holds 60 minutes, or BTC moves 2% or more in 1 hour: open signals against BTC are closed |
| Veto closes | Kept out of the win rate and the headline net. Shown on their own line |
Settlement
| Price source | Binance 1-minute candles of the signal's own market, spot or futures |
|---|---|
| Entry | Filled when a candle inside the entry window trades at the entry price. The entry candle only fills |
| Straddling candle | A candle that touches both the stop and the target counts as the stop |
| Observation | Entry window plus 72 hours. Untouched positions close at the last price |
| Fees | 0.05% per side, 0.10% per trade |
| Funding | 0.02% per trade, paid by longs, received by shorts |
| Cost per trade | 0.12% on a long, 0.08% on a short |
| Maturity guard | A result enters a ratio only 96 hours after publication |
What counts as what
| Win | Target reached, or stop at entry after the +1R half was banked with a positive total |
|---|---|
| Loss | Stop hit before the stop moved to entry |
| Break-even | Stop at entry with nothing banked (within 0.05%). Not in the win rate, in the net |
| Expired, entered | No barrier touched in the window. Not in the win rate, in the net |
| Closed by BTC veto | Not in the win rate, not in the headline net |
| Not filled | Entry never traded. No money at risk, in no figure |
| Win rate | Matured wins / (matured wins + matured losses), always printed with n |
| Net | Sum of per-trade net moves in percentage points, not an account return |