Raptor 2: Flip the Action

Full report

The complete backtest report and investment assessment for Raptor 2: naming guide, full numbers table, the five gates, the action-flip insight, honest limitations, the investability checklist, and the cleanup guide. Published on this site as the official record.

The exact change

In OnTick, the raw D1 bias is computed exactly as P2C computes it and fed unchanged into the detector. Only the detector's returned action is inverted before CreateSetup. Zero stays zero.

int rawBias = BiasDirection();
int baseSig = DetectPattern(rates, atr, rsiBuf[0], zoneBuf[0], rawBias);
int sig = -baseSig;

Base +1 BUY becomes -1 SELL, base -1 SELL becomes +1 BUY. The on-chart label shows Bias(PERIOD_D1) raw: UP | ACTION INVERTED so the inversion can never be misread.

Verification · 2026-09-30

Single-variable confirmed. Token-by-token diff against P2C: the action inversion is the only behavior change. Everything else is identical.

Source review: no issues found. In-house token diff plus an independent external MQL5 review (2026-09-30) found zero syntax problems across all 1,019 lines. Not a MetaEditor compile, that still needs F7 on a real terminal. Full independent report (PDF).

Status: compiled and backtested 2026-10-01. Verdict: RAPTOR, 5 of 5 gates on extended history.

What it does

• P2C BUY signal becomes P2E SELL: same pullback location, same ATR stop, mirrored to the short side.
• P2C SELL signal becomes P2E BUY: same pullback location, same ATR stop, mirrored to the long side.
• P2C no signal stays no signal.

Hypothesis

The detector is right about timing but wrong about direction. P2C faded pullbacks in the direction of the daily bias and lost. P2E takes the same moment and trades the other way, betting the pullback extends into a full reversal.

Unchanged from P2C

Raw D1 bias, pattern geometry, bar-shape and RSI filters, pullback zone, ATR stops, 1.5R / 3.0R targets, 75% partial, runner, breakeven +0.1R, scale-in, 1% risk, 12-bar pending expiry, 21:00 cutoff, 3% daily loss halt, max setups, both trading gates off. The existing BUY/SELL setup construction handles the mirrored SL/TP automatically. If P2E wins where P2C lost, it is purely the action flip.

Identity and safety

File

Blackfish_EA_P2E.mq5

Base

P2C (Blackfish_EA_P2C.mq5). P2B abandoned.

Magic number

20260931

Trading gates

InpLiveTrading = false, InpAllowOrders = false (defaults)

Demo only

No live trading. No real money.

Experiment status

Gates · 5 of 5 PASS

First variant to clear every gate. Backtest only, demo account, no live capital. Next: frozen-parameter out-of-sample, then forward demo.

Backtest results · extended history · 2026-10-01

EURUSD H1, 2023-10-01 to 2026-09-22, every tick based on real ticks, 18,460 bars, 99% history quality. Tester gates InpLiveTrading=true, InpAllowOrders=true.

Total trades

142

Win rate

29.58%

Profit factor

1.10

Net

+$922.19 (+9.22%)

Expected payoff

+6.49

Max drawdown

18.87%

Gross profit / loss

+$10,399.87 / -$9,477.68

Verdict: RAPTOR. The first run (2025-01-01 to 2026-09-22) gave 78 trades, PF 1.15, net +7.35%, passing 4 of 5 gates but short of the 100-trade minimum. Extending history back to 2023-10-01 brought 142 trades and cleared all five gates. The edge held across the longer window.

Download

Complete .mq5 source. Compile in MetaEditor before attaching to any chart.