05 Verified 1 Year Benchmark

📈 05. Verified 1-Year Benchmark

To prevent data-mining and backtest overfitting, TriDomainMoE (Enhanced Live v2) was subjected to a continuous 1-year intra-bar backtest across 105,078 M5 bars (September 12, 2025 to September 12, 2026) using the high-resolution execution engine (src/backtest/bar_engine.py).

---

1. Backtest Simulation Parameters

• Instrument: BTCUSD (Bitcoin / US Dollar)
• Timeframe: M5 Continuous 24/7 Market
• Total Evaluated Bars: 105,078 M5 bars (365 calendar days)
• Starting Equity: $10,000.00 USD
• Risk per Trade: 0.10% ($10.00 cash risk baseline)
• Execution Model: Intra-bar tick emulation with recorded broker floating spreads ($\approx $65$ on BTC) and slippage penalty.
• Stop-Loss Execution: Conservative worst-case fill on adverse gap.

---

2. Master Statistical Evidence Summary

Performance Metric	Realized Value	Institutional Standard	Compliance Status
Total Executed Trades	443 trades	Statistical Significance ($N > 250$)	PASS
Win Rate	75.85% (336 W / 107 L)	$> 65.0%$	EXCEEDED
Profit Factor	3.37	$> 2.00$	CONFIRMED
Gross Profit / Gross Loss	+$2,156.51 / -$639.26	Positive Payoff Asymmetry	Superior Edge
Net Profit ($ / %)	+$1,517.25 (+15.17%)	Capital Growth	Robust Compounding
MAX TRAILING DRAWDOWN	0.3753% ($38.03)	Hard Prop Limit: < 2.50%	PASSED (6.6x Safety Cushion)
Annualized Sharpe Ratio	9.47	$&gt; 3.00$	INSTITUTIONAL GRADE
Annualized Sortino Ratio	39.05	$&gt; 5.00$	MINIMAL DOWNSIDE VOLATILITY
Calmar Ratio	40.43	$&gt; 10.0$	WORLD-CLASS CAPITAL PRESERVATION
Deflated Sharpe (DSR)	1.0000	$\ge 0.95$	Statistically Significant ($p &lt; 0.0001$)
Runs Test Z-Score	-0.578 ($p=0.56$)	$	Z
Max Consecutive Wins	25 consecutive trades	Momentum Runs	High Run Potential
Max Consecutive Losses	4 consecutive trades	Clamped Drawdown Cluster	Maximum 4 Losses

---

3. Monthly Compounding Breakdown (100% Calendar Consistency)

Every single calendar month achieved positive compounding returns:

Calendar Month	Closed Trades	Win Rate (%)	Profit Factor	Net PnL ($)	Return (%)	Ending Equity ($)
2025-09	18	66.67%	2.60	+$70.91	+0.71%	$10,070.91
2025-10	31	67.74%	2.64	+$139.82	+1.39%	$10,210.74
2025-11	48	75.00%	2.96	+$175.41	+1.72%	$10,386.15
2025-12	40	82.50%	4.52	+$148.94	+1.43%	$10,535.09
2026-01	36	83.33%	6.22	+$174.68	+1.66%	$10,709.78
2026-02	39	82.05%	4.41	+$123.00	+1.15%	$10,832.78
2026-03	43	88.37%	9.57	+$238.89	+2.21%	$11,071.67
2026-04	37	78.38%	3.74	+$120.81	+1.09%	$11,192.48
2026-05	34	79.41%	3.77	+$113.92	+1.02%	$11,306.40
2026-06	37	59.46%	2.20	+$90.80	+0.80%	$11,397.20
2026-07	37	64.86%	1.88	+$55.68	+0.49%	$11,452.87
2026-08	28	71.43%	1.48	+$19.82	+0.17%	$11,472.69
2026-09	15	80.00%	3.98	+$44.56	+0.39%	$11,517.25
TOTAL	443	75.85%	3.37	+$1,517.25	+15.17%	$11,517.25

---

4. Verified Data Artifacts

Raw trade logs and bar-by-bar equity curves are tracked directly in the repository:

• Trade Records: data/backtest/btcusd_1y_m5_enhanced_trades.parquet
• Equity Curve: data/backtest/btcusd_1y_m5_enhanced_equity.parquet