test: add comprehensive Hurst strategy simulations and backtest engine enhancements (v1.7.11)

- Add Hurst_simulations.md with detailed performance data from Jan 2025 to Mar 2026 across 1m-1h TFs. - Enhance backtest_engine.py: - Add --starting_equity for memory-efficient chunked testing. - Add Max Drawdown tracking. - Add --interval override flag. - Add --brake_period/--brake_interval for safety filter testing.
2026-03-10 13:02:29 +01:00
parent cd66a976de
commit b462651f60
2 changed files with 241 additions and 11 deletions
--- a/Hurst_simulations.md
+++ b/Hurst_simulations.md
@ -0,0 +1,180 @@
+# Comprehensive Hurst Strategy Simulations (since 2025-01-01)
+Comparison of different Hurst Timeframes and Entry Filters.
+
+# Timeframe: 1m
+
+## Scenario: Without Filter
+| Period | Start Bal | End Equity | PnL | Trades | Max DD |
+| :--- | :--- | :--- | :--- | :--- | :--- |
+| 2025-01-01 to 2025-03-31 | $1000.00 | $3192.86 | $2192.86 | 13239 | 7.40% |
+| 2025-04-01 to 2025-06-30 | $3192.86 | $5665.00 | $2472.14 | 13242 | 3.39% |
+| 2025-07-01 to 2025-09-30 | $5665.00 | $7520.61 | $1855.61 | 13720 | 1.02% |
+| 2025-10-01 to 2025-12-31 | $7520.61 | $8891.62 | $1371.01 | 13584 | 0.56% |
+| 2026-01-01 to 2026-03-10 | $8891.62 | $10437.95 | $1546.33 | 10120 | 0.52% |
+
+**Final Results for 1m (Without Filter):**
+- Final Equity: **$10437.95**
+- Total ROI: **943.80%**
+- Total Trades: **63905**
+- Max Overall Drawdown: **7.40%**
+
+## Scenario: With 1H SMA 200 Filter
+| Period | Start Bal | End Equity | PnL | Trades | Max DD |
+| :--- | :--- | :--- | :--- | :--- | :--- |
+| 2025-01-01 to 2025-03-31 | $1000.00 | $2124.83 | $1124.83 | 10469 | 7.57% |
+| 2025-04-01 to 2025-06-30 | $2124.83 | $3729.53 | $1604.70 | 11016 | 5.09% |
+| 2025-07-01 to 2025-09-30 | $3729.53 | $4673.10 | $943.57 | 10923 | 2.25% |
+| 2025-10-01 to 2025-12-31 | $4673.10 | $5428.01 | $754.91 | 11319 | 1.49% |
+| 2026-01-01 to 2026-03-10 | $5428.01 | $6307.05 | $879.04 | 8485 | 1.24% |
+
+**Final Results for 1m (With 1H SMA 200 Filter):**
+- Final Equity: **$6307.05**
+- Total ROI: **530.71%**
+- Total Trades: **52212**
+- Max Overall Drawdown: **7.57%**
+
+# Timeframe: 3m
+
+## Scenario: Without Filter
+| Period | Start Bal | End Equity | PnL | Trades | Max DD |
+| :--- | :--- | :--- | :--- | :--- | :--- |
+| 2025-01-01 to 2025-03-31 | $1000.00 | $2309.37 | $1309.37 | 4233 | 8.10% |
+| 2025-04-01 to 2025-06-30 | $2309.37 | $3801.67 | $1492.30 | 4331 | 2.07% |
+| 2025-07-01 to 2025-09-30 | $3801.67 | $4988.92 | $1187.25 | 4513 | 1.60% |
+| 2025-10-01 to 2025-12-31 | $4988.92 | $5912.62 | $923.70 | 4370 | 1.05% |
+| 2026-01-01 to 2026-03-10 | $5912.62 | $6740.90 | $828.28 | 3306 | 1.01% |
+
+**Final Results for 3m (Without Filter):**
+- Final Equity: **$6740.90**
+- Total ROI: **574.09%**
+- Total Trades: **20753**
+- Max Overall Drawdown: **8.10%**
+
+## Scenario: With 1H SMA 200 Filter
+| Period | Start Bal | End Equity | PnL | Trades | Max DD |
+| :--- | :--- | :--- | :--- | :--- | :--- |
+| 2025-01-01 to 2025-03-31 | $1000.00 | $1470.40 | $470.40 | 3288 | 10.19% |
+| 2025-04-01 to 2025-06-30 | $1470.40 | $2366.47 | $896.07 | 3597 | 4.96% |
+| 2025-07-01 to 2025-09-30 | $2366.47 | $2844.79 | $478.32 | 3478 | 3.23% |
+| 2025-10-01 to 2025-12-31 | $2844.79 | $3250.87 | $406.08 | 3597 | 3.43% |
+| 2026-01-01 to 2026-03-10 | $3250.87 | $3643.24 | $392.37 | 2739 | 1.90% |
+
+**Final Results for 3m (With 1H SMA 200 Filter):**
+- Final Equity: **$3643.24**
+- Total ROI: **264.32%**
+- Total Trades: **16699**
+- Max Overall Drawdown: **10.19%**
+
+# Timeframe: 5m
+
+## Scenario: Without Filter
+| Period | Start Bal | End Equity | PnL | Trades | Max DD |
+| :--- | :--- | :--- | :--- | :--- | :--- |
+| 2025-01-01 to 2025-03-31 | $1000.00 | $1878.15 | $878.15 | 2482 | 14.06% |
+| 2025-04-01 to 2025-06-30 | $1878.15 | $3053.49 | $1175.34 | 2601 | 3.26% |
+| 2025-07-01 to 2025-09-30 | $3053.49 | $3944.73 | $891.24 | 2689 | 2.08% |
+| 2025-10-01 to 2025-12-31 | $3944.73 | $4578.13 | $633.40 | 2491 | 1.33% |
+| 2026-01-01 to 2026-03-10 | $4578.13 | $5122.25 | $544.12 | 1966 | 1.39% |
+
+**Final Results for 5m (Without Filter):**
+- Final Equity: **$5122.25**
+- Total ROI: **412.23%**
+- Total Trades: **12229**
+- Max Overall Drawdown: **14.06%**
+
+## Scenario: With 1H SMA 200 Filter
+| Period | Start Bal | End Equity | PnL | Trades | Max DD |
+| :--- | :--- | :--- | :--- | :--- | :--- |
+| 2025-01-01 to 2025-03-31 | $1000.00 | $1304.55 | $304.55 | 1915 | 14.36% |
+| 2025-04-01 to 2025-06-30 | $1304.55 | $1926.70 | $622.15 | 2134 | 7.03% |
+| 2025-07-01 to 2025-09-30 | $1926.70 | $2253.45 | $326.75 | 2052 | 4.25% |
+| 2025-10-01 to 2025-12-31 | $2253.45 | $2539.78 | $286.33 | 2038 | 2.83% |
+| 2026-01-01 to 2026-03-10 | $2539.78 | $2774.98 | $235.20 | 1583 | 2.75% |
+
+**Final Results for 5m (With 1H SMA 200 Filter):**
+- Final Equity: **$2774.98**
+- Total ROI: **177.50%**
+- Total Trades: **9722**
+- Max Overall Drawdown: **14.36%**
+
+# Timeframe: 15m
+
+## Scenario: Without Filter
+| Period | Start Bal | End Equity | PnL | Trades | Max DD |
+| :--- | :--- | :--- | :--- | :--- | :--- |
+| 2025-01-01 to 2025-03-31 | $1000.00 | $1373.61 | $373.61 | 785 | 12.40% |
+| 2025-04-01 to 2025-06-30 | $1373.61 | $1824.22 | $450.61 | 725 | 6.62% |
+| 2025-07-01 to 2025-09-30 | $1824.22 | $2212.35 | $388.13 | 807 | 2.60% |
+| 2025-10-01 to 2025-12-31 | $2212.35 | $2535.13 | $322.78 | 765 | 3.14% |
+| 2026-01-01 to 2026-03-10 | $2535.13 | $2821.05 | $285.92 | 607 | 1.92% |
+
+**Final Results for 15m (Without Filter):**
+- Final Equity: **$2821.05**
+- Total ROI: **182.11%**
+- Total Trades: **3689**
+- Max Overall Drawdown: **12.40%**
+
+## Scenario: With 1H SMA 200 Filter
+| Period | Start Bal | End Equity | PnL | Trades | Max DD |
+| :--- | :--- | :--- | :--- | :--- | :--- |
+| 2025-01-01 to 2025-03-31 | $1000.00 | $993.20 | $-6.80 | 578 | 20.65% |
+| 2025-04-01 to 2025-06-30 | $993.20 | $1127.20 | $134.00 | 559 | 10.21% |
+| 2025-07-01 to 2025-09-30 | $1127.20 | $1328.46 | $201.26 | 595 | 2.96% |
+| 2025-10-01 to 2025-12-31 | $1328.46 | $1394.87 | $66.41 | 606 | 5.36% |
+| 2026-01-01 to 2026-03-10 | $1394.87 | $1379.42 | $-15.45 | 455 | 7.15% |
+
+**Final Results for 15m (With 1H SMA 200 Filter):**
+- Final Equity: **$1379.42**
+- Total ROI: **37.94%**
+- Total Trades: **2793**
+- Max Overall Drawdown: **20.65%**
+
+# Timeframe: 37m
+
+## Scenario: Without Filter
+| Period | Start Bal | End Equity | PnL | Trades | Max DD |
+| :--- | :--- | :--- | :--- | :--- | :--- |
+| 2025-01-01 to 2025-03-31 | $1000.00 | $1120.15 | $120.15 | 293 | 10.05% |
+| 2025-04-01 to 2025-06-30 | $1120.15 | $1487.65 | $367.50 | 282 | 6.98% |
+| 2025-07-01 to 2025-09-30 | $1487.65 | $1520.11 | $32.46 | 289 | 4.01% |
+| 2025-10-01 to 2025-12-31 | $1520.11 | $1575.24 | $55.13 | 292 | 9.73% |
+| 2026-01-01 to 2026-03-10 | $1575.24 | $1748.85 | $173.61 | 246 | 3.27% |
+
+**Final Results for 37m (Without Filter):**
+- Final Equity: **$1748.85**
+- Total ROI: **74.88%**
+- Total Trades: **1402**
+- Max Overall Drawdown: **10.05%**
+
+## Scenario: With 1H SMA 200 Filter
+| Period | Start Bal | End Equity | PnL | Trades | Max DD |
+| :--- | :--- | :--- | :--- | :--- | :--- |
+| 2025-01-01 to 2025-03-31 | $1000.00 | $1003.56 | $3.56 | 206 | 10.69% |
+| 2025-04-01 to 2025-06-30 | $1003.56 | $1130.86 | $127.30 | 208 | 8.98% |
+| 2025-07-01 to 2025-09-30 | $1130.86 | $1154.90 | $24.04 | 202 | 2.81% |
+| 2025-10-01 to 2025-12-31 | $1154.90 | $1108.85 | $-46.05 | 224 | 9.50% |
+| 2026-01-01 to 2026-03-10 | $1108.85 | $1117.52 | $8.67 | 179 | 4.01% |
+
+**Final Results for 37m (With 1H SMA 200 Filter):**
+- Final Equity: **$1117.52**
+- Total ROI: **11.75%**
+- Total Trades: **1019**
+- Max Overall Drawdown: **10.69%**
+
+# Timeframe: 1h
+
+## Scenario: Without Filter
+| Period | Start Bal | End Equity | PnL | Trades | Max DD |
+| :--- | :--- | :--- | :--- | :--- | :--- |
+| 2025-01-01 to 2025-03-31 | $1000.00 | $1109.99 | $109.99 | 162 | 7.36% |
+| 2025-04-01 to 2025-06-30 | $1109.99 | $1305.05 | $195.06 | 145 | 7.26% |
+| 2025-07-01 to 2025-09-30 | $1305.05 | $1342.48 | $37.43 | 180 | 2.83% |
+| 2025-10-01 to 2025-12-31 | $1342.48 | $1448.42 | $105.94 | 166 | 6.25% |
+| 2026-01-01 to 2026-03-10 | $1448.42 | $1552.57 | $104.15 | 135 | 4.56% |
+
+**Final Results for 1h (Without Filter):**
+- Final Equity: **$1552.57**
+- Total ROI: **55.26%**
+- Total Trades: **788**
+- Max Overall Drawdown: **7.36%**
+
--- a/src/strategies/backtest_engine.py
+++ b/src/strategies/backtest_engine.py
@ -13,8 +13,8 @@ from ping_pong_bot import PingPongStrategy
 load_dotenv()

 class BacktestEngine:
-    def __init__(self, config_path="config/ping_pong_config.yaml"):
-        self.version = "1.7.9"
+    def __init__(self, config_path="config/ping_pong_config.yaml", starting_equity=1000.0):
+        self.version = "1.7.11"
        with open(config_path, 'r') as f:
            self.config = yaml.safe_load(f)
        
@ -23,12 +23,17 @@ class BacktestEngine:
        self.strategy.direction = self.direction
        
        # Virtual Exchange State
-        self.balance = 1000.0  # Starting USD
-        self.equity = 1000.0
+        self.start_equity = starting_equity
+        self.balance = starting_equity
+        self.equity = starting_equity
        self.position_size = 0.0 # BTC
        self.position_value = 0.0 # USD
        self.entry_price = 0.0
        
+        # Performance Tracking
+        self.max_equity = starting_equity
+        self.max_drawdown = 0.0
+        
        # Settings
        self.fee_rate = 0.0005 # 0.05% Taker
        self.leverage = 5.0 # Will be updated based on mode
@ -39,6 +44,9 @@ class BacktestEngine:
        self.stop_loss_pct = 0.0 # 0.0 = Disabled
        self.stop_on_hurst_break = False
        
+        # Safety Brake Settings
+        self.use_brake = False
+        
        self.trades = []
        self.equity_curve = []

@ -75,7 +83,7 @@ class BacktestEngine:
        df[['open', 'high', 'low', 'close', 'volume']] = df[['open', 'high', 'low', 'close', 'volume']].astype(float)
        return df

-    def run(self, df, ma_df=None, ma_period=None):
+    def run(self, df, ma_df=None, ma_period=None, brake_df=None, brake_period=None):
        if df.empty:
            print("No data to run backtest.")
            return
@ -95,6 +103,17 @@ class BacktestEngine:
            ma_values = df['ma'].values
            print(f"Regime Switching enabled (MA {ma_period})")

+        # Prepare Brake MA
+        brake_values = None
+        if brake_df is not None and brake_period:
+            brake_df['brake_ma'] = brake_df['close'].rolling(window=brake_period).mean()
+            brake_subset = brake_df[['time', 'brake_ma']].rename(columns={'time': 'brake_time'})
+            df = pd.merge_asof(df.sort_values('time'), brake_subset.sort_values('brake_time'), 
+                               left_on='time', right_on='brake_time', direction='backward')
+            brake_values = df['brake_ma'].values
+            self.use_brake = True
+            print(f"Safety Brake enabled (MA {brake_period})")
+
        start_idx = max(self.config['rsi']['period'], self.config['hurst']['period'], 100)
        if start_idx >= len(df):
            print(f"Error: Not enough candles. Need {start_idx}, got {len(df)}")
@ -125,10 +144,19 @@ class BacktestEngine:
                    signal = None 

            if signal == "open":
-                self.open_position(price, time)
+                # Apply Safety Brake
+                if self.use_brake and brake_values is not None and not np.isnan(brake_values[i]):
+                    if self.direction == "short" and price > brake_values[i]:
+                        signal = None # Brake: Don't short in uptrend
+                    elif self.direction == "long" and price < brake_values[i]:
+                        signal = None # Brake: Don't long in downtrend
+                
+                if signal == "open":
+                    self.open_position(price, time)
            elif signal == "close" and abs(self.position_size) > 0:
                self.close_partial_position(price, time)

+            # Mark to Market Equity
            unrealized = 0
            if self.direction == "long":
                unrealized = self.position_size * (price - self.entry_price) if self.position_size > 0 else 0
@ -136,6 +164,15 @@ class BacktestEngine:
                unrealized = abs(self.position_size) * (self.entry_price - price) if self.position_size < 0 else 0
            
            self.equity = self.balance + unrealized
+            
+            # Max Drawdown Tracking
+            if self.equity > self.max_equity:
+                self.max_equity = self.equity
+            
+            dd = (self.max_equity - self.equity) / self.max_equity
+            if dd > self.max_drawdown:
+                self.max_drawdown = dd
+                
            self.equity_curve.append({"time": time, "equity": self.equity})

        self.print_results()
@ -186,8 +223,8 @@ class BacktestEngine:
        self.trades.append({"time": time, "type": reason, "price": price, "pnl": pnl, "fee": fee})

    def print_results(self):
-        total_pnl = self.equity - 1000.0
-        roi = (total_pnl / 1000.0) * 100
+        total_pnl = self.equity - self.start_equity
+        roi = (total_pnl / self.start_equity) * 100
        fees = sum(t['fee'] for t in self.trades)
        sl_hits = len([t for t in self.trades if "Stop Loss" in t['type']])
        print("\n" + "="*30)
@ -198,6 +235,7 @@ class BacktestEngine:
        print(f"Final Equity:  ${self.equity:.2f}")
        print(f"Total PnL:     ${total_pnl:.2f}")
        print(f"ROI:           {roi:.2f}%")
+        print(f"Max Drawdown:  {self.max_drawdown*100:.2f}%")
        print(f"Total Fees:    ${fees:.2f}")
        print("="*30)

@ -205,17 +243,21 @@ async def main():
    parser = argparse.ArgumentParser(description='Ping-Pong Strategy Backtester')
    parser.add_argument('--config', type=str, default='config/ping_pong_config.yaml')
    parser.add_argument('--limit', type=int, default=10000)
+    parser.add_argument('--interval', type=str, help='Strategy Interval (e.g. 5m, 15m)')
    parser.add_argument('--start_date', type=str)
    parser.add_argument('--end_date', type=str)
    parser.add_argument('--ma_period', type=int)
    parser.add_argument('--ma_interval', type=str, default='1h')
+    parser.add_argument('--brake_period', type=int, help='Safety Brake MA Period')
+    parser.add_argument('--brake_interval', type=str, default='1h', help='Safety Brake MA Interval')
    parser.add_argument('--direction', type=str, choices=['long', 'short'])
    parser.add_argument('--stop_loss', type=float, default=0.0, help='Stop Loss % (e.g. 0.02 for 2%)')
    parser.add_argument('--hurst_stop', action='store_true', help='Enable Stop Loss on Hurst break')
    parser.add_argument('--maker_fee', type=float, help='Override fee rate for Maker simulation (e.g. 0.0002)')
+    parser.add_argument('--starting_equity', type=float, default=1000.0, help='Initial balance')
    
    args = parser.parse_args()
-    engine = BacktestEngine(config_path=args.config)
+    engine = BacktestEngine(config_path=args.config, starting_equity=args.starting_equity)
    
    if args.maker_fee:
        engine.fee_rate = args.maker_fee
@ -224,19 +266,27 @@ async def main():
    engine.stop_loss_pct = args.stop_loss
    engine.stop_on_hurst_break = args.hurst_stop
    
+    # Base Data
    symbol = engine.config['symbol'].replace("USDT", "").replace("USD", "")
-    df = await engine.load_data(symbol, "1m", limit=args.limit, start_date=args.start_date, end_date=args.end_date)
+    data_interval = args.interval if args.interval else engine.config['interval']
+    if data_interval.isdigit(): data_interval += "m"
+    
+    df = await engine.load_data(symbol, data_interval, limit=args.limit, start_date=args.start_date, end_date=args.end_date)
    if df.empty: return

    ma_df = None
    if args.ma_period:
        ma_df = await engine.load_data(symbol, args.ma_interval, limit=5000, start_date=None, end_date=args.end_date)

+    brake_df = None
+    if args.brake_period:
+        brake_df = await engine.load_data(symbol, args.brake_interval, limit=5000, start_date=None, end_date=args.end_date)
+
    if args.direction:
        engine.direction = args.direction
        engine.strategy.direction = args.direction
    
-    engine.run(df, ma_df=ma_df, ma_period=args.ma_period)
+    engine.run(df, ma_df=ma_df, ma_period=args.ma_period, brake_df=brake_df, brake_period=args.brake_period)

 if __name__ == "__main__":
    asyncio.run(main())