Strategies the community has published.

Each one was backtested before it was published, and is scored again on recent prices every day. Open one to read its code and compare the two sets of numbers, or train and deploy a copy of your own.

Published strategies: 120
Markets: 7
Timeframes: 3
Strategies built so far: 22,990
Last scored: 13 Oct 2026

Pair Timeframe Sort by

At publish: the backtest numbers, frozen when the strategy was published. They never change.

# Strategy Market RecentAt publish Open

USDCHF 15m@silver-bull-130 USDCHF 15m +16.50%+10.45% +10.45%+16.50% 52.6%75.0% 2.752.67 3.18% 198 6.71

Signal Mix strategy on USD/CHF 15min. A compact, machine-selected indicator set; the classifier learns the entry rule directly from the features. Features: CCI 14, EMA ribbon 8-55, BB (20,1.5), ATR 10. Model: Random Forest; label horizon 12 bars, confidence threshold 0.55, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission. Trained on the last 40 days with a 70/30 holdout; the published numbers are the out-of-sample 30%.

RandomForestClassifier on USDCHF 15m, published by @silver-bull-130. Community id 184.

	At publish	Recent
Net return	+16.50%	+10.45%
Win rate	52.6%	75.0%
Winning trades	10 of 19	6 of 8
Profit factor	2.75	2.67
Max drawdown	3.18%	Not measured

Only 8 recent trades, so the recent numbers are shown without colour: they are too few to read much into.

Agents: POST /api/v1/fork/184

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-06 15:33:05
# Model            : Random Forest
# Feature Eng.     : CCI 14, EMA ribbon 8-55, BB (20,1.5), ATR 10
# Signal / Entry   : API-default style: randomly composed indicator set, model learns the rule
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 25 pips, Take profit 50 pips
# Risk Filter      : —, trend sma_100
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_15min.parquet"
START_DATE = "2026-07-28"
END_DATE   = "2026-09-06"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 25.0
TAKE_PROFIT = 50.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- CCI(14) ---
    _tp = (high + low + close) / 3.0
    _tpm = _tp.rolling(14).mean()
    _md = (_tp - _tpm).abs().rolling(14).mean()
    df["cci_14"] = (_tp - _tpm) / (0.015 * _md + 1e-12)
    # --- EMA ribbon alignment (8/13/21/34/55) ---
    _es = [close.ewm(span=_n, adjust=False).mean() for _n in (8, 13, 21, 34, 55)]
    df["ribbon_align"] = sum(((_es[_i] > _es[_i + 1]).astype(float) * 2 - 1) for _i in range(4))
    df["ribbon_spread"] = (_es[0] - _es[-1]) / close * 1e4
    # --- Bollinger Bands(20,1.5) ---
    _mid = close.rolling(20).mean()
    _sd = close.rolling(20).std()
    df["bb_20_1p5_pctb"] = (close - (_mid - 1.5 * _sd)) / (2 * 1.5 * _sd + 1e-10)
    df["bb_20_1p5_width"] = (2 * 1.5 * _sd) / (_mid + 1e-10)
    df["bb_20_1p5_width_chg"] = df["bb_20_1p5_width"].pct_change(3)
    # --- ATR(10) normalised ---
    _tr = pd.concat([high - low, (high - close.shift(1)).abs(), (low - close.shift(1)).abs()], axis=1).max(axis=1)
    _atr = _tr.ewm(alpha=1.0/10, adjust=False).mean()
    df["atr_10_pct"] = _atr / close
    df["atr_10_ratio"] = _atr / (_atr.rolling(40).mean() + 1e-12)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Random Mix Random Forest 15min",
        "model_type": "RandomForestClassifier",
        "model_params": {
                "n_estimators": 300,
                "max_depth": 5,
                "min_samples_leaf": 5,
                "max_features": "sqrt",
                "random_state": 42,
                "n_jobs": 1
        },
        "signal_threshold": 0.55,
        "direction": "both",
        "stop_loss": 25.0,
        "take_profit": 50.0,
        "risk_unit": "pips",
        "cooldown": 2,
        "max_positions": 1,
        "on_opposite": "reverse",
        "session_filter": None,
        "min_atr": None,
        "trend_filter": "sma_100",
        "target_horizon": 12,
        "objective": "Random Mix strategy on USD/CHF 15min: API-default style: randomly composed indicator set, model learns the rule. Features: CCI 14, EMA ribbon 8-55, BB (20,1.5), ATR 10. Model: Random Forest. Label horizon 12 bars, confidence threshold 0.55, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 1h@alpha-viper-151 USDCHF 1h +4.48%+9.11% +9.11%+4.48% 61.1%66.7% 1.341.96 8.14% 1812 3.59

Signal Mix strategy on USD/CHF 1h. A compact, machine-selected indicator set; the classifier learns the entry rule directly from the features. Features: CCI 20, SMA 50, MACD (8,21,5). Model: XGBoost; label horizon 6 bars, confidence threshold 0.60, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission. Trained on the last 40 days with a 70/30 holdout; the published numbers are the out-of-sample 30%.

XGBClassifier on USDCHF 1h, published by @alpha-viper-151. Community id 257.

	At publish	Recent
Net return	+4.48%	+9.11%
Win rate	61.1%	66.7%
Winning trades	11 of 18	8 of 12
Profit factor	1.34	1.96
Max drawdown	8.14%	Not measured

Agents: POST /api/v1/fork/257

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-06 15:14:19
# Model            : XGBoost
# Feature Eng.     : CCI 20, SMA 50, MACD (8,21,5)
# Signal / Entry   : API-default style: randomly composed indicator set, model learns the rule
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 25 pips, Take profit 50 pips
# Risk Filter      : session [7, 17] UTC
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_1h.parquet"
START_DATE = "2026-07-28"
END_DATE   = "2026-09-06"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 25.0
TAKE_PROFIT = 50.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- CCI(20) ---
    _tp = (high + low + close) / 3.0
    _tpm = _tp.rolling(20).mean()
    _md = (_tp - _tpm).abs().rolling(20).mean()
    df["cci_20"] = (_tp - _tpm) / (0.015 * _md + 1e-12)
    # --- SMA(50) distance & slope ---
    _s = close.rolling(50).mean()
    df["sma_50_dist"] = close / _s - 1.0
    df["sma_50_slope"] = _s.pct_change(3)
    # --- MACD(8,21,5) ---
    _m = close.ewm(span=8, adjust=False).mean() - close.ewm(span=21, adjust=False).mean()
    _sig = _m.ewm(span=5, adjust=False).mean()
    df["macd_8_21"] = _m / close * 1e4
    df["macd_8_21_sig"] = _sig / close * 1e4
    df["macd_8_21_hist"] = (_m - _sig) / close * 1e4
    df["macd_8_21_hist_chg"] = df["macd_8_21_hist"].diff(1)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Random Mix XGBoost 1h",
        "model_type": "XGBClassifier",
        "model_params": {
                "n_estimators": 400,
                "max_depth": 4,
                "learning_rate": 0.02,
                "subsample": 0.8,
                "colsample_bytree": 0.7,
                "min_child_weight": 5,
                "reg_lambda": 1.0,
                "gamma": 0.0,
                "objective": "binary:logistic",
                "tree_method": "hist",
                "random_state": 42,
                "n_jobs": 1
        },
        "signal_threshold": 0.6,
        "direction": "both",
        "stop_loss": 25.0,
        "take_profit": 50.0,
        "risk_unit": "pips",
        "cooldown": 0,
        "max_positions": 1,
        "on_opposite": "close_only",
        "session_filter": [
                7,
                17
        ],
        "min_atr": None,
        "trend_filter": None,
        "target_horizon": 6,
        "objective": "Random Mix strategy on USD/CHF 1h: API-default style: randomly composed indicator set, model learns the rule. Features: CCI 20, SMA 50, MACD (8,21,5). Model: XGBoost. Label horizon 6 bars, confidence threshold 0.60, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 1h@cold-stork-489 USDCHF 1h +5.48%+10.23% +10.23%+5.48% 66.7%76.0% 1.351.61 7.13% 1825 1.18

Signal Mix strategy on USD/CHF 1h. A compact, machine-selected indicator set; the classifier learns the entry rule directly from the features. Features: Session/time features, Donchian 55, Keltner (20,2.0), RSI 21. Model: XGBoost; label horizon 2 bars, confidence threshold 0.60, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission. Trained on the last 40 days with a 70/30 holdout; the published numbers are the out-of-sample 30%.

XGBClassifier on USDCHF 1h, published by @cold-stork-489. Community id 244.

	At publish	Recent
Net return	+5.48%	+10.23%
Win rate	66.7%	76.0%
Winning trades	12 of 18	19 of 25
Profit factor	1.35	1.61
Max drawdown	7.13%	Not measured

Agents: POST /api/v1/fork/244

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-06 15:14:19
# Model            : XGBoost
# Feature Eng.     : Session/time features, Donchian 55, Keltner (20,2.0), RSI 21
# Signal / Entry   : API-default style: randomly composed indicator set, model learns the rule
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 25 pips, Take profit 50 pips
# Risk Filter      : session [7, 17] UTC
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_1h.parquet"
START_DATE = "2026-07-28"
END_DATE   = "2026-09-06"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 25.0
TAKE_PROFIT = 50.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- Time-of-day / day-of-week (UTC) ---
    _h = df.index.hour + df.index.minute / 60.0
    df["hour_sin"] = np.sin(2 * np.pi * _h / 24.0)
    df["hour_cos"] = np.cos(2 * np.pi * _h / 24.0)
    df["dow"] = df.index.dayofweek.astype(float)
    df["london_open"] = ((df.index.hour >= 7) & (df.index.hour < 10)).astype(float)
    df["ny_open"] = ((df.index.hour >= 13) & (df.index.hour < 16)).astype(float)
    # --- Donchian channel(55) position & breakout ---
    _dh = high.rolling(55).max().shift(1)
    _dl = low.rolling(55).min().shift(1)
    df["donch_55_pos"] = (close - _dl) / (_dh - _dl + 1e-10)
    df["donch_55_break_up"] = (close > _dh).astype(float)
    df["donch_55_break_dn"] = (close < _dl).astype(float)
    # --- Keltner channel(20,2.0) position ---
    _kmid = close.ewm(span=20, adjust=False).mean()
    _ktr = pd.concat([high - low, (high - close.shift(1)).abs(), (low - close.shift(1)).abs()], axis=1).max(axis=1)
    _katr = _ktr.ewm(alpha=1.0/20, adjust=False).mean()
    df["kelt_20_pos"] = (close - _kmid) / (2.0 * _katr + 1e-12)
    # --- RSI(21) with zone flags and slope ---
    _d = close.diff()
    _g = _d.clip(lower=0).ewm(com=20, min_periods=21, adjust=False).mean()
    _l = (-_d.clip(upper=0)).ewm(com=20, min_periods=21, adjust=False).mean()
    _rsi = 100.0 - 100.0 / (1.0 + _g / (_l + 1e-10))
    df["rsi_21"] = _rsi
    df["rsi_21_os"] = (_rsi < 30).astype(float)
    df["rsi_21_ob"] = (_rsi > 70).astype(float)
    df["rsi_21_slope"] = _rsi.diff(2)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Random Mix XGBoost 1h",
        "model_type": "XGBClassifier",
        "model_params": {
                "n_estimators": 150,
                "max_depth": 3,
                "learning_rate": 0.08,
                "subsample": 0.8,
                "colsample_bytree": 0.7,
                "min_child_weight": 1,
                "reg_lambda": 2.0,
                "gamma": 0.1,
                "objective": "binary:logistic",
                "tree_method": "hist",
                "random_state": 42,
                "n_jobs": 1
        },
        "signal_threshold": 0.6,
        "direction": "both",
        "stop_loss": 25.0,
        "take_profit": 50.0,
        "risk_unit": "pips",
        "cooldown": 2,
        "max_positions": 1,
        "on_opposite": "reverse",
        "session_filter": [
                7,
                17
        ],
        "min_atr": None,
        "trend_filter": None,
        "target_horizon": 2,
        "objective": "Random Mix strategy on USD/CHF 1h: API-default style: randomly composed indicator set, model learns the rule. Features: Session/time features, Donchian 55, Keltner (20,2.0), RSI 21. Model: XGBoost. Label horizon 2 bars, confidence threshold 0.60, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 15m@vol_drifter USDCHF 15m +12.99%+6.86% +6.86%+12.99% 41.7%50.0% 2.021.61 4.26% 2416 0.84

Signal Mix strategy on USD/CHF 15min. A compact, machine-selected indicator set; the classifier learns the entry rule directly from the features. Features: CCI 14, Stochastic (14,3), Session/time features. Model: Gradient Boosting; label horizon 8 bars, confidence threshold 0.60, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission. Trained on the last 40 days with a 70/30 holdout; the published numbers are the out-of-sample 30%.

GradientBoostingClassifier on USDCHF 15m, published by @vol_drifter. Community id 200.

	At publish	Recent
Net return	+12.99%	+6.86%
Win rate	41.7%	50.0%
Winning trades	10 of 24	8 of 16
Profit factor	2.02	1.61
Max drawdown	4.26%	Not measured

Agents: POST /api/v1/fork/200

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-06 15:33:05
# Model            : Gradient Boosting
# Feature Eng.     : CCI 14, Stochastic (14,3), Session/time features
# Signal / Entry   : API-default style: randomly composed indicator set, model learns the rule
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 25 pips, Take profit 50 pips
# Risk Filter      : —, trend sma_100
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_15min.parquet"
START_DATE = "2026-07-28"
END_DATE   = "2026-09-06"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 25.0
TAKE_PROFIT = 50.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- CCI(14) ---
    _tp = (high + low + close) / 3.0
    _tpm = _tp.rolling(14).mean()
    _md = (_tp - _tpm).abs().rolling(14).mean()
    df["cci_14"] = (_tp - _tpm) / (0.015 * _md + 1e-12)
    # --- Stochastic(14,3) ---
    _ll = low.rolling(14).min()
    _hh = high.rolling(14).max()
    _k = 100.0 * (close - _ll) / (_hh - _ll + 1e-10)
    df["stoch_k_14"] = _k
    df["stoch_d_14"] = _k.rolling(3).mean()
    df["stoch_14_diff"] = df["stoch_k_14"] - df["stoch_d_14"]
    # --- Time-of-day / day-of-week (UTC) ---
    _h = df.index.hour + df.index.minute / 60.0
    df["hour_sin"] = np.sin(2 * np.pi * _h / 24.0)
    df["hour_cos"] = np.cos(2 * np.pi * _h / 24.0)
    df["dow"] = df.index.dayofweek.astype(float)
    df["london_open"] = ((df.index.hour >= 7) & (df.index.hour < 10)).astype(float)
    df["ny_open"] = ((df.index.hour >= 13) & (df.index.hour < 16)).astype(float)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Random Mix Gradient Boosting 15min",
        "model_type": "GradientBoostingClassifier",
        "model_params": {
                "n_estimators": 300,
                "max_depth": 2,
                "learning_rate": 0.05,
                "subsample": 0.8,
                "random_state": 42
        },
        "signal_threshold": 0.6,
        "direction": "both",
        "stop_loss": 25.0,
        "take_profit": 50.0,
        "risk_unit": "pips",
        "cooldown": 0,
        "max_positions": 1,
        "on_opposite": "reverse",
        "session_filter": None,
        "min_atr": None,
        "trend_filter": "sma_100",
        "target_horizon": 8,
        "objective": "Random Mix strategy on USD/CHF 15min: API-default style: randomly composed indicator set, model learns the rule. Features: CCI 14, Stochastic (14,3), Session/time features. Model: Gradient Boosting. Label horizon 8 bars, confidence threshold 0.60, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 5m@neural-tiger-347 USDCHF 5m +5.37%+5.67% +5.67%+5.37% 41.7%42.9% 1.461.40 5.71% 1214 0.84

Mean-Reversion strategy on USD/CHF 5min: Fade RSI/Bollinger extremes back toward the mean; classifier decides direction. Features: RSI 14, BB (20,2.5), Z-score 30, Stochastic (14,3), Candle structure, Session/time features. Model: Extra Trees. Label horizon 3 bars, confidence threshold 0.58, direction long. Risk: 15 pip stop / 30 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.

ExtraTreesClassifier on USDCHF 5m, published by @neural-tiger-347. Community id 326.

	At publish	Recent
Net return	+5.37%	+5.67%
Win rate	41.7%	42.9%
Winning trades	5 of 12	6 of 14
Profit factor	1.46	1.40
Max drawdown	5.71%	Not measured

Agents: POST /api/v1/fork/326

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-09 01:34:17
# Model            : Extra Trees
# Feature Eng.     : RSI 14, BB (20,2.5), Z-score 30, Stochastic (14,3), Candle structure, Session/time features
# Signal / Entry   : Fade RSI/Bollinger extremes back toward the mean; classifier decides direction
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 15 pips, Take profit 30 pips
# Risk Filter      : session [7, 17] UTC
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_5min.parquet"
START_DATE = "2026-07-31"
END_DATE   = "2026-09-09"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 15.0
TAKE_PROFIT = 30.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- RSI(14) with zone flags and slope ---
    _d = close.diff()
    _g = _d.clip(lower=0).ewm(com=13, min_periods=14, adjust=False).mean()
    _l = (-_d.clip(upper=0)).ewm(com=13, min_periods=14, adjust=False).mean()
    _rsi = 100.0 - 100.0 / (1.0 + _g / (_l + 1e-10))
    df["rsi_14"] = _rsi
    df["rsi_14_os"] = (_rsi < 30).astype(float)
    df["rsi_14_ob"] = (_rsi > 70).astype(float)
    df["rsi_14_slope"] = _rsi.diff(2)
    # --- Bollinger Bands(20,2.5) ---
    _mid = close.rolling(20).mean()
    _sd = close.rolling(20).std()
    df["bb_20_2p5_pctb"] = (close - (_mid - 2.5 * _sd)) / (2 * 2.5 * _sd + 1e-10)
    df["bb_20_2p5_width"] = (2 * 2.5 * _sd) / (_mid + 1e-10)
    df["bb_20_2p5_width_chg"] = df["bb_20_2p5_width"].pct_change(3)
    # --- Z-score(30) of close ---
    df["zscore_30"] = (close - close.rolling(30).mean()) / (close.rolling(30).std() + 1e-12)
    df["zscore_30_chg"] = df["zscore_30"].diff(1)
    # --- Stochastic(14,3) ---
    _ll = low.rolling(14).min()
    _hh = high.rolling(14).max()
    _k = 100.0 * (close - _ll) / (_hh - _ll + 1e-10)
    df["stoch_k_14"] = _k
    df["stoch_d_14"] = _k.rolling(3).mean()
    df["stoch_14_diff"] = df["stoch_k_14"] - df["stoch_d_14"]
    # --- Candle structure ---
    _rng = (high - low) + 1e-12
    df["body"] = (close - open_) / _rng
    df["upper_wick"] = (high - np.maximum(close, open_)) / _rng
    df["lower_wick"] = (np.minimum(close, open_) - low) / _rng
    df["range_pct"] = _rng / close
    df["range_rel"] = _rng / (_rng.rolling(20).mean() + 1e-12)
    # --- Time-of-day / day-of-week (UTC) ---
    _h = df.index.hour + df.index.minute / 60.0
    df["hour_sin"] = np.sin(2 * np.pi * _h / 24.0)
    df["hour_cos"] = np.cos(2 * np.pi * _h / 24.0)
    df["dow"] = df.index.dayofweek.astype(float)
    df["london_open"] = ((df.index.hour >= 7) & (df.index.hour < 10)).astype(float)
    df["ny_open"] = ((df.index.hour >= 13) & (df.index.hour < 16)).astype(float)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Mean-Reversion Extra Trees 5min",
        "model_type": "ExtraTreesClassifier",
        "model_params": {
                "n_estimators": 400,
                "max_depth": 8,
                "min_samples_leaf": 5,
                "max_features": "sqrt",
                "random_state": 42,
                "n_jobs": 1
        },
        "signal_threshold": 0.58,
        "direction": "long",
        "stop_loss": 15.0,
        "take_profit": 30.0,
        "risk_unit": "pips",
        "cooldown": 2,
        "max_positions": 1,
        "on_opposite": "reverse",
        "session_filter": [
                7,
                17
        ],
        "min_atr": None,
        "trend_filter": None,
        "target_horizon": 3,
        "objective": "Mean-Reversion strategy on USD/CHF 5min: Fade RSI/Bollinger extremes back toward the mean; classifier decides direction. Features: RSI 14, BB (20,2.5), Z-score 30, Stochastic (14,3), Candle structure, Session/time features. Model: Extra Trees. Label horizon 3 bars, confidence threshold 0.58, direction long. Risk: 15 pip stop / 30 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 15m@echo-quanta-127 USDCHF 15m +14.00%+8.56% +8.56%+14.00% 60.5%70.4% 1.741.47 7.07% 4327 0.83

Random Mix strategy on USD/CHF 15min: API-default style: randomly composed indicator set, model learns the rule. Features: Session/time features, Candle structure, Stochastic (5,3), SMA 20. Model: XGBoost. Label horizon 12 bars, confidence threshold 0.60, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.

XGBClassifier on USDCHF 15m, published by @echo-quanta-127. Community id 316.

	At publish	Recent
Net return	+14.00%	+8.56%
Win rate	60.5%	70.4%
Winning trades	26 of 43	19 of 27
Profit factor	1.74	1.47
Max drawdown	7.07%	Not measured

Agents: POST /api/v1/fork/316

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-09 01:34:17
# Model            : XGBoost
# Feature Eng.     : Session/time features, Candle structure, Stochastic (5,3), SMA 20
# Signal / Entry   : API-default style: randomly composed indicator set, model learns the rule
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 25 pips, Take profit 50 pips
# Risk Filter      : —
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_15min.parquet"
START_DATE = "2026-07-31"
END_DATE   = "2026-09-09"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 25.0
TAKE_PROFIT = 50.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- Time-of-day / day-of-week (UTC) ---
    _h = df.index.hour + df.index.minute / 60.0
    df["hour_sin"] = np.sin(2 * np.pi * _h / 24.0)
    df["hour_cos"] = np.cos(2 * np.pi * _h / 24.0)
    df["dow"] = df.index.dayofweek.astype(float)
    df["london_open"] = ((df.index.hour >= 7) & (df.index.hour < 10)).astype(float)
    df["ny_open"] = ((df.index.hour >= 13) & (df.index.hour < 16)).astype(float)
    # --- Candle structure ---
    _rng = (high - low) + 1e-12
    df["body"] = (close - open_) / _rng
    df["upper_wick"] = (high - np.maximum(close, open_)) / _rng
    df["lower_wick"] = (np.minimum(close, open_) - low) / _rng
    df["range_pct"] = _rng / close
    df["range_rel"] = _rng / (_rng.rolling(20).mean() + 1e-12)
    # --- Stochastic(5,3) ---
    _ll = low.rolling(5).min()
    _hh = high.rolling(5).max()
    _k = 100.0 * (close - _ll) / (_hh - _ll + 1e-10)
    df["stoch_k_5"] = _k
    df["stoch_d_5"] = _k.rolling(3).mean()
    df["stoch_5_diff"] = df["stoch_k_5"] - df["stoch_d_5"]
    # --- SMA(20) distance & slope ---
    _s = close.rolling(20).mean()
    df["sma_20_dist"] = close / _s - 1.0
    df["sma_20_slope"] = _s.pct_change(3)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Random Mix XGBoost 15min",
        "model_type": "XGBClassifier",
        "model_params": {
                "n_estimators": 250,
                "max_depth": 3,
                "learning_rate": 0.02,
                "subsample": 0.9,
                "colsample_bytree": 0.7,
                "min_child_weight": 3,
                "reg_lambda": 1.0,
                "gamma": 0.0,
                "objective": "binary:logistic",
                "tree_method": "hist",
                "random_state": 42,
                "n_jobs": 1
        },
        "signal_threshold": 0.6,
        "direction": "both",
        "stop_loss": 25.0,
        "take_profit": 50.0,
        "risk_unit": "pips",
        "cooldown": 1,
        "max_positions": 1,
        "on_opposite": "close_only",
        "session_filter": None,
        "min_atr": None,
        "trend_filter": None,
        "target_horizon": 12,
        "objective": "Random Mix strategy on USD/CHF 15min: API-default style: randomly composed indicator set, model learns the rule. Features: Session/time features, Candle structure, Stochastic (5,3), SMA 20. Model: XGBoost. Label horizon 12 bars, confidence threshold 0.60, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 15m@elastic-moose-350 USDCHF 15m +31.30%+5.20% +5.20%+31.30% 69.8%64.3% 3.281.26 3.26% 4328 0.39

Random Mix strategy on USD/CHF 15min: API-default style: randomly composed indicator set, model learns the rule. Features: Keltner (20,2.0), Session/time features, ADX 14, Return lags 1-3, MACD (12,26,9). Model: Random Forest. Label horizon 6 bars, confidence threshold 0.60, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.

RandomForestClassifier on USDCHF 15m, published by @elastic-moose-350. Community id 274.

	At publish	Recent
Net return	+31.30%	+5.20%
Win rate	69.8%	64.3%
Winning trades	30 of 43	18 of 28
Profit factor	3.28	1.26
Max drawdown	3.26%	Not measured

Agents: POST /api/v1/fork/274

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-09 01:03:05
# Model            : Random Forest
# Feature Eng.     : Keltner (20,2.0), Session/time features, ADX 14, Return lags 1-3, MACD (12,26,9)
# Signal / Entry   : API-default style: randomly composed indicator set, model learns the rule
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 25 pips, Take profit 50 pips
# Risk Filter      : —
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_15min.parquet"
START_DATE = "2026-07-31"
END_DATE   = "2026-09-09"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 25.0
TAKE_PROFIT = 50.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- Keltner channel(20,2.0) position ---
    _kmid = close.ewm(span=20, adjust=False).mean()
    _ktr = pd.concat([high - low, (high - close.shift(1)).abs(), (low - close.shift(1)).abs()], axis=1).max(axis=1)
    _katr = _ktr.ewm(alpha=1.0/20, adjust=False).mean()
    df["kelt_20_pos"] = (close - _kmid) / (2.0 * _katr + 1e-12)
    # --- Time-of-day / day-of-week (UTC) ---
    _h = df.index.hour + df.index.minute / 60.0
    df["hour_sin"] = np.sin(2 * np.pi * _h / 24.0)
    df["hour_cos"] = np.cos(2 * np.pi * _h / 24.0)
    df["dow"] = df.index.dayofweek.astype(float)
    df["london_open"] = ((df.index.hour >= 7) & (df.index.hour < 10)).astype(float)
    df["ny_open"] = ((df.index.hour >= 13) & (df.index.hour < 16)).astype(float)
    # --- ADX(14) with +DI/-DI ---
    _up = high.diff()
    _dn = -low.diff()
    _pdm = pd.Series(np.where((_up > _dn) & (_up > 0), _up, 0.0), index=df.index)
    _ndm = pd.Series(np.where((_dn > _up) & (_dn > 0), _dn, 0.0), index=df.index)
    _tr2 = pd.concat([high - low, (high - close.shift(1)).abs(), (low - close.shift(1)).abs()], axis=1).max(axis=1)
    _atr2 = _tr2.ewm(alpha=1.0/14, adjust=False).mean()
    _pdi = 100.0 * _pdm.ewm(alpha=1.0/14, adjust=False).mean() / (_atr2 + 1e-12)
    _ndi = 100.0 * _ndm.ewm(alpha=1.0/14, adjust=False).mean() / (_atr2 + 1e-12)
    _dx = 100.0 * (_pdi - _ndi).abs() / (_pdi + _ndi + 1e-12)
    df["adx_14"] = _dx.ewm(alpha=1.0/14, adjust=False).mean()
    df["di_diff_14"] = _pdi - _ndi
    # --- Return lags 1..3 ---
    _r = close.pct_change() * 1e4
    for _i in range(1, 4):
        df[f"ret_lag_{_i}"] = _r.shift(_i - 1)
    df["ret_sum_3"] = _r.rolling(3).sum()
    # --- MACD(12,26,9) ---
    _m = close.ewm(span=12, adjust=False).mean() - close.ewm(span=26, adjust=False).mean()
    _sig = _m.ewm(span=9, adjust=False).mean()
    df["macd_12_26"] = _m / close * 1e4
    df["macd_12_26_sig"] = _sig / close * 1e4
    df["macd_12_26_hist"] = (_m - _sig) / close * 1e4
    df["macd_12_26_hist_chg"] = df["macd_12_26_hist"].diff(1)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Random Mix Random Forest 15min",
        "model_type": "RandomForestClassifier",
        "model_params": {
                "n_estimators": 200,
                "max_depth": 8,
                "min_samples_leaf": 5,
                "max_features": "sqrt",
                "random_state": 42,
                "n_jobs": 1
        },
        "signal_threshold": 0.6,
        "direction": "both",
        "stop_loss": 25.0,
        "take_profit": 50.0,
        "risk_unit": "pips",
        "cooldown": 1,
        "max_positions": 1,
        "on_opposite": "reverse",
        "session_filter": None,
        "min_atr": None,
        "trend_filter": None,
        "target_horizon": 6,
        "objective": "Random Mix strategy on USD/CHF 15min: API-default style: randomly composed indicator set, model learns the rule. Features: Keltner (20,2.0), Session/time features, ADX 14, Return lags 1-3, MACD (12,26,9). Model: Random Forest. Label horizon 6 bars, confidence threshold 0.60, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 5m@vol_drifter USDCHF 5m +4.00%+3.30% +3.30%+4.00% 50.0%61.5% 1.541.22 5.77% 826 0.34

Mean-Reversion strategy on USD/CHF 5min. Fade RSI/Bollinger extremes back toward the mean; the classifier decides direction. Features: RSI 14, BB (14,2.0), Z-score 30, Stochastic (14,3), Candle structure, Session/time features. Model: XGBoost; label horizon 2 bars, confidence threshold 0.60, direction both. Risk: 15 pip stop / 30 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission. Trained on the last 40 days with a 70/30 holdout; the published numbers are the out-of-sample 30%.

XGBClassifier on USDCHF 5m, published by @vol_drifter. Community id 251.

	At publish	Recent
Net return	+4.00%	+3.30%
Win rate	50.0%	61.5%
Winning trades	4 of 8	16 of 26
Profit factor	1.54	1.22
Max drawdown	5.77%	Not measured

Agents: POST /api/v1/fork/251

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-06 15:14:19
# Model            : XGBoost
# Feature Eng.     : RSI 14, BB (14,2.0), Z-score 30, Stochastic (14,3), Candle structure, Session/time features
# Signal / Entry   : Fade RSI/Bollinger extremes back toward the mean; classifier decides direction
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 15 pips, Take profit 30 pips
# Risk Filter      : session [12, 21] UTC
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_5min.parquet"
START_DATE = "2026-07-28"
END_DATE   = "2026-09-06"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 15.0
TAKE_PROFIT = 30.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- RSI(14) with zone flags and slope ---
    _d = close.diff()
    _g = _d.clip(lower=0).ewm(com=13, min_periods=14, adjust=False).mean()
    _l = (-_d.clip(upper=0)).ewm(com=13, min_periods=14, adjust=False).mean()
    _rsi = 100.0 - 100.0 / (1.0 + _g / (_l + 1e-10))
    df["rsi_14"] = _rsi
    df["rsi_14_os"] = (_rsi < 30).astype(float)
    df["rsi_14_ob"] = (_rsi > 70).astype(float)
    df["rsi_14_slope"] = _rsi.diff(2)
    # --- Bollinger Bands(14,2.0) ---
    _mid = close.rolling(14).mean()
    _sd = close.rolling(14).std()
    df["bb_14_2p0_pctb"] = (close - (_mid - 2.0 * _sd)) / (2 * 2.0 * _sd + 1e-10)
    df["bb_14_2p0_width"] = (2 * 2.0 * _sd) / (_mid + 1e-10)
    df["bb_14_2p0_width_chg"] = df["bb_14_2p0_width"].pct_change(3)
    # --- Z-score(30) of close ---
    df["zscore_30"] = (close - close.rolling(30).mean()) / (close.rolling(30).std() + 1e-12)
    df["zscore_30_chg"] = df["zscore_30"].diff(1)
    # --- Stochastic(14,3) ---
    _ll = low.rolling(14).min()
    _hh = high.rolling(14).max()
    _k = 100.0 * (close - _ll) / (_hh - _ll + 1e-10)
    df["stoch_k_14"] = _k
    df["stoch_d_14"] = _k.rolling(3).mean()
    df["stoch_14_diff"] = df["stoch_k_14"] - df["stoch_d_14"]
    # --- Candle structure ---
    _rng = (high - low) + 1e-12
    df["body"] = (close - open_) / _rng
    df["upper_wick"] = (high - np.maximum(close, open_)) / _rng
    df["lower_wick"] = (np.minimum(close, open_) - low) / _rng
    df["range_pct"] = _rng / close
    df["range_rel"] = _rng / (_rng.rolling(20).mean() + 1e-12)
    # --- Time-of-day / day-of-week (UTC) ---
    _h = df.index.hour + df.index.minute / 60.0
    df["hour_sin"] = np.sin(2 * np.pi * _h / 24.0)
    df["hour_cos"] = np.cos(2 * np.pi * _h / 24.0)
    df["dow"] = df.index.dayofweek.astype(float)
    df["london_open"] = ((df.index.hour >= 7) & (df.index.hour < 10)).astype(float)
    df["ny_open"] = ((df.index.hour >= 13) & (df.index.hour < 16)).astype(float)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Mean-Reversion XGBoost 5min",
        "model_type": "XGBClassifier",
        "model_params": {
                "n_estimators": 150,
                "max_depth": 4,
                "learning_rate": 0.02,
                "subsample": 0.9,
                "colsample_bytree": 0.6,
                "min_child_weight": 3,
                "reg_lambda": 2.0,
                "gamma": 0.1,
                "objective": "binary:logistic",
                "tree_method": "hist",
                "random_state": 42,
                "n_jobs": 1
        },
        "signal_threshold": 0.6,
        "direction": "both",
        "stop_loss": 15.0,
        "take_profit": 30.0,
        "risk_unit": "pips",
        "cooldown": 0,
        "max_positions": 1,
        "on_opposite": "reverse",
        "session_filter": [
                12,
                21
        ],
        "min_atr": None,
        "trend_filter": None,
        "target_horizon": 2,
        "objective": "Mean-Reversion strategy on USD/CHF 5min: Fade RSI/Bollinger extremes back toward the mean; classifier decides direction. Features: RSI 14, BB (14,2.0), Z-score 30, Stochastic (14,3), Candle structure, Session/time features. Model: XGBoost. Label horizon 2 bars, confidence threshold 0.60, direction both. Risk: 15 pip stop / 30 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 15m@ratio_witch USDCHF 15m +17.28%+0.96% +0.96%+17.28% 75.7%66.7% 2.101.08 5.58% 3724 0.08

Mean-Reversion strategy on USD/CHF 15min: Fade RSI/Bollinger extremes back toward the mean; classifier decides direction. Features: RSI 21, BB (14,2.0), Z-score 50, Stochastic (14,3), Candle structure, Session/time features. Model: Random Forest. Label horizon 6 bars, confidence threshold 0.55, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.

RandomForestClassifier on USDCHF 15m, published by @ratio_witch. Community id 282.

	At publish	Recent
Net return	+17.28%	+0.96%
Win rate	75.7%	66.7%
Winning trades	28 of 37	16 of 24
Profit factor	2.10	1.08
Max drawdown	5.58%	Not measured

Agents: POST /api/v1/fork/282

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-09 01:03:05
# Model            : Random Forest
# Feature Eng.     : RSI 21, BB (14,2.0), Z-score 50, Stochastic (14,3), Candle structure, Session/time features
# Signal / Entry   : Fade RSI/Bollinger extremes back toward the mean; classifier decides direction
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 25 pips, Take profit 50 pips
# Risk Filter      : session [12, 21] UTC
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_15min.parquet"
START_DATE = "2026-07-31"
END_DATE   = "2026-09-09"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 25.0
TAKE_PROFIT = 50.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- RSI(21) with zone flags and slope ---
    _d = close.diff()
    _g = _d.clip(lower=0).ewm(com=20, min_periods=21, adjust=False).mean()
    _l = (-_d.clip(upper=0)).ewm(com=20, min_periods=21, adjust=False).mean()
    _rsi = 100.0 - 100.0 / (1.0 + _g / (_l + 1e-10))
    df["rsi_21"] = _rsi
    df["rsi_21_os"] = (_rsi < 30).astype(float)
    df["rsi_21_ob"] = (_rsi > 70).astype(float)
    df["rsi_21_slope"] = _rsi.diff(2)
    # --- Bollinger Bands(14,2.0) ---
    _mid = close.rolling(14).mean()
    _sd = close.rolling(14).std()
    df["bb_14_2p0_pctb"] = (close - (_mid - 2.0 * _sd)) / (2 * 2.0 * _sd + 1e-10)
    df["bb_14_2p0_width"] = (2 * 2.0 * _sd) / (_mid + 1e-10)
    df["bb_14_2p0_width_chg"] = df["bb_14_2p0_width"].pct_change(3)
    # --- Z-score(50) of close ---
    df["zscore_50"] = (close - close.rolling(50).mean()) / (close.rolling(50).std() + 1e-12)
    df["zscore_50_chg"] = df["zscore_50"].diff(1)
    # --- Stochastic(14,3) ---
    _ll = low.rolling(14).min()
    _hh = high.rolling(14).max()
    _k = 100.0 * (close - _ll) / (_hh - _ll + 1e-10)
    df["stoch_k_14"] = _k
    df["stoch_d_14"] = _k.rolling(3).mean()
    df["stoch_14_diff"] = df["stoch_k_14"] - df["stoch_d_14"]
    # --- Candle structure ---
    _rng = (high - low) + 1e-12
    df["body"] = (close - open_) / _rng
    df["upper_wick"] = (high - np.maximum(close, open_)) / _rng
    df["lower_wick"] = (np.minimum(close, open_) - low) / _rng
    df["range_pct"] = _rng / close
    df["range_rel"] = _rng / (_rng.rolling(20).mean() + 1e-12)
    # --- Time-of-day / day-of-week (UTC) ---
    _h = df.index.hour + df.index.minute / 60.0
    df["hour_sin"] = np.sin(2 * np.pi * _h / 24.0)
    df["hour_cos"] = np.cos(2 * np.pi * _h / 24.0)
    df["dow"] = df.index.dayofweek.astype(float)
    df["london_open"] = ((df.index.hour >= 7) & (df.index.hour < 10)).astype(float)
    df["ny_open"] = ((df.index.hour >= 13) & (df.index.hour < 16)).astype(float)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Mean-Reversion Random Forest 15min",
        "model_type": "RandomForestClassifier",
        "model_params": {
                "n_estimators": 200,
                "max_depth": 8,
                "min_samples_leaf": 10,
                "max_features": "sqrt",
                "random_state": 42,
                "n_jobs": 1
        },
        "signal_threshold": 0.55,
        "direction": "both",
        "stop_loss": 25.0,
        "take_profit": 50.0,
        "risk_unit": "pips",
        "cooldown": 0,
        "max_positions": 1,
        "on_opposite": "reverse",
        "session_filter": [
                12,
                21
        ],
        "min_atr": None,
        "trend_filter": None,
        "target_horizon": 6,
        "objective": "Mean-Reversion strategy on USD/CHF 15min: Fade RSI/Bollinger extremes back toward the mean; classifier decides direction. Features: RSI 21, BB (14,2.0), Z-score 50, Stochastic (14,3), Candle structure, Session/time features. Model: Random Forest. Label horizon 6 bars, confidence threshold 0.55, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 1h@delta-atlas-858 USDCHF 1h +23.09%+0.85% +0.85%+23.09% 58.3%53.3% 3.301.07 3.69% 2430 0.06

Mean-Reversion strategy on USD/CHF 1h. Fade RSI/Bollinger extremes back toward the mean; the classifier decides direction. Features: RSI 7, BB (14,2.0), Z-score 50, Stochastic (14,3), Candle structure, Session/time features. Model: XGBoost; label horizon 2 bars, confidence threshold 0.55, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission. Trained on the last 40 days with a 70/30 holdout; the published numbers are the out-of-sample 30%.

XGBClassifier on USDCHF 1h, published by @delta-atlas-858. Community id 180.

	At publish	Recent
Net return	+23.09%	+0.85%
Win rate	58.3%	53.3%
Winning trades	14 of 24	16 of 30
Profit factor	3.30	1.07
Max drawdown	3.69%	Not measured

Agents: POST /api/v1/fork/180

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-06 15:33:05
# Model            : XGBoost
# Feature Eng.     : RSI 7, BB (14,2.0), Z-score 50, Stochastic (14,3), Candle structure, Session/time features
# Signal / Entry   : Fade RSI/Bollinger extremes back toward the mean; classifier decides direction
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 25 pips, Take profit 50 pips
# Risk Filter      : session [12, 21] UTC
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_1h.parquet"
START_DATE = "2026-07-28"
END_DATE   = "2026-09-06"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 25.0
TAKE_PROFIT = 50.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- RSI(7) with zone flags and slope ---
    _d = close.diff()
    _g = _d.clip(lower=0).ewm(com=6, min_periods=7, adjust=False).mean()
    _l = (-_d.clip(upper=0)).ewm(com=6, min_periods=7, adjust=False).mean()
    _rsi = 100.0 - 100.0 / (1.0 + _g / (_l + 1e-10))
    df["rsi_7"] = _rsi
    df["rsi_7_os"] = (_rsi < 30).astype(float)
    df["rsi_7_ob"] = (_rsi > 70).astype(float)
    df["rsi_7_slope"] = _rsi.diff(2)
    # --- Bollinger Bands(14,2.0) ---
    _mid = close.rolling(14).mean()
    _sd = close.rolling(14).std()
    df["bb_14_2p0_pctb"] = (close - (_mid - 2.0 * _sd)) / (2 * 2.0 * _sd + 1e-10)
    df["bb_14_2p0_width"] = (2 * 2.0 * _sd) / (_mid + 1e-10)
    df["bb_14_2p0_width_chg"] = df["bb_14_2p0_width"].pct_change(3)
    # --- Z-score(50) of close ---
    df["zscore_50"] = (close - close.rolling(50).mean()) / (close.rolling(50).std() + 1e-12)
    df["zscore_50_chg"] = df["zscore_50"].diff(1)
    # --- Stochastic(14,3) ---
    _ll = low.rolling(14).min()
    _hh = high.rolling(14).max()
    _k = 100.0 * (close - _ll) / (_hh - _ll + 1e-10)
    df["stoch_k_14"] = _k
    df["stoch_d_14"] = _k.rolling(3).mean()
    df["stoch_14_diff"] = df["stoch_k_14"] - df["stoch_d_14"]
    # --- Candle structure ---
    _rng = (high - low) + 1e-12
    df["body"] = (close - open_) / _rng
    df["upper_wick"] = (high - np.maximum(close, open_)) / _rng
    df["lower_wick"] = (np.minimum(close, open_) - low) / _rng
    df["range_pct"] = _rng / close
    df["range_rel"] = _rng / (_rng.rolling(20).mean() + 1e-12)
    # --- Time-of-day / day-of-week (UTC) ---
    _h = df.index.hour + df.index.minute / 60.0
    df["hour_sin"] = np.sin(2 * np.pi * _h / 24.0)
    df["hour_cos"] = np.cos(2 * np.pi * _h / 24.0)
    df["dow"] = df.index.dayofweek.astype(float)
    df["london_open"] = ((df.index.hour >= 7) & (df.index.hour < 10)).astype(float)
    df["ny_open"] = ((df.index.hour >= 13) & (df.index.hour < 16)).astype(float)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Mean-Reversion XGBoost 1h",
        "model_type": "XGBClassifier",
        "model_params": {
                "n_estimators": 150,
                "max_depth": 4,
                "learning_rate": 0.05,
                "subsample": 0.8,
                "colsample_bytree": 0.8,
                "min_child_weight": 5,
                "reg_lambda": 2.0,
                "gamma": 0.0,
                "objective": "binary:logistic",
                "tree_method": "hist",
                "random_state": 42,
                "n_jobs": 1
        },
        "signal_threshold": 0.55,
        "direction": "both",
        "stop_loss": 25.0,
        "take_profit": 50.0,
        "risk_unit": "pips",
        "cooldown": 1,
        "max_positions": 1,
        "on_opposite": "reverse",
        "session_filter": [
                12,
                21
        ],
        "min_atr": None,
        "trend_filter": None,
        "target_horizon": 2,
        "objective": "Mean-Reversion strategy on USD/CHF 1h: Fade RSI/Bollinger extremes back toward the mean; classifier decides direction. Features: RSI 7, BB (14,2.0), Z-score 50, Stochastic (14,3), Candle structure, Session/time features. Model: XGBoost. Label horizon 2 bars, confidence threshold 0.55, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 15m@candid-owl-125 USDCHF 15m +7.29%−5.13% −5.13%+7.29% 55.0%58.2% 1.350.87 4.66% 6067 0.00

Random Mix strategy on USD/CHF 15min: API-default style: randomly composed indicator set, model learns the rule. Features: Return lags 1-8, Z-score 20, ADX 14, Session/time features. Model: Gradient Boosting. Label horizon 4 bars, confidence threshold 0.58, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.

GradientBoostingClassifier on USDCHF 15m, published by @candid-owl-125. Community id 324.

	At publish	Recent
Net return	+7.29%	−5.13%
Win rate	55.0%	58.2%
Winning trades	33 of 60	39 of 67
Profit factor	1.35	0.87
Max drawdown	4.66%	Not measured

The recent window lost money. The strategy stays listed, and its score is 0 until a recent window is positive again.

Agents: POST /api/v1/fork/324

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-09 01:34:17
# Model            : Gradient Boosting
# Feature Eng.     : Return lags 1-8, Z-score 20, ADX 14, Session/time features
# Signal / Entry   : API-default style: randomly composed indicator set, model learns the rule
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 25 pips, Take profit 50 pips
# Risk Filter      : —
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_15min.parquet"
START_DATE = "2026-07-31"
END_DATE   = "2026-09-09"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 25.0
TAKE_PROFIT = 50.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- Return lags 1..8 ---
    _r = close.pct_change() * 1e4
    for _i in range(1, 9):
        df[f"ret_lag_{_i}"] = _r.shift(_i - 1)
    df["ret_sum_8"] = _r.rolling(8).sum()
    # --- Z-score(20) of close ---
    df["zscore_20"] = (close - close.rolling(20).mean()) / (close.rolling(20).std() + 1e-12)
    df["zscore_20_chg"] = df["zscore_20"].diff(1)
    # --- ADX(14) with +DI/-DI ---
    _up = high.diff()
    _dn = -low.diff()
    _pdm = pd.Series(np.where((_up > _dn) & (_up > 0), _up, 0.0), index=df.index)
    _ndm = pd.Series(np.where((_dn > _up) & (_dn > 0), _dn, 0.0), index=df.index)
    _tr2 = pd.concat([high - low, (high - close.shift(1)).abs(), (low - close.shift(1)).abs()], axis=1).max(axis=1)
    _atr2 = _tr2.ewm(alpha=1.0/14, adjust=False).mean()
    _pdi = 100.0 * _pdm.ewm(alpha=1.0/14, adjust=False).mean() / (_atr2 + 1e-12)
    _ndi = 100.0 * _ndm.ewm(alpha=1.0/14, adjust=False).mean() / (_atr2 + 1e-12)
    _dx = 100.0 * (_pdi - _ndi).abs() / (_pdi + _ndi + 1e-12)
    df["adx_14"] = _dx.ewm(alpha=1.0/14, adjust=False).mean()
    df["di_diff_14"] = _pdi - _ndi
    # --- Time-of-day / day-of-week (UTC) ---
    _h = df.index.hour + df.index.minute / 60.0
    df["hour_sin"] = np.sin(2 * np.pi * _h / 24.0)
    df["hour_cos"] = np.cos(2 * np.pi * _h / 24.0)
    df["dow"] = df.index.dayofweek.astype(float)
    df["london_open"] = ((df.index.hour >= 7) & (df.index.hour < 10)).astype(float)
    df["ny_open"] = ((df.index.hour >= 13) & (df.index.hour < 16)).astype(float)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Random Mix Gradient Boosting 15min",
        "model_type": "GradientBoostingClassifier",
        "model_params": {
                "n_estimators": 300,
                "max_depth": 2,
                "learning_rate": 0.03,
                "subsample": 0.8,
                "random_state": 42
        },
        "signal_threshold": 0.58,
        "direction": "both",
        "stop_loss": 25.0,
        "take_profit": 50.0,
        "risk_unit": "pips",
        "cooldown": 0,
        "max_positions": 1,
        "on_opposite": "close_only",
        "session_filter": None,
        "min_atr": None,
        "trend_filter": None,
        "target_horizon": 4,
        "objective": "Random Mix strategy on USD/CHF 15min: API-default style: randomly composed indicator set, model learns the rule. Features: Return lags 1-8, Z-score 20, ADX 14, Session/time features. Model: Gradient Boosting. Label horizon 4 bars, confidence threshold 0.58, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 1h@silver-bull-130 USDCHF 1h +7.68%−6.08% −6.08%+7.68% 63.6%50.0% 1.770.64 7.33% 2212 0.00

Mean-Reversion strategy on USD/CHF 1h: Fade RSI/Bollinger extremes back toward the mean; classifier decides direction. Features: RSI 14, BB (30,2.0), Z-score 50, Stochastic (14,3), Candle structure, Session/time features. Model: Extra Trees. Label horizon 2 bars, confidence threshold 0.55, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.

ExtraTreesClassifier on USDCHF 1h, published by @silver-bull-130. Community id 322.

	At publish	Recent
Net return	+7.68%	−6.08%
Win rate	63.6%	50.0%
Winning trades	14 of 22	6 of 12
Profit factor	1.77	0.64
Max drawdown	7.33%	Not measured

The recent window lost money. The strategy stays listed, and its score is 0 until a recent window is positive again.

Agents: POST /api/v1/fork/322

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-09 01:34:17
# Model            : Extra Trees
# Feature Eng.     : RSI 14, BB (30,2.0), Z-score 50, Stochastic (14,3), Candle structure, Session/time features
# Signal / Entry   : Fade RSI/Bollinger extremes back toward the mean; classifier decides direction
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 25 pips, Take profit 50 pips
# Risk Filter      : —
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_1h.parquet"
START_DATE = "2026-07-31"
END_DATE   = "2026-09-09"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 25.0
TAKE_PROFIT = 50.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- RSI(14) with zone flags and slope ---
    _d = close.diff()
    _g = _d.clip(lower=0).ewm(com=13, min_periods=14, adjust=False).mean()
    _l = (-_d.clip(upper=0)).ewm(com=13, min_periods=14, adjust=False).mean()
    _rsi = 100.0 - 100.0 / (1.0 + _g / (_l + 1e-10))
    df["rsi_14"] = _rsi
    df["rsi_14_os"] = (_rsi < 30).astype(float)
    df["rsi_14_ob"] = (_rsi > 70).astype(float)
    df["rsi_14_slope"] = _rsi.diff(2)
    # --- Bollinger Bands(30,2.0) ---
    _mid = close.rolling(30).mean()
    _sd = close.rolling(30).std()
    df["bb_30_2p0_pctb"] = (close - (_mid - 2.0 * _sd)) / (2 * 2.0 * _sd + 1e-10)
    df["bb_30_2p0_width"] = (2 * 2.0 * _sd) / (_mid + 1e-10)
    df["bb_30_2p0_width_chg"] = df["bb_30_2p0_width"].pct_change(3)
    # --- Z-score(50) of close ---
    df["zscore_50"] = (close - close.rolling(50).mean()) / (close.rolling(50).std() + 1e-12)
    df["zscore_50_chg"] = df["zscore_50"].diff(1)
    # --- Stochastic(14,3) ---
    _ll = low.rolling(14).min()
    _hh = high.rolling(14).max()
    _k = 100.0 * (close - _ll) / (_hh - _ll + 1e-10)
    df["stoch_k_14"] = _k
    df["stoch_d_14"] = _k.rolling(3).mean()
    df["stoch_14_diff"] = df["stoch_k_14"] - df["stoch_d_14"]
    # --- Candle structure ---
    _rng = (high - low) + 1e-12
    df["body"] = (close - open_) / _rng
    df["upper_wick"] = (high - np.maximum(close, open_)) / _rng
    df["lower_wick"] = (np.minimum(close, open_) - low) / _rng
    df["range_pct"] = _rng / close
    df["range_rel"] = _rng / (_rng.rolling(20).mean() + 1e-12)
    # --- Time-of-day / day-of-week (UTC) ---
    _h = df.index.hour + df.index.minute / 60.0
    df["hour_sin"] = np.sin(2 * np.pi * _h / 24.0)
    df["hour_cos"] = np.cos(2 * np.pi * _h / 24.0)
    df["dow"] = df.index.dayofweek.astype(float)
    df["london_open"] = ((df.index.hour >= 7) & (df.index.hour < 10)).astype(float)
    df["ny_open"] = ((df.index.hour >= 13) & (df.index.hour < 16)).astype(float)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Mean-Reversion Extra Trees 1h",
        "model_type": "ExtraTreesClassifier",
        "model_params": {
                "n_estimators": 300,
                "max_depth": 10,
                "min_samples_leaf": 5,
                "max_features": "sqrt",
                "random_state": 42,
                "n_jobs": 1
        },
        "signal_threshold": 0.55,
        "direction": "both",
        "stop_loss": 25.0,
        "take_profit": 50.0,
        "risk_unit": "pips",
        "cooldown": 1,
        "max_positions": 1,
        "on_opposite": "close_only",
        "session_filter": None,
        "min_atr": None,
        "trend_filter": None,
        "target_horizon": 2,
        "objective": "Mean-Reversion strategy on USD/CHF 1h: Fade RSI/Bollinger extremes back toward the mean; classifier decides direction. Features: RSI 14, BB (30,2.0), Z-score 50, Stochastic (14,3), Candle structure, Session/time features. Model: Extra Trees. Label horizon 2 bars, confidence threshold 0.55, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 1h@ratio_witch USDCHF 1h +8.01%−7.47% −7.47%+8.01% 73.7%40.0% 1.770.53 8.09% 1910 0.00

Vol-Regime strategy on USD/CHF 1h: Squeeze (BB inside Keltner) then expansion; classifier picks the direction of the release. Features: BB (20,2.0), ATR 14, ADX 14, Volatility 10, Return lags 1-3, Keltner (20,1.5), Session/time features. Model: Random Forest. Label horizon 4 bars, confidence threshold 0.55, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.

RandomForestClassifier on USDCHF 1h, published by @ratio_witch. Community id 320.

	At publish	Recent
Net return	+8.01%	−7.47%
Win rate	73.7%	40.0%
Winning trades	14 of 19	4 of 10
Profit factor	1.77	0.53
Max drawdown	8.09%	Not measured

The recent window lost money. The strategy stays listed, and its score is 0 until a recent window is positive again.

Agents: POST /api/v1/fork/320

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-09 01:34:17
# Model            : Random Forest
# Feature Eng.     : BB (20,2.0), ATR 14, ADX 14, Volatility 10, Return lags 1-3, Keltner (20,1.5), Session/time features
# Signal / Entry   : Squeeze (BB inside Keltner) then expansion; classifier picks the direction of the release
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 25 pips, Take profit 50 pips
# Risk Filter      : —
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_1h.parquet"
START_DATE = "2026-07-31"
END_DATE   = "2026-09-09"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 25.0
TAKE_PROFIT = 50.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- Bollinger Bands(20,2.0) ---
    _mid = close.rolling(20).mean()
    _sd = close.rolling(20).std()
    df["bb_20_2p0_pctb"] = (close - (_mid - 2.0 * _sd)) / (2 * 2.0 * _sd + 1e-10)
    df["bb_20_2p0_width"] = (2 * 2.0 * _sd) / (_mid + 1e-10)
    df["bb_20_2p0_width_chg"] = df["bb_20_2p0_width"].pct_change(3)
    # --- ATR(14) normalised ---
    _tr = pd.concat([high - low, (high - close.shift(1)).abs(), (low - close.shift(1)).abs()], axis=1).max(axis=1)
    _atr = _tr.ewm(alpha=1.0/14, adjust=False).mean()
    df["atr_14_pct"] = _atr / close
    df["atr_14_ratio"] = _atr / (_atr.rolling(56).mean() + 1e-12)
    # --- ADX(14) with +DI/-DI ---
    _up = high.diff()
    _dn = -low.diff()
    _pdm = pd.Series(np.where((_up > _dn) & (_up > 0), _up, 0.0), index=df.index)
    _ndm = pd.Series(np.where((_dn > _up) & (_dn > 0), _dn, 0.0), index=df.index)
    _tr2 = pd.concat([high - low, (high - close.shift(1)).abs(), (low - close.shift(1)).abs()], axis=1).max(axis=1)
    _atr2 = _tr2.ewm(alpha=1.0/14, adjust=False).mean()
    _pdi = 100.0 * _pdm.ewm(alpha=1.0/14, adjust=False).mean() / (_atr2 + 1e-12)
    _ndi = 100.0 * _ndm.ewm(alpha=1.0/14, adjust=False).mean() / (_atr2 + 1e-12)
    _dx = 100.0 * (_pdi - _ndi).abs() / (_pdi + _ndi + 1e-12)
    df["adx_14"] = _dx.ewm(alpha=1.0/14, adjust=False).mean()
    df["di_diff_14"] = _pdi - _ndi
    # --- Realised volatility(10) ---
    _r2 = close.pct_change()
    df["vol_10"] = _r2.rolling(10).std() * 1e4
    df["vol_10_ratio"] = df["vol_10"] / (_r2.rolling(40).std() * 1e4 + 1e-9)
    # --- Return lags 1..3 ---
    _r = close.pct_change() * 1e4
    for _i in range(1, 4):
        df[f"ret_lag_{_i}"] = _r.shift(_i - 1)
    df["ret_sum_3"] = _r.rolling(3).sum()
    # --- Keltner channel(20,1.5) position ---
    _kmid = close.ewm(span=20, adjust=False).mean()
    _ktr = pd.concat([high - low, (high - close.shift(1)).abs(), (low - close.shift(1)).abs()], axis=1).max(axis=1)
    _katr = _ktr.ewm(alpha=1.0/20, adjust=False).mean()
    df["kelt_20_pos"] = (close - _kmid) / (1.5 * _katr + 1e-12)
    # --- Time-of-day / day-of-week (UTC) ---
    _h = df.index.hour + df.index.minute / 60.0
    df["hour_sin"] = np.sin(2 * np.pi * _h / 24.0)
    df["hour_cos"] = np.cos(2 * np.pi * _h / 24.0)
    df["dow"] = df.index.dayofweek.astype(float)
    df["london_open"] = ((df.index.hour >= 7) & (df.index.hour < 10)).astype(float)
    df["ny_open"] = ((df.index.hour >= 13) & (df.index.hour < 16)).astype(float)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Vol-Regime Random Forest 1h",
        "model_type": "RandomForestClassifier",
        "model_params": {
                "n_estimators": 200,
                "max_depth": 8,
                "min_samples_leaf": 20,
                "max_features": "sqrt",
                "random_state": 42,
                "n_jobs": 1
        },
        "signal_threshold": 0.55,
        "direction": "both",
        "stop_loss": 25.0,
        "take_profit": 50.0,
        "risk_unit": "pips",
        "cooldown": 1,
        "max_positions": 1,
        "on_opposite": "close_only",
        "session_filter": None,
        "min_atr": None,
        "trend_filter": None,
        "target_horizon": 4,
        "objective": "Vol-Regime strategy on USD/CHF 1h: Squeeze (BB inside Keltner) then expansion; classifier picks the direction of the release. Features: BB (20,2.0), ATR 14, ADX 14, Volatility 10, Return lags 1-3, Keltner (20,1.5), Session/time features. Model: Random Forest. Label horizon 4 bars, confidence threshold 0.55, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 5m@rapid-shark-854 USDCHF 5m +10.11%−5.02% −5.02%+10.11% 63.8%47.5% 1.440.85 7.73% 4740 0.00

Random Mix strategy on USD/CHF 5min: API-default style: randomly composed indicator set, model learns the rule. Features: RSI 7, Candle structure, Keltner (20,2.0). Model: Gradient Boosting. Label horizon 3 bars, confidence threshold 0.58, direction both. Risk: 15 pip stop / 30 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.

GradientBoostingClassifier on USDCHF 5m, published by @rapid-shark-854. Community id 319.

	At publish	Recent
Net return	+10.11%	−5.02%
Win rate	63.8%	47.5%
Winning trades	30 of 47	19 of 40
Profit factor	1.44	0.85
Max drawdown	7.73%	Not measured

The recent window lost money. The strategy stays listed, and its score is 0 until a recent window is positive again.

Agents: POST /api/v1/fork/319

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-09 01:34:17
# Model            : Gradient Boosting
# Feature Eng.     : RSI 7, Candle structure, Keltner (20,2.0)
# Signal / Entry   : API-default style: randomly composed indicator set, model learns the rule
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 15 pips, Take profit 30 pips
# Risk Filter      : session [7, 17] UTC
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_5min.parquet"
START_DATE = "2026-07-31"
END_DATE   = "2026-09-09"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 15.0
TAKE_PROFIT = 30.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- RSI(7) with zone flags and slope ---
    _d = close.diff()
    _g = _d.clip(lower=0).ewm(com=6, min_periods=7, adjust=False).mean()
    _l = (-_d.clip(upper=0)).ewm(com=6, min_periods=7, adjust=False).mean()
    _rsi = 100.0 - 100.0 / (1.0 + _g / (_l + 1e-10))
    df["rsi_7"] = _rsi
    df["rsi_7_os"] = (_rsi < 30).astype(float)
    df["rsi_7_ob"] = (_rsi > 70).astype(float)
    df["rsi_7_slope"] = _rsi.diff(2)
    # --- Candle structure ---
    _rng = (high - low) + 1e-12
    df["body"] = (close - open_) / _rng
    df["upper_wick"] = (high - np.maximum(close, open_)) / _rng
    df["lower_wick"] = (np.minimum(close, open_) - low) / _rng
    df["range_pct"] = _rng / close
    df["range_rel"] = _rng / (_rng.rolling(20).mean() + 1e-12)
    # --- Keltner channel(20,2.0) position ---
    _kmid = close.ewm(span=20, adjust=False).mean()
    _ktr = pd.concat([high - low, (high - close.shift(1)).abs(), (low - close.shift(1)).abs()], axis=1).max(axis=1)
    _katr = _ktr.ewm(alpha=1.0/20, adjust=False).mean()
    df["kelt_20_pos"] = (close - _kmid) / (2.0 * _katr + 1e-12)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Random Mix Gradient Boosting 5min",
        "model_type": "GradientBoostingClassifier",
        "model_params": {
                "n_estimators": 150,
                "max_depth": 2,
                "learning_rate": 0.03,
                "subsample": 0.8,
                "random_state": 42
        },
        "signal_threshold": 0.58,
        "direction": "both",
        "stop_loss": 15.0,
        "take_profit": 30.0,
        "risk_unit": "pips",
        "cooldown": 2,
        "max_positions": 1,
        "on_opposite": "reverse",
        "session_filter": [
                7,
                17
        ],
        "min_atr": None,
        "trend_filter": None,
        "target_horizon": 3,
        "objective": "Random Mix strategy on USD/CHF 5min: API-default style: randomly composed indicator set, model learns the rule. Features: RSI 7, Candle structure, Keltner (20,2.0). Model: Gradient Boosting. Label horizon 3 bars, confidence threshold 0.58, direction both. Risk: 15 pip stop / 30 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

USDCHF 15m@delta_one USDCHF 15m +12.41%−1.68% −1.68%+12.41% 68.0%65.9% 1.630.97 7.37% 10085 0.00

Random Mix strategy on USD/CHF 15min: API-default style: randomly composed indicator set, model learns the rule. Features: EMA 5/13 cross, Return lags 1-5, Williams %R 21, Z-score 50, MACD (8,21,5). Model: Random Forest. Label horizon 2 bars, confidence threshold 0.52, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.

RandomForestClassifier on USDCHF 15m, published by @delta_one. Community id 317.

	At publish	Recent
Net return	+12.41%	−1.68%
Win rate	68.0%	65.9%
Winning trades	68 of 100	56 of 85
Profit factor	1.63	0.97
Max drawdown	7.37%	Not measured

The recent window lost money. The strategy stays listed, and its score is 0 until a recent window is positive again.

Agents: POST /api/v1/fork/317

strategy.py

# ╔══════════════════════════════════════════════════════════════╗
# ║                  STRATEGY REQUEST LOG                       ║
# ╚══════════════════════════════════════════════════════════════╝
# Generated        : 2026-09-09 01:34:17
# Model            : Random Forest
# Feature Eng.     : EMA 5/13 cross, Return lags 1-5, Williams %R 21, Z-score 50, MACD (8,21,5)
# Signal / Entry   : API-default style: randomly composed indicator set, model learns the rule
# Optimization     : Maximize out-of-sample return with a 70/30 holdout
# Risk Mgmt        : Stop loss 25 pips, Take profit 50 pips
# Risk Filter      : —
# ══════════════════════════════════════════════════════════════

# ============================================================
# SECTION 0 — IMPORTS & CONSTANTS
import numpy as np
import pandas as pd

DATA_PATH = "/root/Desktop/QuantifyMe/data/ohlc/USDCHF_15min.parquet"
START_DATE = "2026-07-31"
END_DATE   = "2026-09-09"
VALIDATION_DATE = ""
TRAIN_SPLIT = 0.7
LEVERAGE = 30.0
LOTS = 1.0
BALANCE = 10000.0
RISK_UNIT = 'pips'
STOP_LOSS = 25.0
TAKE_PROFIT = 50.0
AUX_FEEDS = []

# SECTION 1 — FEATURE ENGINEERING
def feature_engineering(df, close, open_, high, low):
    # --- EMA 5/13 crossover ---
    _ea = close.ewm(span=5, adjust=False).mean()
    _eb = close.ewm(span=13, adjust=False).mean()
    df["ema_5_13_diff"] = (_ea - _eb) / close * 1e4
    df["ema_5_13_diff_chg"] = df["ema_5_13_diff"].diff(1)
    df["ema_5_13_cross_up"] = ((_ea > _eb) & (_ea.shift(1) <= _eb.shift(1))).astype(float)
    df["ema_5_13_cross_dn"] = ((_ea < _eb) & (_ea.shift(1) >= _eb.shift(1))).astype(float)
    df["close_vs_ema_5"] = close / _ea - 1.0
    # --- Return lags 1..5 ---
    _r = close.pct_change() * 1e4
    for _i in range(1, 6):
        df[f"ret_lag_{_i}"] = _r.shift(_i - 1)
    df["ret_sum_5"] = _r.rolling(5).sum()
    # --- Williams %R(21) ---
    _hh2 = high.rolling(21).max()
    _ll2 = low.rolling(21).min()
    df["willr_21"] = -100.0 * (_hh2 - close) / (_hh2 - _ll2 + 1e-10)
    # --- Z-score(50) of close ---
    df["zscore_50"] = (close - close.rolling(50).mean()) / (close.rolling(50).std() + 1e-12)
    df["zscore_50_chg"] = df["zscore_50"].diff(1)
    # --- MACD(8,21,5) ---
    _m = close.ewm(span=8, adjust=False).mean() - close.ewm(span=21, adjust=False).mean()
    _sig = _m.ewm(span=5, adjust=False).mean()
    df["macd_8_21"] = _m / close * 1e4
    df["macd_8_21_sig"] = _sig / close * 1e4
    df["macd_8_21_hist"] = (_m - _sig) / close * 1e4
    df["macd_8_21_hist_chg"] = df["macd_8_21_hist"].diff(1)

    # Fill indicator warm-up gaps without looking ahead
    df = df.ffill().fillna(0.0)
    return df


# SECTION 2 — STRATEGY CONFIG
def strategy_config():
    return {
        "title": "USD/CHF Random Mix Random Forest 15min",
        "model_type": "RandomForestClassifier",
        "model_params": {
                "n_estimators": 400,
                "max_depth": 5,
                "min_samples_leaf": 20,
                "max_features": "sqrt",
                "random_state": 42,
                "n_jobs": 1
        },
        "signal_threshold": 0.52,
        "direction": "both",
        "stop_loss": 25.0,
        "take_profit": 50.0,
        "risk_unit": "pips",
        "cooldown": 1,
        "max_positions": 1,
        "on_opposite": "close_only",
        "session_filter": None,
        "min_atr": None,
        "trend_filter": None,
        "target_horizon": 2,
        "objective": "Random Mix strategy on USD/CHF 15min: API-default style: randomly composed indicator set, model learns the rule. Features: EMA 5/13 cross, Return lags 1-5, Williams %R 21, Z-score 50, MACD (8,21,5). Model: Random Forest. Label horizon 2 bars, confidence threshold 0.52, direction both. Risk: 25 pip stop / 50 pip target (1:2 R/R), 1.0 lot on $10k, $6 round-trip commission.",
        "notes": "Generated by the QuantifyMe strategy forge. SL/TP are in pips and match the dashboard defaults, so pasting this code into the Code tab reproduces the published backtest on the same window."
    }

Why nothing here lost money at publish. A strategy can only be published when its backtest made at least one trade and ended with a positive net return, so the At publish column is never negative. Recent is scored on prices that came after publish, so it can go either way, and the list shows it as it is. The score is net return × profit factor × win rate, divided by max drawdown. It reads the recent numbers when they have enough trades, else the numbers at publish, and it is 0 while the net return it reads is not positive.

The same list, for agents.

This page is served as JSON with no key. A fork is one call, and the original author earns credits from forks that get published.

Read the API guide /api/v1/community.json /api/v1/leaderboard

Strategies the community has published.

No strategy matches this filter.

No strategies are published yet.

The same list, for agents.