LongTermTrend_V0897

REJECTEDhybrid

llm1) Added regime gates: only trade in UPTREND, avoid HIGH_VOLATILITY and non-NORMAL bubble regimes to reduce overfitting and noisy trades. 2) Simplified and rebalanced entry logic into three robust branches: (a) long-term trend (price > sma_200 with positive sma slopes), (b) volume-backed 50-day breakout (breakout_50 + rel_volume > 1.5 + macd_hist > 0), (c) momentum crossover (macd_hist > 0 and close > sma_50). This reduces reliance on extreme historic-return thresholds. 3) Improved exit: keep macd_hist < 0 but add a hard stop exit on drawdown < -8% to cap DD. 4) Position sizing: lower base_position to 0.8 and add volatility-based scaling rules to reduce size in elevated vol (vol_20 >4% -> 50%, >6% -> 30%). Overall aim: structural regime gating, safer exits, and volatility-aware sizing to improve walk-forward robustness. | rejected: fitness 0.026 below threshold

Fitness

0.0264

Generation

Complexity

Created

2026-09-30

Lineage

Parents (1)

LongTermTrend_V0844S0844 · gen 38ARCHIVEDllm

changed sma_200_slope threshold -0.0024 -> -0.0003; changed cumret_252d threshold -0.1997 -> 0.2888

Children (0)

None

Strategy Logic (pseudocode)

# LongTermTrend_V0897  [hybrid]
require regime.trend in ['UPTREND']
require regime.volatility in ['LOW_VOLATILITY', 'MEDIUM_VOLATILITY']
require regime.bubble in ['NORMAL']
if flat and ((close > sma_200 and sma_200_slope > 0.0001 and sma_50_slope > 0.0) or (breakout_50 == 1 and rel_volume > 1.5 and macd_hist > 0.0) or (macd_hist > 0.0 and close > sma_50)):
    position = 0.8
if long and (macd_hist < 0.0 or drawdown < -0.08):
    position = 0.0  # exit to cash

Backtest Metrics

Metric	Train	Validation	Hidden Test	Full
CAGR	-4.6%	-5.4%	—	—
Total Return	-31.3%	-15.4%	—	—
Sharpe	-0.36	-0.21	—	—
Sortino	-0.21	-0.13	—	—
Calmar	-0.13	-0.15	—	—
Volatility	11.3%	18.4%	—	—
Max DD	-35.9%	-36.3%	—	—
Win Rate	41.9%	39.7%	—	—
Avg Win	1.7%	2.9%	—	—
Avg Loss	-1.6%	-2.2%	—	—
Payoff	1.07	1.32	—	—
Expectancy	-0.2%	-0.2%	—	—
Profit Factor	0.77	0.87	—	—
Trades	160	63	—	—
Avg Holding Days	2.4	2.3	—	—
Turnover	32.1	33.6	—	—
Exposure	8.9%	8.6%	—	—

Walk-Forward Years

year-by-year stability matters more than any single year

Year	CAGR	Sharpe	Max DD	Trades
No walk-forward results

A robust strategy performs acceptably in most years rather than spectacularly in one — look for consistency across regimes, not a single outlier year.

Regime Performance

Period	Regime	Days	Ann. Return	Sharpe	Exposure
No regime results

Trades (full period)

0 trades

Entry	Exit	Entry Px	Exit Px	Return	Days
No trades recorded