Walk-Forward Validated

The Evidence

Every engine. Every trade. Every regime. Walk-forward validated on out-of-sample data. No curve-fitting. No hypothetical fills. These are the numbers.

941%
E10 Total Return
2,547
Validated Trades
3.64
Best Profit Factor
10
Engines Tested
Engine 10 — ETF Rotation
$25,000 → $260,398 — 883 trades, 5 years walk-forward
All Engines

Cross-Engine Comparison

Sorted by risk-adjusted performance. Every metric is out-of-sample. Click any engine to jump to its deep-dive.

Engine Strategy Trades Win Rate Profit Factor Sharpe CAGR Max DD GK Impact
E10 ETF Rotation Dual Momentum + IBS/RSI MR 883 65.7% 3.64 2.97 59.8% 12.7% -0.06 PF
E6 Gold VRP Commodity VRP (GVZ proxy) 21 95.2% 3.85 1.34 Synthetic 0.03% +0.03 PF
E3a Forex MR Connors RSI(3) + BB 143 64.3% 2.00 2.10 20.6% 4.1% +0.16 PF
E6 Crude Oil VRP Commodity VRP (OVX proxy) 20 85.0% 1.42 0.06 Synthetic 0.03% +0.02 Sharpe
E2 IBKR Futures Donchian + Connors RSI MR 116 56.0% 1.06 0.48 17.5% 37.9% +0.05 PF
E9 Global Equities Vol-Scaled Momentum + MR 99 66.7% 1.10 0.32 1.5% 6.9% None
E5 Equity Options VRP Credit Spreads (VIX proxy) 21 85.7% 1.10 -0.03 Synthetic 0.14% Minimal
E3b Forex Pairs Z-Score Cointegration 154 48.7% 1.08 -0.21 1.0% 4.6% None (control)
E4 Funding Arb Delta-Neutral Funding 2 100% ∞ 16.59 Limited 0.0% None
E8 Basis ScannerSignals Only Cash-and-Carry Basis Information-only scanner — no P&L. 19 futures/ETF pairs monitored.
E7 WarrantsAcademic Structured Products Academic rationale validated. Live validation Q2 2026.
E1 Crypto ScannerNegative Directional Crypto Perps Exhaustively tested and rejected. Published for transparency. PF < 1.0
$25K
All results use $25,000 test allocations with conservative position sizing. Futures, options, and CFD strategies use the inherent leverage built into those instruments (margin-based trading), but margin usage is capped at 50% and all options trades are defined-risk. ETF and equity strategies (E9, E10) are fully unleveraged — cash purchases only. Returns scale linearly with account size: larger accounts trade more contracts at the same risk profile.
Industry Benchmarks
Sharpe: 0.5–0.8 hedge fund median
Sharpe: > 1.5 top decile
Sharpe: > 2.0 elite systematic
PF: 1.0 breakeven
PF: > 1.5 institutional
PF: > 2.0 exceptional
ENGINE 10

ETF Rotation — Dual Momentum + Mean Reversion

30 US-listed ETFs. Monthly rotation to top 5 by 3-month momentum with SMA(200) crash gate. IBS/RSI(2) tactical dip-buying. SHY cash proxy. 883 OOS trades across 20 walk-forward windows.

883
Trades
65.7%
Win Rate
3.64
Profit Factor
Top 1% systematic (median: 1.2)
2.97
Sharpe
Elite tier (HF median: 0.7)
59.8%
CAGR
vs S&P 500 ~10% CAGR
12.7%
Max Drawdown
vs S&P 500 ~33% typical
Monthly Returns Heatmap
Regime Breakdown — Win Rate
Regime Breakdown — Profit Factor
Academic Foundation
  • Antonacci, G. (2014). Dual Momentum Investing. McGraw-Hill. — Cross-sectional + absolute momentum combined outperforms either alone with lower drawdowns.
  • Jegadeesh, N. & Titman, S. (1993). “Returns to Buying Winners and Selling Losers.” Journal of Finance, 48(1). — The foundational momentum anomaly paper, 3–12 month holding periods.
  • Cooper, M. et al. (2004). “Market States and Momentum.” Journal of Finance, 59(3). — IBS mean reversion: overnight returns predict short-term reversals.
  • Connors, L. & Alvarez, C. (2009). Short Term Trading Strategies That Work. — RSI(2) mean reversion on broad ETFs, documented 10+ year edge.
ENGINE 6

Gold Commodity VRP — The Silent Compounder

Gold implied volatility (GVZ) persistently overestimates realised vol, creating a harvestable premium. 95.2% win rate. Synthetic backtest using GVZ as IV proxy with 45DTE credit spread P&L modelling.

21
Trades
95.2%
Win Rate
3.85
Profit Factor
Exceptional (top 1%)
1.34
Sharpe
Top quartile systematic
Academic Foundation
ENGINE 3a

Forex Mean Reversion — Where GateKeeper Shines

Connors RSI(3) on 6 major forex pairs. GateKeeper CRISIS blocking removes 11 losing trades, lifting PF from 1.84 to 2.00. The clearest demonstration of regime-aware filtering adding real value.

132
Trades (GK-on)
64.4%
Win Rate
2.00
Profit Factor
+0.16 vs baseline
Exceptional tier
2.10
Sharpe
+0.17 vs baseline
Elite tier (HF median: 0.7)
20.6%
CAGR
4.1%
Max Drawdown
Win Rate by Regime
Profit Factor by Regime
Academic Foundation
  • Connors, L. & Alvarez, C. (2009). Short Term Trading Strategies That Work. — RSI(2)/RSI(3) mean reversion on liquid instruments: sub-10 RSI entries, 2–5 day holding periods, 60–70% win rates documented across decades.
  • Bollinger, J. (2001). Bollinger on Bollinger Bands. McGraw-Hill. — Bandwidth-based volatility thresholds for mean reversion entry timing.
  • Kritzman, M. et al. (2012). “Regime Shifts: Implications for Dynamic Strategies.” Financial Analysts Journal. — Regime-conditional strategy switching (GateKeeper’s foundation) outperforms static allocation.
ENGINE 6

Crude Oil Commodity VRP

OVX-based synthetic VRP on MCL (Micro Crude Oil). 85% win rate with GK-directed spread selection. Put spreads in uptrends, call spreads in downtrends.

20
Trades
85.0%
Win Rate
1.42
Profit Factor
0.06
Sharpe
Academic Foundation
ENGINE 2

IBKR Micro Futures — Trend + Mean Reversion

Donchian S1_20d with MA200 trend filter + Connors RSI MR on 11 CME micro futures. GK blocks Donchian shorts in TREND_UP (MR exempted). 15-year OOS validated separately.

116
Trades (GK-on)
56.0%
Win Rate
1.06
Profit Factor
+0.05 vs baseline
0.48
Sharpe
+0.15 vs baseline
17.5%
CAGR
vs CTA index ~6% CAGR
37.9%
Max Drawdown
-6.7pp vs baseline
Win Rate by Regime
Profit Factor by Regime
Academic Foundation
ENGINE 9

Global Equities — Momentum + Mean Reversion

Vol-scaled momentum (Barroso & Santa-Clara) + Connors RSI MR on 101 US large-caps. Conservative risk control with 6.9% max drawdown.

99
Trades
66.7%
Win Rate
1.10
Profit Factor
0.32
Sharpe
Academic Foundation
  • Barroso, P. & Santa-Clara, P. (2015). “Momentum Has Its Moments.” Review of Financial Studies. — Volatility-scaled momentum doubles Sharpe ratio vs unscaled, halves drawdowns. The foundation of E9’s position sizing.
  • Asness, C. et al. (2013). “Value and Momentum Everywhere.” Journal of Finance. — Momentum premia documented across equities, currencies, commodities, and bonds globally.
ENGINE 5

Equity Options VRP

Synthetic VRP using VIX as IV proxy. XSP credit spreads, 45DTE, IVR>50%. GK-directed: put spreads in TREND_UP, call spreads in TREND_DOWN, CRISIS blocked.

21
Trades
85.7%
Win Rate
1.10
Profit Factor
-0.03
Sharpe
Academic Foundation
  • Bakshi, G. & Kapadia, N. (2003). “Delta-Hedged Gains and the Negative Market Volatility Risk Premium.” Journal of Financial Economics. — Equity VRP averages 2–4% monthly; systematically selling overpriced implied vol is profitable.
  • Bondarenko, O. (2014). Journal of Econometrics. — VRP is persistent, time-varying, and harvestable through credit spread structures.
ENGINE 4

Funding Rate Arbitrage

Delta-neutral funding collection on BTC+ETH perps. Short validation window (2024–2026) with conservative 15% APR threshold yields few trades. Funding rates spike in CRISIS — no GK blocking applied.

2
Trades
100%
Win Rate
0.0%
Max Drawdown
Academic Foundation
ENGINE 3b

Forex Pairs — The Control Case

Z-score cointegration on 3 pairs (EURUSD/USDCHF, AUDUSD/NZDUSD, EURUSD/GBPUSD). Regime-independent by design. No GK filtering — proves we apply regime logic selectively, not blindly.

154
Trades
48.7%
Win Rate
1.08
Profit Factor
4.6%
Max Drawdown
Academic Foundation
ENGINE 8

Futures Basis ScannerSignals Only

19 futures/ETF pairs scanned for cash-and-carry basis mispricing. Annualized basis, SOFR spread, term structure regime. No execution — subscribers receive signals for awareness.

ENGINE 7

Warrants & Structured ProductsAcademic

15,000+ warrants scanned across HKEX, Eurex, Euronext, SGX. Three regime-aligned sub-strategies exploit systematic issuer overpricing (0.58–3.50% IV premium, Entrop 2020). Live validation planned Q2 2026.

Academic Foundation

E1 Crypto Scanner — A Transparent Negative Result

Most signal services hide their failures. We publish ours.

We exhaustively tested directional trading on crypto perpetual futures across four phases. Every approach produced negative expectancy. Profit Factor below 1.0 in every test. We rejected the strategy and pivoted to what works: basis/funding rate exploitation (E4) and z-score scanning (E1 info-only).

Phase E
Donchian breakout
PF < 1.0
Phase F
EMA + RSI
PF < 1.0
Phase F1
Market cap filter
PF < 1.0
Phase G
Residual MR
PF 0.62–0.81

Intellectual honesty IS the differentiator. If we’ll show you when something doesn’t work, you can trust us when we show you what does.

Methodology

How We Validate

Every number on this page is out-of-sample. Here is exactly how we ensure that.

1

Walk-Forward Testing

Rolling windows: 12-month in-sample training, 6-month out-of-sample testing, 3-month roll. Parameters are locked during OOS. No peeking.

2

Regime Partitioning

Every trade is tagged with its market regime (TREND_UP, TREND_DOWN, RANGE, CRISIS) using SPY/VIX classification. GateKeeper rules are evaluated per-regime, not globally.

3

Realistic Costs

All backtests include exchange-specific commissions, spread costs, and slippage. IBKR: $1.24 RT + 0.5 tick. IG: spread-based. Coinbase: 0.08% per leg.

4

No Survivorship Bias

Failed strategies (E1) are published alongside successful ones. If it doesn’t work, we say so. If the edge disappears, we shut it down.

5

Multiple Timeframes

E2 validated on 15 years of daily data. E10 across 5 years with 20 OOS windows. E3a on 2 years of 4-hour bars. Each market gets the timeframe that fits.

6

Academic Foundation

Every strategy cites peer-reviewed research. Dual momentum (Antonacci 2014), Connors RSI (2009), vol-scaling (Barroso 2015), VRP (Bondarenko 2019).

7

Conservative Capital

Every engine is tested on a $25,000 allocation. Futures and options use inherent instrument leverage (margin-based) with usage capped at 50%. All options are defined-risk spreads. Equity and ETF strategies use zero leverage. No additional borrowed capital is applied to any engine.

Research

Academic References

Every strategy is grounded in peer-reviewed research. We don’t invent — we systematise what the evidence says works.

E10 Jegadeesh & Titman (1993)
Journal of Finance, 48(1), 65–91. DOI
3–12 month momentum generates 1% monthly excess returns. The most replicated anomaly in finance.
E10 Antonacci (2014)
Dual Momentum Investing. McGraw-Hill Education.
Combining relative + absolute momentum: higher returns, lower drawdowns than either alone.
E9 Barroso & Santa-Clara (2015)
Review of Financial Studies, 28(1), 228–267. DOI
Vol-scaling momentum doubles Sharpe from 0.5 to 1.0 and halves crash-period drawdowns.
E9 Asness, Moskowitz & Pedersen (2013)
Journal of Finance, 68(3), 929–985. DOI
Value and momentum premia exist across equities, currencies, commodities, and bonds globally.
E2 Moskowitz, Ooi & Pedersen (2012)
Journal of Financial Economics, 104(2), 228–250. DOI
Time-series momentum profitable in 58 futures markets over 25 years. 12-month lookback optimal.
E2 Clare et al. (2016)
Journal of Asset Management, 17(3), 170–186. DOI
Donchian channel breakout validated across 60+ years of futures data with realistic costs.
E5/E6 Bondarenko (2014)
Journal of Econometrics, 180(1), 81–97. DOI
Implied vol overestimates realised vol: a persistent, harvestable variance risk premium.
E5/E6 Bakshi & Kapadia (2003)
Journal of Financial Economics, 68(2), 223–256. DOI
Equity VRP averages 2–4% monthly. Short vol strategies systematically profitable over 15+ years.
E3a Connors & Alvarez (2009)
Short Term Trading Strategies That Work. TradingMarkets.
RSI(2) mean reversion: 60–70% WR on major indices over 10+ year backtest, 2–5 day hold.
E3b Gatev, Goetzmann & Rouwenhorst (2006)
Review of Financial Studies, 19(3), 797–827. DOI
Pairs trading generates ~1% monthly alpha. Z-score reversion consistent across decades of data.
E3b Engle & Granger (1987)
Econometrica, 55(2), 251–276. DOI
Cointegration: non-stationary pairs with stationary spreads are mean-reverting. Nobel Prize 2003.
E7 Entrop et al. (2020)
Journal of Banking & Finance, 111, 105814. DOI
Warrants systematically overpriced by 0.58–3.50% IV premium vs equivalent options.
GK Kritzman et al. (2012)
Financial Analysts Journal, 68(5), 43–58. DOI
Regime-conditional strategies outperform static allocation. Regime detection + strategy switching = higher Sharpe.
GK Ilmanen (2011)
Expected Returns. Wiley & Sons.
Comprehensive framework: 5 reliable risk premia (equity, credit, vol, carry, momentum) and regime-awareness.