← all research
STUDY

"Combined Book — Robustness & Significance Test"

2026-08-26

All combined stats over the overlap window 2022-12..2026-07 (~3.5 yrs, bull). Low correlation + higher combined Sharpe must hold across parameters and sub-periods to be believed. sqrt(365).

1. Bootstrap significance (combined 50/50 @1.5x)

Small sample -> p is suggestive, NOT conclusive.

2. Parameter sensitivity

2a. Gate60 (SMA, ROC) on crypto leg

SMAROCCrypto CAGRCrypto SharpeCrypto maxDDCorr w/ eq
15030+17.1%0.74-26.9%0.11
15060+22.2%0.84-23.5%0.13
15090+20.8%0.76-30.2%0.15
20030+18.6%0.81-21.1%0.11
20060+22.2%0.86-21.0%0.14
20090+23.6%0.85-23.1%0.15
25030+16.4%0.75-24.2%0.11
25060+21.4%0.84-21.0%0.14
25090+18.8%0.72-24.0%0.17

2b. Equity momentum (lookback L) leg

LEq CAGREq SharpeEq maxDDCorr w/ crypto
63+36.2%1.34-49.7%0.16
126+37.2%1.37-58.6%0.14
189+34.7%1.29-64.6%0.15
252+32.4%1.23-60.5%0.16

3. Sub-period consistency (combined 50/50 @1.5x)

PeriodCAGRSharpemaxDDCorr
2023+51.2%1.89-21.5%0.11
2024+80.1%1.91-31.9%0.24
2025+100.5%2.27-31.6%0.16
2026+138.7%2.36-31.0%nan

4. Diversification vs raw leverage

BookCAGRSharpemaxDD
50/50 @1.5x+78.7%1.99-31.9%
2x equity only+196.2%2.05-66.7%

If 50/50@1.5x matches 2x-equity CAGR at LOWER drawdown, the crypto leg earns its place.

Verdict