Backtests · Package A

$100k became $1,301,279.

Simulated 02 Jan 2018 → 21 Aug 2026. Same engines as the paper desk. PEAD 45% / equity momentum 45% / risk parity 10% of 1.5× paper equity. Momentum book is 15 names.

Terminal NAV

$1,301,279

13.0× starting capital · 4.0× SPY wealth

CAGR

34.6%

Sharpe

1.08

Max drawdown

-36.0%

Leverage

1.5×

Vs SPY CAGR

14.6%

Momentum names

The path

Weekly marks. Package A at 1.5× against the unlevered book and SPY.

Updated 21 Aug 2026

Package A @ 1.5x
Package A @ 1.0x
SPY buy & hold

Year by year

Calendar returns for Package A at 1.5× versus SPY. 2022 is real. So is 2020.

2018
-1.3%SPY -5.2%
2019
+14.6%SPY +31.2%
2020
+163.0%SPY +18.3%
2021
+55.2%SPY +28.7%
2022
-20.5%SPY -18.2%
2023
+31.8%SPY +26.2%
2024
+66.4%SPY +24.9%
2025
+17.7%SPY +17.7%
2026
+37.5%SPY +12.9%

What changed the money

The mix shift is modest. Almost all of the extra terminal wealth is the 1.5× overlay versus the unlevered book — and versus just holding SPY.

Live target

Package A @ 1.5x

$1,301,279

CAGR: 34.6%
Sharpe: 1.08
Max DD: -36.0%

Compare

Package A @ 1.0x

$609,079

CAGR: 23.3%
Sharpe: 1.08
Max DD: -24.9%

Compare

Prior 48/32/20 @ 1.0x

$511,931

CAGR: 20.8%
Sharpe: 1.10
Max DD: -24.1%

Compare

SPY buy & hold

$324,520

CAGR: 14.6%
Sharpe: 0.80
Max DD: -33.7%

Sleeves

Standalone sleeve engines, then combined at the live weights. Plain English on Strategies.

PEAD

45%

Sharpe 0.99 · CAGR 21.2% · DD -33.8%

Equity momentum

45%

Sharpe 0.90 · CAGR 26.8% · DD -39.5%

Risk parity

10%

Sharpe 0.68 · CAGR 5.4% · DD -19.1%

Sleeve	Alloc	Sharpe	CAGR	Max DD	OOS Sharpe
PEAD	45%of 1.5× notional	0.99	21.2%	-33.8%	1.11
Equity momentum	45%of 1.5× notional	0.90	26.8%	-39.5%	1.25
Risk parity	10%of 1.5× notional	0.68	5.4%	-19.1%	1.33

Out of sample (from 2024)

Sharpe

1.33

CAGR

45.6%

Max DD

-31.8%

Vol

30.7%

Methodology

Daily sleeve returns from the PEAD, quality-momentum, and risk-parity engines that run the Alpaca equities paper book. Current mix is 45 / 45 / 10 of deployed notional at constant 1.5× gross. Transaction costs are inside each sleeve. Out-of-sample is 2024 onward. Curves start at $100,000.

Closer to realism

Next-day P&L on prior weights (no same-bar free lunch)
Costs scale with turnover inside each sleeve
Live rebalancer uses the same mix, name count, and 1.5× cap
Out-of-sample window published beside the full sample

Still not fully realistic

Daily close-to-close bars; no intraday path or partial fills
1.5× is constant leverage with no margin interest (paper)
Slippage is a flat percent — real impact rises with size
Paper fills are optimistic versus live capital

Questions on the research book, or soft interest as an allocator, builder, or learner?

Connect · Privacy