System backtests: each Asset Class's cash rules against buy and hold
By Gareth David HoweAcutis Capital Founder
Each Asset Class on the dashboard has two cash gauges. When a gauge reads below −0.1, half of that Asset Class moves to cash; when both are below −0.1, all of it does. This article tests each pair on its own, on a single asset, and compares the result with simply buying and holding that asset.
- SPY (the S&P 500) traded with Financial Conditions and the Equities Regime, from January 2007.
- Bitcoin traded with Bitcoin Trend and the Bitcoin Regime, from January 2016.
- Gold traded with Gold Trend and the Macro Gold Regime, from January 2007.
- XLU (utility shares) traded with the Defensive Regime and Global Liquidity, from February 2008, when the Global Liquidity history begins.
SPY traded with Financial Conditions and Equities Regime: each signal below −0.1 moves 50% to cash.
CAGR
System
17.5%
Buy & hold
14.3%
Volatility
System
10.9%
Buy & hold
20.0%
Deepest drawdown
System
-9.4%
Buy & hold
-33.7%
Hover over the charts to see the values for a date.
Fall from previous high
Drag the handles to zoom; drag the selected window to scroll.
3 Jan 2020 – 16 Nov 2026
SPY: 3 Jan 2007 – 16 Nov 2026. Financial Conditions includes the Momentum early buy-back dates. Before the Equities Regime begins in January 2012, Financial Conditions alone sets SPY cash (red = 100% cash). The figures above use daily closes from the first date shown; the chart plots weekly points, so falls on it can look slightly smaller. Past results do not guarantee future results.
How the test works
The test reuses the signal histories and cash rules behind the dashboard and the model backtests. Each day's closing readings set the invested share for the next day, and the order fills at the next day's close. The invested share is 100%, 50% or 0%. Money in cash earns the US 3-month Treasury bill rate. Buy and hold stays fully invested throughout. Share prices include dividends; Gold uses the futures price.
Financial Conditions includes its Momentum early buy-backs. On seven dates (3 November 2014, 14 January 2019, 6 April 2020, 3 April 2023, 22 August 2024, 25 April 2025 and 15 April 2026) a sharp fall in market stress turns a red Financial Conditions reading green early. It stays green until the reading itself recovers or the next recorded Financial Conditions cash raise, whichever comes first. On 3 April 2023 and 15 April 2026 the reading had already recovered, so those dates change nothing.
There are two exceptions to the current rules. Before the Equities Regime history begins in January 2012, Financial Conditions alone decides SPY cash, as in the Dynamic Quartet's early history. XLU waits for the Global Liquidity history, so its test starts in February 2008 rather than January 2007.
What the results show
Figures here run to 16 November 2026. The charts update every Wednesday.
SPY. Since January 2007 the system grew 12.1% a year, compared with 10.7% for buy and hold. Its deepest fall was 9.4%, against 55.2% for buy and hold, with roughly half the volatility. It was invested about 60% of the time and changed its cash level about eight or nine times a year. The early buy-backs added about 0.7 points a year without deepening the worst fall.
Bitcoin. Since January 2016 the system grew 76.8% a year, compared with 58.5% for buy and hold. Its deepest fall was 59.6%, against 83.4%. Like SPY and XLU, the system beat buy and hold on both return and deepest fall over the full history. It also traded the most, about eleven cash changes a year.
Gold. Since January 2007 the system grew 9.5% a year, compared with 9.6% for buy and hold. Its deepest fall was 27.6%, against 44.4%. Gold changed cash level least often, about four times a year. Its Macro Gold Regime averages global inflation, US real yields and gold against the S&P 500 in equal thirds (from 17 November 2026).
XLU. Since February 2008 the system grew 8.7% a year, compared with 8.1% for buy and hold. Its deepest fall was 27.2%, against 43.8%, with about a third less volatility and about six cash changes a year. Its Defensive Regime gives global inflation half the weight and splits the rest equally between Treasury bond prices, utilities against the Nasdaq-100, Fed direction and utilities' dividend yield against Treasuries (from 17 November 2026).
For Gold, the system gave up a little yearly return in exchange for much smaller falls, which shows in its higher Sharpe and Calmar ratios. The SPY, Bitcoin and XLU systems added return as well. The period matters. Since January 2020 the systems beat buy and hold on yearly return for SPY (17.6% against 14.3%), Bitcoin (48.8% against 37.7%) and XLU (9.3% against 8.1%), and trailed it slightly for Gold (13.8% against 14.4%), with smaller deepest falls for all four. Use the Since 2007 and Since 2020 buttons above the chart to compare.

cutis
cutis