Dual Momentum Allocation: Gold vs Bitcoin

Research into a rules-based tactical allocation model using dual momentum to rotate between physical gold and BTC. Focuses on systematic portfolio construction and risk-adjusted returns of 'digital gold'.

Strategy Decoder Editorial · 2026-05-06 · 4 min read

Key takeaways

A dual momentum strategy can systematically allocate between physical gold (GLD) and Bitcoin (IBIT) based on relative and absolute performance.
The strategy uses a lookback period (e.g., 4 to 12 weeks) to identify the stronger asset or revert to a cash position if neither shows positive momentum.
Integrating a volatility cap (e.g., 20% annualized) significantly reduces overall portfolio volatility and maximum drawdown, albeit at a cost to absolute returns.
Weekly rebalancing is chosen as a practical sweet spot, balancing signal responsiveness with transaction costs for these assets.
The study highlights the potential for systematic strategies to combine disparate assets like gold and Bitcoin for improved risk-adjusted outcomes.

Algorithmic and quantitative traders constantly seek robust strategies that can adapt to changing market conditions and manage diversified portfolios effectively. A recent study, as reported by QuantPedia, delves into a systematic approach for tactically allocating between physical gold and Bitcoin using a dual momentum framework. This research focuses on whether a rules-based model can leverage the distinct characteristics of these two assets, often referred to as 'digital gold' and 'physical gold,' to enhance portfolio performance and manage risk.

The Dual Nature of Store-of-Value Assets

Gold has long been recognized as a store of value, primarily due to its scarcity – new mining production adds only a small percentage to the existing global supply annually. Bitcoin, with its programmed scarcity and halving mechanism, mirrors this characteristic, leading many proponents to label it 'digital gold.' However, empirical observations often show Bitcoin exhibiting correlations with traditional risk assets during periods of market stress, challenging its role as a completely uncorrelated safe haven.

This tension between theoretical characteristics and real-world behavior forms the basis for investigating whether a systematic, momentum-driven allocation can harness the strengths of both assets while mitigating their individual weaknesses.

Data and Methodology for a Systematic Approach

The study utilized liquid, exchange-traded products for practical implementation: the SPDR Gold Trust (GLD) for physical gold exposure and the iShares Bitcoin Trust (IBIT) for Bitcoin. Data analysis spanned from December 31, 2018, to April 2026, encompassing various market cycles, including periods of significant volatility and growth in both assets.

A crucial aspect of the methodology was the data alignment and rebalancing frequency. Raw Bitcoin price data from Bitfinex was carefully matched to GLD's NYSE trading days and closing times to ensure consistent, point-in-time comparisons. The strategy adopted a weekly rebalancing schedule, specifically at Wednesday's close. This frequency was chosen as a balance between capturing momentum signals effectively in fast-moving markets (which daily rebalancing would overemphasize with high transaction costs) and avoiding sluggishness (which monthly rebalancing might incur).

Benchmark Performance

To contextualize the strategy's performance, three passive benchmarks were established:

Bitcoin Buy-and-Hold: Demonstrated high absolute returns but significantly higher volatility and deep drawdowns.
GLD Buy-and-Hold: Offered lower volatility and drawdowns but also commensurately lower returns.
50/50 Bitcoin/GLD Blend: Provided intermediate characteristics but still suffered substantial drawdowns, indicating that basic diversification alone isn't a silver bullet for managing the unique risks of these assets.

Dual Momentum Framework

The core of the strategy is built on a dual momentum framework, which incorporates both relative and absolute momentum principles. This systematic model has a single adjustable parameter: the lookback period (X) for momentum calculation, tested across various weekly intervals (e.g., 1, 2, 3, 4, 6, 8, 12, 20, 24, and 28 weeks).

At each weekly rebalancing point, the allocation rule is straightforward:

Long IBIT: If IBIT's return over the X-week period is greater than GLD's AND IBIT's return is positive.
Long GLD: If GLD's return over the X-week period is greater than IBIT's AND GLD's return is positive.
Flat (Cash): If neither asset meets its positive momentum and outperformance criteria.

This structure allows the strategy to tactically switch between Bitcoin, gold, or a cash position, never holding both simultaneously. The requirement for positive returns acts as an implicit downside protection mechanism, moving the portfolio to cash when neither asset demonstrates upward momentum.

Incorporating Volatility Targeting

Recognizing Bitcoin's inherently higher volatility compared to gold, the strategy introduced a crucial risk management component: a volatility cap. This mechanism aims to constrain portfolio risk to a maximum of 20% annualized volatility, acting as a hard upper bound rather than a flexible target.

At each rebalancing, once an asset is selected by the dual momentum signal, its 12-week rolling standard deviation is calculated and annualized. The position sizing is then determined by the formula: Position Weight = min(20% / Annualized Volatility). This ensures that if the selected asset's annualized volatility exceeds 20%, the allocation to that asset is reduced proportionally, with the remainder held in cash. This dynamically scales back exposure during high-volatility periods, directly confronting drawdown risk.

Performance Outcomes

The pure dual momentum strategy, without the volatility cap, showed a performance sweet spot between 4 and 12-week lookback periods. For instance, the 8-week variant delivered very high annualized returns and an impressive Sharpe ratio, significantly outperforming both the 50/50 blend and Bitcoin buy-and-hold benchmarks. However, the maximum drawdown remained substantial, approaching 50% in some cases, highlighting its inability to completely avoid systemic risks when both assets decline.

When the 20% volatility cap was introduced, the strategy's risk profile dramatically improved. The composite strategy's annualized volatility dropped significantly, even below that of standalone gold, and the maximum drawdown was substantially curtailed. While this came at the expense of absolute returns, the risk-adjusted metrics, such as the Sharpe and Calmar ratios, remained attractive. This demonstrates that systematic volatility control can effectively manage risk without completely sacrificing the strategy's underlying alpha generation capabilities.

Why it matters for algo traders

For algorithmic and quantitative traders, this research offers several critical insights. Firstly, it provides a solid foundation for designing systematic allocation strategies between traditionally disparate assets like gold and Bitcoin. The dual momentum framework, with its relative and absolute components, is a powerful tool for navigating market trends. Secondly, the meticulous approach to data handling – ensuring consistent, synchronized time series with liquid instruments – is a blueprint for robust backtesting and implementation. Thirdly, the effectiveness of volatility targeting as a dynamic risk management layer is particularly relevant. Algorithmic traders can integrate similar mechanisms to automatically adjust position sizing based on real-time volatility, not only for Bitcoin and gold but across any asset class. This can lead to more stable equity curves and better control over portfolio drawdowns, which are paramount in live trading environments. The study underscores that combining tactical allocation with intelligent risk controls can yield superior risk-adjusted performance for systematic portfolios.

Frequently asked questions

What is dual momentum in the context of this strategy?

Dual momentum is a strategy that combines relative momentum (choosing the better performing asset between two) and absolute momentum (only investing if the chosen asset has positive returns over a lookback period). If neither asset shows positive momentum, the strategy moves to cash.

How does volatility targeting improve the strategy's risk profile?

Volatility targeting caps the maximum annualized portfolio volatility at a predefined level (e.g., 20%). If the selected asset's volatility exceeds this cap, the strategy reduces its allocation to that asset, holding the remainder in cash, thereby systematically lowering exposure during high-risk periods.

Why was weekly rebalancing chosen over daily or monthly?

Weekly rebalancing was chosen to strike a balance between signal responsiveness and implementation feasibility. Daily rebalancing was deemed too costly due to transaction fees and noise, while monthly rebalancing was considered too slow to effectively capture momentum signals in volatile markets like cryptocurrency.

Tags: momentum, bitcoin, gold, tactical asset allocation

Based on reporting by QuantPedia.