Golden Dual-Momentum Rotation
A Systematic Way to Trade Bitcoin and Gold
Gold has been humanity’s go-to store of value for thousands of years. Bitcoin, with its built-in scarcity and fixed supply schedule, has been dubbed “digital gold” by its proponents. But should you own one, the other or both and in what proportion?
A recent paper by Vojtko and Dujava proposes an elegant, rules-based answer: don’t pick one and stick with it forever. Instead, let momentum signals tell you which one to own at any given time or whether to own neither at all.
I recommend reading the paper by Vojtko and Dujava because it’s the basis for this post.
TL;DR:
Bitcoin delivers spectacular returns, but the ride is brutal — a nearly 78% peak-to-trough loss would test the resolve of even the most committed long-term investor. Gold is far more stable but leaves significant return on the table. And the intuitive “split the difference” approach of a 50/50 portfolio doesn’t really solve anything: you give up a meaningful chunk of Bitcoin’s returns while still suffering a gut-wrenching 51% drawdown.
Simply blending the two assets doesn’t resolve the fundamental tension between them.
The proposed solution draws on a well-established concept in quantitative finance called dual momentum, originally developed by Gary Antonacci. The idea combines two distinct momentum signals:
Relative momentum: Which of the two assets has performed better recently?
Absolute momentum: Has the winning asset (step 1) actually gone up or just fallen less than the other?
The allocation rule is straightforward:
Look back over a set number of weeks
Own Bitcoin if Bitcoin has outperformed gold and Bitcoin’s return is positive
Own Gold if gold has outperformed Bitcoin and gold’s return is positive
Hold cash if neither asset has posted a positive return over the lookback period
Testing the Strategy
For the Bitcoin allocation, the analysis uses two ETFs to reflect the available investment vehicles over different periods. From 2018 through early 2024, Bitcoin exposure is represented by GBTC, while from 2024 onward the analysis transitions to IBIT following the launch of spot Bitcoin ETFs in the United States. Gold exposure is represented by GLD. Rather than holding pure and simple cash, the strategy allocates it to SHV, which tracks short-term U.S. Treasury securities and serves as a low-risk cash equivalent with yield generation.
Momentum is evaluated using multiple lookback periods. Specifically, the strategy measures momentum over 11, 21, 32, 42, 63, and 84 trading days. In addition to the individual momentum strategies, the analysis also includes a composite strategy that combines the 21-, 42-, and 63-day lookback periods. Rather than relying on one isolated measurement period, the composite strategy aims to create a more robust strategy that mitigates parameter sensitivity and may perform more consistently across changing market conditions.
Following the approach of Vojtko and Dujava, the analysis will evaluate two different versions of the strategy. The first version does not apply any position sizing rule, meaning that the portfolio will always be fully invested in the asset selected by the momentum model. In this case, the entire portfolio capital is allocated to the chosen asset regardless of its current level of volatility.
The second version introduces a position sizing mechanism based on a volatility cap. Rather than targeting a fixed volatility level, the rule simply limits portfolio exposure whenever the estimated volatility exceeds a predefined threshold. Specifically, annualized volatility is capped at 20%, which means that position sizes are reduced during periods of elevated market risk but are not levered up when volatility falls below the threshold. This distinction is important because the rule acts as a pure risk-control mechanism rather than a volatility-targeting strategy.
The volatility is estimated using a 63-day lookback period. By adjusting exposure according to recent market fluctuations, the strategy seeks to smooth returns and reduce the impact of extreme moves, especially in Bitcoin. The resulting return profile is intended to resemble the volatility characteristics typically associated with equities, thereby making the strategy more comparable to traditional stock market investments from a risk perspective.
The Results
The sample period is from 2018-2025.

The composite CAGR of 42.36% suggests that, a combination of lookback windows generates substantial annualized returns. Interestingly, the shorter 11-day lookback produces only 6.93% CAGR, while performance accelerates significantly as the window expands. Returns peak at 52.15% CAGR for the 63-day lookback, before easing slightly at the 84-day horizon. This pattern indicates that the strategy benefits from allowing trends more time to develop rather than reacting too quickly to short-term market noise.
Volatility, however, remains relatively stable across all lookback periods. The volatility figures cluster tightly around the 50–52% range, implying that changing the lookback period does not materially alter the strategy’s overall risk exposure. In practical terms, the strategy consistently operates in a high-volatility regime regardless of parameter selection. Such a volatility is way too high for an implementation of such a strategy. Therefore the mentioned volatility cap is needed.

Like the earlier results without the volatility cap, the 11-day lookback period again produces the weakest performance, generating a CAGR of just 6.55%. As the lookback horizon increases, returns improve steadily, with performance peaking at a CAGR of 18.12% for the 63-day configuration before moderating slightly at the 84-day horizon. This pattern reinforces the earlier observation that the strategy benefits from allowing trends sufficient time to develop, rather than reacting too aggressively to short-term market fluctuations and noise.
The strongest overall return is achieved by the composite strategy, which combines signals across multiple lookback periods and delivers a CAGR of 18.51%.
The most notable effect of introducing the volatility cap is the substantial reduction in overall portfolio risk compared to Table 1. Volatility figures now cluster tightly within the 16–17% range across all lookbacks. In practical terms, the volatility cap creates a far more stable and consistent risk profile, allowing the strategy to maintain equity-like levels of risk regardless of parameter selection.
Risk-adjusted performance, measured by the Sharpe ratio, also improves meaningfully as the lookback horizon lengthens. The 11-day strategy records the weakest Sharpe ratio at 0.40, while the 63-day strategy achieves the highest standalone value at 1.06. The composite approach, however, outperforms every individual configuration on both an absolute and risk-adjusted basis. This indicates that combining signals across multiple lookback horizons produces a more balanced and robust trading strategy. Additionally, it has a maximum drawdown of 18% and a Calmar ratio of 1.03.
Further Work
One important extension would be testing the strategy using spot Bitcoin data rather than a Bitcoin ETF. While ETFs provide a practical and accessible investment vehicle, they also introduce structural characteristics such as trading-hour limitations. Spot Bitcoin trades continuously in a 24/7 global market and may therefore exhibit different dynamics.
Another area for future work is the implementation of a volatility-targeted framework. A volatility-targeted approach would dynamically adjust position sizing based on recent market volatility, reducing exposure during periods of heightened uncertainty and increasing exposure when volatility is lower and more stable. The objective of this framework is to create a more consistent level of portfolio risk through time.
Key Takeaways
A 50/50 allocation to Bitcoin and Gold alone is not ideal. Bitcoin offers spectacular returns but comes with devastating drawdowns of nearly 78%, while Gold provides stability at the cost of significantly lower returns. A simple 50/50 blend fails to resolve this tension, still producing a painful 51% drawdown.
Momentum signals can guide smarter allocation. Rather than holding a fixed position, the dual momentum framework dynamically rotates between Bitcoin, Gold, and cash based on both relative performance and absolute performance.
Longer lookback periods outperform shorter ones. The 11-day lookback consistently produces the weakest results, while performance improves meaningfully as the window extends. This suggests trends need time to develop before they become reliable trading signals.
Volatility control is essential, not optional. Without position sizing, strategy volatility clusters around 50–52%, far too high for practical use. Introducing a 20% annualized volatility cap brings risk down to a manageable 16–17% range, comparable to traditional equity investing.
The composite strategy delivers the best risk-adjusted results. Combining 21-, 42-, and 63-day lookback signals produces a CAGR of 18.51% and a Sharpe ratio of 1.11, outperforming every individual lookback configuration on both an absolute and risk-adjusted basis.
Conclusion
Rather than forcing a choice between the explosive but volatile world of cryptocurrency and the steady but modest returns of precious metals, the strategy lets momentum do the decision-making for you.
The results make a strong case for this kind of disciplined rotation. Without any risk management, the strategy generates impressive returns, but it does so at a level of volatility that is far too high for practical use. The introduction of a volatility cap fundamentally changes the character of the strategy. By scaling back exposure during turbulent periods, it transforms what would otherwise be an extremely aggressive approach into something far more practical, delivering equity-like risk with returns that meaningfully exceed what either asset could offer on a standalone basis.
Perhaps the most important insight from this analysis is that the composite approach, which blends signals across multiple lookback windows, outperforms any single lookback.
It is worth emphasizing, however, that past performance in any systematic strategy is not a guarantee of future results. The sample period from 2018 to 2025 captures a unique era in the evolution of both Bitcoin and Gold as investment assets, and conditions may shift in ways that challenge the assumptions underlying this approach.
Ultimately, the dual momentum strategy demonstrates it is possible to participate in the remarkable return potential of Bitcoin while maintaining a level of portfolio discipline that makes the journey far more manageable. For investors looking to engage with these two assets in a thoughtful and systematic way, dual momentum provides a genuinely attractive framework to use or build upon.
Disclaimer
The above article constitutes my or the authors’ personal views and is for entertainment purposes only. It is not to be construed as financial advice in any shape or form. Please do your own research and seek your own advice from a qualified financial advisor. I / The authors may from time to time hold positions in the aforementioned securities consistent with the views and opinions expressed in this article. The information provided in this article is not making promises, or guarantees regarding the accuracy of information supplied, nor that you guarantee for the completeness of the information here. The information in this article is opinion-based and that these opinions do not reflect the ideas, ideologies, or points of view of any organization the authors may be potentially affiliated with. The authors reserve the right to change the content of this blog or the above article. The performance represented is historical and that past performance is not a reliable indicator of future results and investors may not recover the full amount invested.

