Skip to content

Artemis: From Deflation to Reflation (Dragon Portfolio): summary and key takeaways

The article reviews how a hypothetical Dragon Portfolio, built from equities, long bonds, gold, long volatility and trend, behaved in 2020. It reports a +52% gross gain with a -11% maximum decline, while 60/40 and risk parity portfolios lagged. The results are hypothetical and come from the firm that created the portfolio.

Cole, C. · 2021 · Artemis Capital Management, 2021 Letter to InvestorsSummary updated 20 November 2026Practitioner research
Published2021Firm research
Asset focusMulti-assetStrategic allocation
TypePortfolio MgmtInflation & Macro
Reading time4 minthis summary

Artemis: From Deflation to Reflation (Dragon Portfolio) at a glance

The question
Does a portfolio that diversifies across market regimes rather than asset classes hold up through the deflation, fiat-debasement and reflation phases of 2020?
Dataset & period
Hypothetical Dragon Portfolio assembled from daily series of actual funds and instruments after estimated fees and trading costs, scaled to 15% target volatility. Forward test covers 2020. Long-run comparison charts cover 1928 to 2020 (93 years). Exact sample counts: Not stated in the article.
Method
The firm defines the Dragon as roughly equal exposure to five regime classes: equities, interest-rate-linked bonds, fiat alternatives such as gold, long volatility and trend. For 2020 it assembled an investable version from proprietary long-volatility fund returns, CTA fund-of-funds returns, Treasury bond futures, equity futures and ETFs, and physical gold with custody fees. Cash is pooled at one prime broker with portfolio margin to reach the volatility target. Results are compared with 60/40 and risk parity portfolios.

Key findings of Artemis: From Deflation to Reflation (Dragon Portfolio)

  • For 2020 the hypothetical Dragon returned +52% before fees at the portfolio level, with a worst decline of -11%. The article says 60/40 and risk parity did far worse.
  • In January to March 2020 the Dragon gained +13% while 60/40 lost -9% and risk parity lost -21%. Long volatility drove the gain.
  • From April to August the Dragon gained +32% against +21% for 60/40, helped by equities and gold during the stimulus-driven rebound.
  • From September to December the Dragon gained +2% against +5% for 60/40. Trend strategies and equities gained as commodity trends emerged, while gold suffered.
  • The article states that 91% of 60/40 price appreciation over 93 years came from 22 years between 1984 and 2007, which it uses to argue against recency bias.
  • A do-it-yourself version using ETPs is described as hampered by idle cash, because one volatility ETP uses only ~2% of net asset value to run its strategy.

Robustness, as reported in the paper

  • The 2020 series draws on after-fee data from actual funds and instruments, with estimated trading costs, rather than a pure backtest.
  • The long history (1928 to 2020) is a hypothetical replication using simplified systematic proxies, and the firm says it was built with hindsight.

Caveats and limits

  • All Dragon results are hypothetical and prepared by the firm that sells related strategies; the disclaimer says they are not live returns.
  • A single year is a very small sample for a regime-diversification claim (inferred).
  • The 2020 long-volatility and trend inputs are the firm's own fund and partner CTA returns, so the headline depends on those managers (inferred).
  • The return is quoted gross, and the leverage and financing needed to reach 15% volatility are not itemised in the article.

Items marked (inferred) are our reading of the design, not statements by the authors.

Why it matters

The note frames stock-bond reliance as a bet on the post-1982 regime and offers a regime-based allocation as the alternative. Practitioners can use it as a description of one regime-diversified design and its operational issues, such as cash efficiency and rebalancing. It is a firm's own case study, so the evidence is illustrative.

This is a firm research note or web article, not a peer-reviewed paper.

Questions about this paper

How did the Dragon Portfolio do in the first quarter of 2020?

The article reports +13% from January to March, against -9% for 60/40 and -21% for risk parity.

Why does the article say a DIY Dragon is hard?

It cites cash inefficiency of volatility and trend products, such as ~2% of net asset value used by one volatility ETP, plus sourcing, manager diversification and rebalancing discipline.

Are the Dragon returns real trading results?

No. The article states that Dragon results are hypothetical and not returns from any live account.

Where can I read the original paper?

The original is at its publisher: https://www.artemiscm.com/. We link to the paper and do not host the PDF.

Follow new quant research

Download MacroFinance market data as CSV or Excel, or pull it by API, and test the ideas in these papers yourself. New summaries are added as we review them.

Create a free account
Summaries are written by MacroFinance from the published papers, in our own words, for information only. They are not investment advice and may omit detail. Papers are not hosted here: each page links to the original at its publisher.