Eight articles. Thirty-eight years of science. One verdict.
Over the past two weeks, we published an eight-part series that asked a question the mainstream of finance has been unable to answer for decades: why do trends exist?
We did not ask this question abstractly. We traced it to its source. In 1987, a small group of physicists and economists gathered at the Santa Fe Institute and dropped the foundational assumption of modern economics. They abandoned equilibrium. What they built in its place, a framework called complexity economics, turns out to be the deepest theoretical validation that diversified systematic trend following has ever received.
This series tells the story of how that happened, from the desert workshop that started it all to the central banks and regulators now adopting its tools, and draws a direct line from the science to the practice of following trends.
Why We Published This Series
The thesis behind this series connects directly to the argument made in The Fractals of Finance: that markets are not random, that feedback drives their behaviour, and that the geometry of price reveals a living system rather than a mechanical one. The earlier Fractals of Finance Research Series presented the empirical evidence for this claim. This series presents the theoretical foundation.
Complexity economics, pioneered by W. Brian Arthur at the Santa Fe Institute, provides the scientific framework that explains why markets behave the way the data shows they do. Agent-based models demonstrate that fat tails, clustered volatility, momentum, and speculative bubbles are not anomalies. They are the structural output of any market populated by adaptive agents. This is the science behind the statistics.
We structured the series as a narrative because that is what the history actually is. A revolution in economic thought, fought across journal rejections, interdisciplinary workshops, artificial markets built from scratch, and a slow institutional awakening that is still unfolding. Each article builds on the last. The argument is cumulative. By the eighth article, the case is not merely suggestive. It is complete.
The Argument, Article by Article
Article 1: Do Anything You Like, Providing It Is Not Conventional
The series opens in 1987 with Brian Arthur’s phone call to Kenneth Arrow and the founding workshop at the Santa Fe Institute. The instruction from Citibank’s chairman: rethink economics from scratch, just not the way it has been done before. The first decision the group made was the most radical: they dropped equilibrium. The article introduces the core framework of complexity economics, previews Arthur’s El Farol problem and increasing returns, and establishes the central distinction of the series: predictive strategies compete inside the market’s ecology and are subject to crowding and collapse; responsive strategies sit outside and harvest the dynamics that prediction failure produces.
Article 2: The Bar Problem That Explains Every Market
Arthur’s El Farol bar problem, unpacked in full. In a reflexive system, any prediction that becomes sufficiently widespread reshapes reality into something that contradicts the prediction. The article formalises the predictive versus responsive distinction and demonstrates it through two real-world case studies: the GameStop short squeeze of January 2021 and the Volmageddon volatility collapse of February 2018. Both are El Farol dynamics in action: ecologies of homogeneous predictions that crowded, collapsed, and released energy into trends. The conclusion: trend following’s edge does not decay because it is not a prediction. It is a structural response to the permanent cycle of prediction failure in reflexive markets.
Article 3: When Small Things Become Unstoppable
The story of Arthur’s six-year fight to publish his theory of increasing returns, rejected by journal after journal because the idea “cannot be an equilibrium.” The article explains how positive feedback creates lock-in in technology markets (VHS over Betamax, Windows over Macintosh) and translates the dynamic directly into finance: a trend is increasing returns operating in the domain of price. The Swiss National Bank’s abandonment of the franc-euro floor in January 2015 is examined as a case of lock-in breaking. The rise of passive investing is identified as a lock-in currently forming. In an increasing-returns world, the responsive strategy that follows feedback is systematically aligned with the most powerful force in the market.
Article 4: All Systems Will Be Gamed
Arthur’s principle that any rule-based system will be exploited by the adaptive agents within it, applied to the 2008 financial crisis. The article details Arthur’s four categories of exploitation and shows how each played out in the mortgage securitisation chain. The crisis is reframed as a complexity event: an ecology of homogeneous predictions about housing safety that crowded and collapsed, producing cascading failure that no equilibrium model could see coming. If all boundaries will break and you cannot know which ones, the only durable architecture is one that is permanently positioned to respond when they do.
Article 5: The Physicist Who Beat the House Twice
The story of J. Doyne Farmer: from building a shoe-mounted computer to beat roulette with the Eudaemons, to co-founding the Prediction Company in Santa Fe, to developing a formal theory of market ecology at the Santa Fe Institute and Oxford. The article traces the divergence between Edward Thorp’s path (finding specific mispricings through prediction) and Farmer’s path (understanding why exploitable structure exists through dynamics). Farmer’s market ecology framework confirms, from a completely independent line of research, that momentum, fat tails, and volatility clustering are structural features of any complex adaptive system.
Article 6: The Artificial Stock Market
The landmark Santa Fe Artificial Stock Market experiment. Arthur, Holland, LeBaron, Palmer, and Tayler built a simulated stock exchange populated by adaptive agents with simple learning rules. When the agents actively explored new strategies, the market spontaneously produced fat tails, clustered volatility, speculative bubbles, sudden crashes, and exploitable momentum. When the agents were passive, the market was efficient and technical trading did not work. The efficient market is a dead ecology. The complex market is the living one, and it is the only one that has ever existed. The returns that trend followers capture come from the ecology itself.
Article 7: The Economy on a Computer
The scaling of complexity economics from a single artificial stock to entire economies. The article documents the failure of DSGE models in 2008, Andrew Haldane’s public break from equilibrium thinking at the Bank of England, and the institutional adoption of agent-based models by central banks and regulators worldwide. It covers Farmer’s Complexity Economics programme at Oxford and the publication of The Economy as an Evolving Complex System IV by the Santa Fe Institute Press. The institutions that underwrote the theoretical framework which said trend following should not work are now adopting the framework that explains why it does.
Article 8: What Complexity Economics Means for Your Trading
The finale distils the entire series into eight actionable principles: the market is an ecology, not a machine; predictions are food for responsive strategies; trends are increasing returns in the domain of price; all boundaries will break; fat tails are the system’s normal output; simple rules outperform in complex environments; diversification is the structural response to an ecology; and the programme is the strategy. Each principle is grounded in the science developed across the preceding seven articles and translated directly into the practice of portfolio construction and risk management.
The Core Finding
Financial markets are not equilibrium systems that occasionally experience shocks. They are complex adaptive systems that naturally produce trends, fat tails, clustered volatility, and speculative bubbles.
Agents inside the market form predictions. Those predictions crowd. The crowding generates instability. The instability produces directional moves that persist long enough for a responsive strategy to capture. This cycle of prediction, crowding, failure, and release is not a market inefficiency that can be arbitraged away. It is the market itself. It is what reflexivity produces, permanently, as a condition of its own existence.
The implications cascade across every domain of finance.
For risk managers: fat tails are not rare events caused by external shocks. They are the system’s normal output whenever a large cluster of predictions fails simultaneously. Models built on Gaussian assumptions do not underestimate risk. They render it invisible.
For investors: the efficient market hypothesis rests on equilibrium. The institutions that regulate global finance are abandoning that premise. If markets are not in equilibrium, prices do not reflect all available information. Trends can exist. Momentum can exist. The empirical features that trend followers have been exploiting for decades are not anomalies. They are structural.
For trend followers: the edge does not decay because the mechanism that produces it is permanent. As long as markets are populated by adaptive agents who form predictions, those predictions will crowd and fail. As long as predictions crowd and fail, directional moves will follow. As long as directional moves follow, a diversified responsive programme will capture them. The science is unanimous.
For the profession: the equilibrium framework that has dominated economics for two centuries is being supplemented, and in critical areas replaced, by complexity-based tools. Agent-based models are now used by central banks, financial regulators, and policymakers. The theoretical framework that said trend following should not work is being replaced by one that explains why it must.
The Books
This series presented the theoretical foundation. The books present the complete argument and its practical applications.
The Fractals of Finance: Determinism, Adaptation and the Geometry of Markets tells the full story of feedback, fractal geometry, and market behaviour, and provides the empirical evidence that the architecture of price is neither random nor efficient. Available now on Amazon.
Three forthcoming books in 2026 extend the argument into the territory this series has opened:
Complex Adaptive Markets describes the living systems inside markets: the feedback loops, the cascading dynamics, the ecology of competing strategies that complexity economics reveals.
Carved by Impossibility argues that robust portfolio structure is not designed. It is what remains after everything fragile has been eliminated.
Trend Following Manifesto (with Niels Kaastrup-Larsen) describes the practical architecture: the diversified systematic programme built to harvest the dynamics that the science says are permanent.
There is no equilibrium. There is only adaptation.
Follow the trend.
Want the theoretical foundation for why trend following works?
The Fractals of Finance: Determinism, Adaptation and the Geometry of Markets
The book explores the full architecture of feedback, fat tails, and fractal structure in financial markets, and what it means for how we trade, invest, and understand risk.
Available now on Amazon in paperback, hardcover, and Kindle.
Want a practical field manual for trading trends and capturing outliers?
The Aussie Turtles Trend Following Guide: A Field Manual for Hunting Outliers adapts the timeless principles of the original Turtle traders into a systematic, rules-based approach for modern markets. Co-authored with Adam Havryliv.
Available now on Amazon in paperback, hardcover, and Kindle.