The Vault

THE EMBEDDED AGENT SERIES | Episode 7 of 7: The Practice of Embedding

What It Means to Build for a World That Includes You

“In the beginner’s mind there are many possibilities, but in the expert’s mind there are few.” - Shunryu Suzuki

This is not a conclusion. Conclusions belong to arguments that resolve. The series you have been reading does not resolve, because the condition it describes does not resolve. You are embedded. You will remain embedded. There is no step you can take that places you outside the system, no insight that grants you the view from above, no strategy so clever that it transcends its own participation in the ecology it exploits.

So this final essay is not an ending. It is a pivot. From understanding to practice, from seeing the condition to living within it. What embeddedness means, you already know; the question now is what it requires you to do.

Because embeddedness is not, in the end, a theoretical problem. It is a practical one. It changes what you build, how you size, what you expect, how you respond, and who you become as a participant in a system that includes your participation as part of its ongoing construction. The question is no longer: do you understand? The question is: given that you understand, how do you act?

The Photograph and the Walk

You have built a model, backtested it, optimised it, stress-tested it. And now you know something that changes the meaning of all that work: your model describes a world that included your absence.

This does not make the model worthless. It makes the model provisional. The backtest is a photograph of a room you were not in. It tells you about the room’s general character: its dimensions, its lighting, its furniture. These things are real information. But the photograph cannot tell you what the room will look like when you enter it, because your entry changes the room.

The practice of embedding begins here: with the way you hold your models. Not as predictions. Not as promises. As orientations. The model tells you what kind of thing you are building. It reveals your strategy’s character, its structural tendencies, its relationship to certain market conditions. You use this information the way a sailor uses a weather chart: not to know what the ocean will do, but to understand what your vessel is built for.

Holding models lightly is not the same as holding them cynically. The cynical trader dismisses models entirely, which is just another form of certainty: certainty that models are useless. The embedded agent holds models with the specific weight they deserve: informative, incomplete, and provisional in equal measure. They are the best tools available, and they are structurally incapable of capturing the world as you will find it. Both things are true, simultaneously, permanently.

Sizing as the Primary Practice

If there is a single discipline that separates the embedded agent from every other kind of market participant, it is sizing.

Sizing is not risk management in the conventional sense. Conventional risk management asks: given the likely distribution of outcomes, how much should I risk? The embedded agent asks a different question: given that the distribution of outcomes includes my own participation, and the participation of everyone who trades like me, and the reflexive feedback of our collective actions, and the emergent dynamics of a computationally irreducible system, how much can I afford to be wrong by and still remain in the game?

This is not a quantitative question that yields a precise answer. It is a qualitative question that yields a practice. The practice is: size smaller than your model suggests. Not because the model is wrong, but because the model describes a world that does not include you. The world that includes you will produce outcomes the model did not contemplate, and those outcomes will come at the worst possible time, because the system’s capacity for surprise is greatest precisely when the most participants believe they have accounted for it.

Sizing is the embedded agent’s primary expression of humility. It is the practical manifestation of everything this series has explored. You know the map is drawn of a territory that shifts when you walk on it. You know your strategy changes the ecology it feeds upon. You know the system computes at a level you cannot access. By now you also know that your beliefs become structural forces, that the boundary between you and the market is porous, and that understanding does not grant prediction.

Sizing is what you do about all of that. You size for the world as it is, not as your model says it is, and for surprises you cannot specify. You size, above all, for the recursive loop that amplifies your losses when everyone who trades like you loses at once. And you size for survival, because survival is the only thing that compounds.

The Discipline of Provisional Commitment

The embedded agent’s relationship to any position, any trade, any belief about the market, is one of provisional commitment. This is harder than it sounds.

Provisional commitment means acting with full conviction on the basis of a view you know to be incomplete. It means entering a trade as though the signal is right, even while knowing the signal describes a world that has already shifted. You follow your rules with discipline, and you hold those rules with the lightness that comes from understanding their structural limitations.

Most traders default to one extreme or the other. The rigid trader follows the rules mechanically, without adaptation, and is eventually destroyed by a regime the rules were not designed for. The flexible trader adapts constantly, abandoning rules at the first sign of discomfort, and is destroyed by the absence of any consistent framework. Neither extreme works, because both misunderstand the nature of the game.

The embedded agent practises something more difficult than either rigidity or flexibility: committing fully while remaining genuinely adaptable. The commitment is to the framework: the rules, the strategy, the systematic approach. The adaptability is in the recognition that the framework is operating within a system that is changing it, and that the framework itself may need to evolve. The discipline is in knowing which to do when: when to hold the line, and when to adjust it.

This sounds like judgment, and it is. But it is judgment informed by a deep understanding of why your framework is both necessary and insufficient. It is not guesswork. It is the practiced, embodied, continuously refined capacity to navigate a system that includes your navigation as part of its dynamics.

Consistency Across Conditions

In a system that is computationally irreducible, you cannot know which regime is coming next, whether the year ahead will be kind or cruel to your strategy, whether the niche you occupy will expand or contract. And because you cannot know these things, the most dangerous temptation is to chase the strategy that would have been right for the regime that just ended.

The embedded agent’s edge is consistency. Not the consistency of doing the same thing regardless of conditions, but the consistency of maintaining the same principles, the same risk framework, the same structural alignment, across conditions you did not foresee.

This consistency is what allows such a trader to capture the returns that come from being present when the system produces outcomes aligned with the strategy. If you are a trend follower, your edge comes from being in the market when trends develop. You do not know when trends will develop. You cannot predict the next trending period. What you can do is ensure that when the trend arrives, you are there to capture it. And the only way to ensure that is to be there always, through the drawdowns, through the flat periods, through the regimes that feel hostile to your approach.

Consistency is the temporal expression of embeddedness. You are not embedded only when conditions are favourable. You are embedded all the time. Your presence in hostile conditions is what earns your presence in favourable ones. Absorbing the drawdown is what positions you to capture the recovery, and your patience through the quiet periods is what lets you participate in the volatile ones.

Most traders fail not because their strategy is wrong but because they are absent when their strategy is right. They abandon the approach during the drawdown, switch to something that worked recently, and miss the turn. The embedded agent stays. Not blindly, not passively, but with the active, disciplined presence that comes from understanding that you cannot time your embedding. You can only sustain it.

The Ontological Shift

There is a deeper shift here, one that goes beyond tactics and strategy and into the way you understand your own relationship to the act of trading.

Most market participants experience uncertainty as a problem to be solved. The goal is to reduce uncertainty, to improve predictions, to narrow the range of possible outcomes until the future becomes, if not certain, at least manageable. Uncertainty is the enemy. Knowledge is the weapon. The aim is to know more, predict better, control the outcome more tightly.

The embedded agent experiences uncertainty differently. Not as a problem to be solved but as a condition to be inhabited. The system is uncertain because it is complex, adaptive, and irreducible. This uncertainty is not a bug. It is the source of the very dynamics that create opportunity. Without uncertainty, there would be no trends to follow, no mispricings to exploit, no regimes to adapt to. Uncertainty is the medium in which the embedded agent operates, and eliminating it would eliminate the activity itself.

This is not merely an epistemological shift, accepting that you cannot know the future. It is an ontological shift, accepting that you are part of a system whose nature is uncertainty, and that your practice must be built for that nature rather than against it.

The difference sounds abstract, but its practical consequences are enormous. The epistemological shift says: I cannot predict, so I will hedge my bets. The ontological shift says: I am inside a system that is inherently unpredictable, so I will build a practice that thrives on unpredictability rather than merely tolerating it. The first is defensive. The second is generative.

Trend following, at its best, is a generative response to uncertainty. It does not merely survive unpredictability. It feeds on it. Every major trend emerges from the system’s inherent uncertainty. Every significant move is a surprise to most participants. The trend follower’s practice is designed to capture precisely these surprises, not by predicting them but by being structurally positioned to benefit when they occur.

The Sculptor and the Clay

There is a metaphor that captures what it means to practise as an embedded agent, and it is not the metaphor of the navigator, though that comes close. It is the metaphor of the sculptor who is also the clay.

A sculptor working with external material has the luxury of standing apart. The sculptor shapes the clay. The clay does not shape the sculptor. The relationship is one-directional. The artist has a vision, and the material submits to it.

But what if the sculptor were made of clay? What if the act of shaping the material also shaped the shaper? What if every press of the thumb altered not just the form being created but the hand creating it? The sculpture would be a mutual production: partly the sculptor’s intention, partly the clay’s response, partly the ongoing feedback between the two. And the sculptor could never step back to see the finished work, because stepping back would change both the sculpture and the sculptor.

This is the embedded agent’s condition. You are shaping the market with your participation, and the market is shaping you. Your strategies, your convictions, your risk tolerances, your emotional responses are all being moulded by the system you are trading. The drawdowns change you, and so do the wins. The long stretches of nothing change you too. You are not the same trader you were before the last crisis, and you will not be the same after the next one.

The practice of embedding is the practice of sculpting and being sculpted simultaneously. It requires attending not just to the market but to yourself within the market. How is the current regime changing your perception? How are your recent results affecting your confidence? How is your experience of the last drawdown shaping your sizing in the next opportunity? These are not psychological questions in the self-help sense. They are structural questions about the feedback between agent and environment.

The Practice That Never Resolves

There is no point at which the embedded agent “arrives,” no level of understanding at which the condition is finally transcended. No strategy is so well-designed that it accounts for every recursive loop, every ecological shift, every emergent dynamic. The practice does not resolve. It deepens.

This is, paradoxically, what makes it sustainable. A practice that aimed for resolution would eventually be defeated by the system’s inexhaustible capacity for novelty. A practice that aims for depth can accommodate novelty because it is not trying to predict the novel event but to respond to it from a foundation of structural understanding.

The embedded agent who has practised for twenty years has not solved the market. Such a trader has developed a richer, more textured, more embodied understanding of what it means to participate in a complex adaptive system. Each year adds not certainty but nuance. Each regime adds not mastery but calibration. And each surprise leaves you not more in control, but more responsive to whatever comes next.

This is the trajectory of any serious practice. The martial artist does not eventually defeat all opponents. The musician does not eventually play all pieces perfectly. The meditator does not eventually stop thinking. Each practitioner deepens their engagement with the discipline, and the deepening is the point. The product of practice is not a result. It is a more refined capacity to practise.

The System Includes You

Let us return to where we began, six essays ago. You are not observing the market. You are inside it. Your models describe a world that included your absence. Your strategy is a species in the ecology it exploits. Your beliefs are building materials. The boundary between you and the market is a useful fiction. Understanding will not yield prediction.

And yet, here you are. Still trading. Still participating. Still embedded.

This is not defeat. It is the most honest starting point available. You are inside a system that includes your presence. Your models are provisional. Your edge is relational. Your risk is co-created. Your understanding is structural, not predictive. And the practice you are engaged in is not a problem to be solved but a condition to be inhabited with increasing skill, increasing patience, and increasing respect for the system that includes you.

The embedded agent does not seek to escape embeddedness. There is nowhere to escape to. The task is to practise within it instead. To build systems that account for their own presence. To size for a world that includes their participation. To hold models with the lightness they deserve. To maintain consistency across conditions they could not foresee. To accept uncertainty not as a problem but as the medium of the practice.

The system includes you. Your practice must include that fact.

A Beginning

There is a tradition in Japanese calligraphy where the master, after years of practice, is asked to produce a single stroke. Not a character, not a word, not a composition. A single stroke. And in that stroke, the master’s entire practice is visible: the years of discipline, the thousands of pages of drills, the patient deepening of a relationship with brush, ink, and paper. The stroke is simple. The practice behind it is not.

The embedded agent’s equivalent of the master’s stroke is the next trade. Not a remarkable trade. Not a brilliant trade. Just the next one. Sized correctly. Entered with discipline. Managed with the structural understanding of everything this series has explored. Held lightly. Released when the system says to release it. And then the next one. And the next.

Each trade is unremarkable. The practice that produces each trade is a lifetime’s work.

This series has not given you a system to trade. It has not given you an edge to exploit. It has not given you a prediction to act upon. What it has given you, if it has succeeded, is a different relationship to the act of participating in markets. A relationship grounded in the structural reality of embeddedness rather than the comfortable fiction of detachment.

You are inside the system. The system includes you. And the only practice that honours this truth is one that is never finished, never resolved, and never far from the simple, irreducible act of placing the next trade with the full weight of understanding behind it and the full acceptance that understanding is not enough.

It was never going to be enough. And that is precisely where the practice begins.

Richard Brennan writes on systematic trading, complex adaptive markets, and the philosophical foundations of trend following at atstradingsolutions.com. His books include The Fractals of Finance and Complex Adaptive Markets. The forthcoming Carved by Impossibility completes the trilogy.

Want the theoretical foundation for why markets adapt?

Complex Adaptive Markets: How Living Systems Shape Finance

The book explores the full architecture of feedback, emergence, and adaptive behaviour 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 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.

Share this post:

Facebook
LinkedIn
X