What the Eye Cannot See, the Pattern Reveals
A configuration is the trace of relationships we cannot inspect one by one
Previous: Episode 4, The Same Pattern at Every Scale
Next: Episode 6, The Practitioner’s Discipline
An experienced trader sometimes says a market feels different before they can name the change. A clinician notices that the combination of ordinary symptoms does not fit an ordinary case. A forester enters a familiar stand and hears fewer ground birds while seeing new growth along the floor. A pilot registers that several small cues together no longer match the conditions expected on approach.
In the flooded forest of the hero image, bare trees and standing water are visible. A visitor might call it a recent flood. Someone returning over seasons might remember that water once receded sooner and vegetation came back along a particular bank. Neither observer yet knows why the place changed. The returning observer has a history of the place that the photograph lacks.
The grammar of a trend
In a vigorous move, pullbacks may be shallow. Reversals may be absorbed quickly. Volatility may expand mainly in the direction of the trend. Related markets may begin to move together. Later, the pattern can change: pullbacks deepen, an attempted reversal is no longer quickly absorbed, widening daily ranges produce little further progress, and related markets begin to diverge.
No single feature gives a verdict. A deep pullback has preceded both exhausted moves and renewed advances. Correlation can rise because of a common opportunity or a common danger. The configuration is the way these features move together through time, under a set of conditions I can specify and compare. It is a research object before it is a trading claim.
The novice waits for a threshold. The expert notices that the system’s grammar has changed. That observation becomes useful when it is turned into a question: which relationships shifted, when were they visible, and how often did the same arrangement lead nowhere?
The forester who has walked the same route for years can compare how long water remains, what vegetation returns and which animals use the site. The recorded associations, not one dramatic visit, make a change worth investigating.
In my own work, the equivalent observation may be less dramatic. A signal remains valid, but its surrounding markets no longer confirm it as often as they did in the earlier sample. The conclusion is not “skip this one”. It is “define what confirmation means, find all comparable signals and discover whether the difference has mattered”. The discipline begins at the moment the impression can still be written down without knowledge of what followed.
A pattern after the answer is known
The most dangerous expert is the one who describes a chart after the turn. Once the outcome is visible, deepening pullbacks and weakening breadth appear to have announced it. But those features may have appeared during previous pauses that resolved upward. The descriptive claim, “these features were present”, and the predictive claim, “they would have helped me decide then”, are different.
To test the second claim, I have to specify the observations before sorting the outcomes. How deep is a pullback relative to recent movement? Over what interval must related markets diverge? Which markets and periods count? What action would the configuration have justified, and what would the action have cost? Then I count all appearances, including the unmemorable ones. I make room for other explanations and later data.
Kahneman and Klein’s examination of expert intuition identifies two conditions that matter here: an environment with regularities that can be learned and a chance to receive useful feedback about decisions. [1] Fireground research by Klein and colleagues shows how recognition can help experienced commanders act rapidly in a domain they know. [2] Markets offer some learnable patterns, but their feedback is noisy. A sound trade may lose, a poor decision may profit, and the people who create the pattern can change their behaviour. This makes disciplined recognition harder, not worthless.
In the clinic, a practitioner’s concern about an unusual combination can trigger a diagnostic test. In the forest, the observation can direct sampling upstream and downstream. Neither expert needs to conceal the initial judgment; neither should pretend the judgment is already the diagnosis. Trading deserves the same separation between noticing and establishing.
Suppose I write, on a particular date, that breakouts preceded by deeper pullbacks seem less likely to produce large winners. My research note must specify how depth is measured, which prior interval supplies the reference, whether the observation applies to long and short trades, and what result would count against it. I then examine the full set of historical trades, with a later period reserved as a check. If the finding evaporates when I alter one threshold slightly or include execution costs, the original impression may still describe some charts. It has not earned a decision rule.
Inside a rule-based strategy
My live trading rules govern entries, size and exits. A chart that looks tired does not give me a private exemption from a valid signal. If I override it without recording and testing an alternative, I have created another strategy that borrows the old one’s backtest without earning it.
I shift from trusting a single threshold to testing a configuration of relationships. Configuration reading belongs in research and scheduled review. Suppose several positions begin losing together. I examine whether the exposure is consistent with the rules, whether correlations and execution have changed, and whether the same combination appeared in earlier periods. I can formulate a new filter or a different sizing rule, test it across the complete record and document its effect on the large winners as well as the failed entries. If the new rule earns adoption, it becomes part of the strategy by an explicit decision. The discipline preserves judgment by giving it a place where it can be evaluated.
This placement answers a misunderstanding about systematic practice. Rules do not eliminate the practitioner. Someone must decide what deserves research, maintain the data, judge whether a contract remains tradable, and examine whether implementation matches the design. Human attention can improve that work without becoming an unaccountable trade-by-trade veto.
I learned a powerful version of relational attention from someone who had never seen my price charts. In Episode 6 I will tell the story of an Aboriginal Elder, a flowering tree and the location of fish downstream. That encounter changed my understanding of what a practitioner can know by attending to associations over time.
What a trained eye actually does
The forester and trader face a shared challenge in deciding when a change is meaningful. A familiar forest can look different after one storm and return to its earlier pattern the next season. A market can show deeper pullbacks for several weeks and then resume its former trend. The trained eye is attentive to the changed association, but it does not declare a new state from a single dramatic visit. It knows what a comparable observation would look like and which changes ought to persist if the proposed explanation is right.
That is a more demanding skill than collecting an inventory. If water remains unusually long, the forester might check upstream drainage, compare nearby sites and look for vegetation that responds to prolonged saturation. If a market’s breakouts repeatedly fail while related instruments diverge, I might examine whether the change is broad, whether execution has deteriorated and whether similar intervals occurred before. The work turns a scene into a set of discriminating questions. It keeps the first impression alive as a hypothesis while making it possible to be wrong.
That skill is not magic and need not be concealed behind “feel”. I can keep a dated note with the observed configuration, the plausible interpretations, the evidence that would favour one over another, and the time at which I will revisit it. Over years, the notes reveal which impressions were useful and which were compelling only in memory. The same method also teaches humility about changing environments. A once reliable configuration may stop having the consequence it had before.
In Bohm’s vocabulary, the explicate observations give us partial access to an implicate structure of relationships and constraints. The label is not the explanation. The work is in tracing how the observed associations arise in this setting and whether they remain informative when tested. When a market changes, a single price may not tell me what has altered. The relationship among price, response and participation may make the change intelligible.
When an interpretation proves wrong, the record should teach me something more exact than “the market fooled me”. Perhaps I defined related markets too broadly, overlooked seasonality, or selected only the episodes that stood out visually. Each failed test narrows the claim. The reward for noticing a pattern is not the right to trade on it. It is a chance to learn what the pattern actually describes.
The forester returns to learn whether water still stands where it once receded. I return to the complete set of breakouts, including the ones that contradict my impression. A changed configuration can point to a hidden relationship. The dated record tells me whether I recognised it in time to act and whether it earned a place in the rules.
The water in the photograph will recede or remain, and a return visit will begin to answer which. Episode 6 asks how a practitioner builds the memory needed to make such a comparison.
Previous: Episode 4, The Same Pattern at Every Scale
Next: Episode 6, The Practitioner’s Discipline
Sources and notes
[1] Daniel Kahneman and Gary Klein, Conditions for Intuitive Expertise: A Failure to Disagree, American Psychologist 64, no. 6 (2009): 515–526.
[2] Gary Klein, Roberta Calderwood and Anne Clinton-Cirocco, Rapid Decision Making on the Fire Ground: The Original Study Plus a Postscript, Journal of Cognitive Engineering and Decision Making 4, no. 3 (2010): 186–209.
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, Complex Adaptive Markets, Carved by Impossibility and The Aussie Turtles Trend Following Guide.
Want to explore why structure exists at all?
Carved by Impossibility: What Remains When Everything Else Is Eliminated
The book explores the architecture of constraint, emergence, and reality itself, and what it means for how we understand markets, life, and the universe.
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.