Stock Market
Trace Evidence

The Market Leaves Residue

In 1910, the French criminalist Edmond Locard proposed a deceptively simple law of physical contact: every meeting of two surfaces leaves a trace on each. A century later, Locard's exchange principle remains the philosophical foundation of forensic science. This paper argues that the same principle governs financial market behavior at the microstructure level.

Institutional accumulation and distribution cannot occur without leaving observable residue — in volume, in float dynamics, in repeated alert convergence, in the shape of price relative to its anchors. These traces are not predictions. They are not advice. They are forensic findings about events that have already occurred or are presently unfolding in the order book — observable, measurable, time-stamped, and amenable to chain-of-custody preservation.

Locard's Exchange Principle

Edmond Locard was born in 1877 in the village of Saint-Chamond, in the Loire region of France. By 1910 he had founded the world's first police forensic laboratory, a two-room facility in the attic of the Lyon courthouse, equipped with little more than a microscope, a spectroscope, and a folding camera. His resources were primitive. His thesis was extraordinary.

The corollary to Locard's principle, made famous by his student Paul Kirk and refined by generations of criminalists, is what we now call the exchange principle: when any two objects come into contact, each transfers material to the other. The criminal carries something away from the scene and leaves something behind. The contact itself is the proof.

Locard's first famous case demonstrated the principle. In 1912, a young woman named Marie Latelle was found strangled in her parents' home in Lyon. Her boyfriend, Émile Gourbin, had a seemingly airtight alibi — he had been playing cards with friends thirty kilometers away, and his friends would swear to it. Locard was not interested in the alibi. He examined Gourbin's fingernails under his microscope and found small flakes of skin and a residue of pink powder. The powder turned out to be a custom-blended cosmetic — a cream of bismuth, magnesium stearate, and a particular shade of rose pigment that matched, exactly, the makeup Marie Latelle had purchased from a particular pharmacist in Lyon two weeks earlier. Confronted with the analysis, Gourbin confessed.

The case is famous not because the science was complex — by modern standards, the analysis was crude — but because of what Locard had proven about the structure of evidence itself. Gourbin's alibi was constructed. His witnesses were paid. The card game was real but the timing was a lie. None of these reconstructed details survived contact with the trace evidence, because the trace evidence was not the product of any actor's intention. It was a physical consequence of contact that no amount of preparation could have anticipated or eliminated.

This is the property that makes trace evidence epistemologically distinctive. A confession can be coerced. A witness can be mistaken. A document can be forged. A weapon can be planted. But the residue of contact, when it can be observed and measured, is involuntary. It records what happened, not what anyone wishes to claim happened.

The forensic discipline that emerged from Locard's work is, at its core, a structured method for reading evidence that no participant intended to leave. The questions a criminalist asks are not metaphysical. They are mechanical. What touched what? In what direction? With what force? Over what duration? These are the questions trace evidence is built to answer. They are the same questions, this paper argues, that should be asked of the order book.

The Anatomy of Trace Evidence

Trace evidence behaves according to characteristic properties that distinguish it from other categories of forensic information. Understanding these properties is a prerequisite to recognizing the same patterns in domains other than physical crime scenes.

These five properties — involuntary, directional, time-decaying, convergence-dependent, and expert-mediated — together describe what makes trace evidence powerful and what bounds its use. They also describe, this paper now argues, the structure of a particular kind of information available in financial markets.

From Crime Scene to Order Book

Markets are not crime scenes. The participants are not, with rare exceptions, criminals. But markets are venues of contact, in the strict Locardian sense. Every transaction is a meeting of two surfaces — a buyer and a seller — and that meeting deposits residue on both sides. The residue accumulates in the public tape, in the order book, in the option chain, in the volume profile, and in the time stamps that record when each contact occurred.

The professional participants in modern markets — institutional desks, market makers, algorithmic execution engines, hedged liquidity providers — operate under the same constraint that bound Émile Gourbin in his card game in Lyon. They cannot conduct their business without leaving residue. The residue is not the trade itself, which may be private, but the structural footprint of the trade in the public record: the volume that printed at a particular price, the breadth of the book that absorbed the order, the rate at which inventory was rebuilt afterward. Each of these is a trace. None is the institution's intention to communicate. All are involuntary consequences of size and speed meeting public infrastructure.

The question for any forensic discipline applied to markets is whether these residues can be measured, classified, and interpreted with the same care that a criminalist applies to physical trace material. The thesis of this paper, and of the forensic methodology underlying the platform that publishes it, is that they can — provided the discipline is held to the same epistemological standards. That means observable phenomena rather than inferences about intent. Documented methodology rather than proprietary black boxes. Time-stamped findings preserved with chain of custody. Convergence requirements rather than single-trait dispositive claims. And — crucially — explicit refusal to cross the line from forensic finding to predictive advice.

Market Trace Evidence — Seven Categories

The traces left by institutional activity in public market data fall into several categories, each loosely analogous to a physical-forensics category and each carrying its own evidentiary properties.

The Convergence Requirement

The single most important methodological principle inherited from physical forensics is that trace evidence is rarely dispositive in isolation. A single brown wool fiber proves nothing. A single high-volume bar on a single chart on a single day proves nothing. A single insider purchase, a single FDA filing, a single bullish news headline, a single oversized option print — none of these, taken alone, supports a forensic conclusion.

The discipline of forensic analysis is in part a discipline of refusing to draw conclusions from individual data points. It requires the analyst to wait for convergence: multiple independent signatures, captured by different instruments, describing the same underlying event from different angles, before any conclusion is offered. In physical forensics this convergence requirement is embedded in evidentiary standards — no responsible analyst will take a single fiber to court as the sole basis for a charge. The same standard, this paper argues, should govern any forensic methodology applied to market data.

The methodology underlying the platform that publishes this paper enforces convergence as a prerequisite for elevating any signal to public visibility. A momentum reading alone is not enough. A news catalyst alone is not enough. A volume anomaly alone is not enough. The standard for surfacing a forensic finding is the alignment of independent signatures — momentum, volume, news, regulatory context, insider activity, market regime — across multiple domains, each computed by separate machinery, each carrying its own confidence score. Only when several of these align, on the same ticker, within a tight time window, does the methodology classify the event as a finding worth publishing.

This Is Not Forecasting

A persistent confusion in the discussion of market analysis is the conflation of forensic finding with predictive forecasting. The two are categorically different activities, and the distinction matters legally as well as methodologically.

A forecast is a statement about the future. It says: this stock will rise, that index will fall, this sector will outperform. Forecasts are by their nature unfalsifiable until the future arrives, and even then they are evaluated only in aggregate, because any individual forecast can be defended as bad luck. Forecasting is the dominant idiom of financial advice, of investment management, and of the regulatory framework that governs personalized financial advisement under the Investment Advisers Act of 1940.

A forensic finding is a statement about the present and the recent past. It says: these traces, observed in this time window, exhibit this convergence pattern. The finding is verifiable at the moment of publication — the underlying data is timestamped and preserved. It does not assert what will happen next. It does not recommend an action. It does not address any individual reader's circumstances. It documents an observable state of the market and discloses the methodology by which that state was characterized.

• No entry prices, exit prices, position sizes, or stop-loss levels are specified for any individual user.
• No signal is described as a recommendation — every output is a forensic finding about observable market state.
• No suitability judgment is offered to any reader's individual circumstances.
• The publisher exclusion of the Investment Advisers Act §202(a)(11)(D) and the disclosure framework of CFTC Rule 4.41 govern the regulatory posture, and that posture is maintained at every customer-facing surface.

Locard, presenting his fiber analysis of Émile Gourbin's fingernails to the magistrate in 1912, did not predict that Gourbin would confess. He documented what the residue showed. The confession was not Locard's product. It was the consequence of Gourbin being confronted with evidence that no constructed alibi could explain. The forensic posture takes the same form: present the evidence, disclose the methodology, document the convergence, and let the reader form their own conclusions about what to do with the finding.

The Forensic Posture

The forensic posture in market analysis is not a metaphor. It is a transferable discipline, with its own epistemology, its own evidentiary categories, and its own methodological commitments. Locard's exchange principle is the foundational law of physical contact, and it has analogues in any domain where contact between participants leaves measurable residue. Markets are such a domain.

The volume on the tape, the asymmetries in the book, the convergence of independent alerts, the residue of insider transactions and regulatory events — these are traces of activity by participants who, like Locard's perpetrators, cannot conduct their business without leaving observable marks.

Reading these traces with the discipline of a criminalist, rather than the bravado of a forecaster, is a different kind of analytical practice than the one that dominates retail finance. It produces fewer claims. The claims it does produce are tied to documented evidence. Each one is preserved with timestamp and methodology disclosure. Each one is governed by convergence rather than inspiration. And each one stays, deliberately, on the impersonal, descriptive, declassified-archive side of the line that separates a publisher from an adviser.

Locard would, perhaps, have appreciated the application. The principle he derived from a microscope in an attic in Lyon proved general enough to remake an entire discipline. There is no obvious reason it should stop at the boundary of the trading floor.