Where We Are and How We Got Here

A price is a coordinate, not a description

Tell me the market is at a certain level and you have given me the start of a question, not an answer. A price is a single number. It tells you where the market is standing. It says almost nothing about the situation the market is standing in.

Picture two markets at exactly the same level. One arrived after a long, patient climb, with fear low and optimism widening. The other arrived days after a violent drop, with prices still shaking and investors braced for further loss. Same number, opposite circumstances. Read only the current price and the two markets look identical, so you treat them the same. They are not the same at all.

This is the problem I set out to solve. I could see context that the raw price was hiding, and OpenK began as an attempt to show a computer what I was seeing. Two ideas do most of the work. I call them STATE and PATH.

STATE: where the market sits

STATE is the market's position relative to its own history. Not relative to one average, one target price, or one long-run return, because a single anchor is too thin a reference. STATE measures the current level against many historical anchors at once, along with the distributions of where the market has been before.

The question is never simply whether today's price is high or low. It is high or low compared to what, and how unusual is that position? A level that looks ordinary against one stretch of history may look extreme against another. By reading position across several references instead of one, STATE captures whether the market sits somewhere familiar or somewhere it has rarely been.

Be clear about what this means. STATE describes position. It is not a prediction and it is not a timing signal. Knowing the market sits in an unusual place tells you something about vulnerability and how much room there is for expectations to change. It does not tell you when they will change. Overvaluation can persist for a long time. Valuation marks vulnerability, not timing, and that distinction runs through everything I do.

PATH: how the market got here

PATH is the second idea, and the one most people underrate. PATH is the road the market traveled to reach its current state. The same endpoint can be reached in many ways, and the way matters.

A sudden crash and a gradual decline can finish at an identical price. The crash carries the signature of fear arriving quickly, forced selling, and dislocation. The slow decline carries the signature of confidence draining out over time. These are different states of the world wearing the same price tag. Investors inside them feel different things, hold different positions, and may react differently to whatever comes next.

Fear and optimism run through the entire cycle, not only at its turning points, and PATH is where much of that history remains visible. I am not searching for a chart from the past that looks exactly like the present. History does not repeat that cleanly. I am reading the character of the journey: whether the move was a sharp break or a long grind, whether it is accelerating or exhausting itself, and whether instability arrived suddenly or accumulated over time.

The journey is information. Price throws most of it away the moment you reduce the market to one current number.

The published evidence that path matters

This is not a hunch I am asking anyone to accept on faith. Guyon and Lekeufack, in "Volatility Is (Mostly) Path-Dependent," show that volatility depends heavily on the path the market followed, not only its current position. Their work provides strong published evidence for the broader premise that the road matters, not just the destination.

Their research anchors that premise. Where OpenK differs is in scope and intent. They study path dependence in volatility. I am after a broader description of the market: where it sits, how it arrived, and what those two things reveal about fear, optimism, vulnerability, and the changing range of future outcomes.

The underlying intuition is in the published record. OpenK extends that intuition beyond volatility, builds its own descriptions of STATE and PATH from price, and keeps the exact construction private.

Many memories at once

Context does not live at one time horizon. Recent motion and structural position are different questions, and answering both requires more than one memory length.

Short memory captures recent motion. It is sensitive to the character of the current move and to fear or optimism as they change. Long memory captures structural position. It moves more slowly, ignores much of the daily noise, and shows where the market sits relative to its deeper history.

Neither is complete alone. A market can appear calm in short memory while sitting at an extreme in long memory. It can move violently in the short run while remaining close to its longer structure. Both descriptions can be true at the same time.

OpenK therefore reads the market through several memory lengths together. No single window can keep the recent and the structural in view at once.

This also helps keep the K curve, OpenK's moving anchor, connected to the changing market around it. The K curve and the economic loop receive their own treatment elsewhere. The point here is simpler: context changes depending on how far back you look, so the model has to remember more than one version of the past.

Price as a compressed record

Everything described here is built from price alone, with no direct macroeconomic inputs by design.

The working hypothesis is that price history is already a compressed record of the cycle. Growth, liquidity, fear, leverage, expectations, and financial conditions all leave traces in the path investors create. OpenK asks whether enough of that context can be recovered from price itself if the market is described in the right way.

This is not a claim that price contains everything. It is a research choice. Before adding more variables, I want to know how much of the market's condition can be reconstructed from the one record every participant has already helped create.

Where the interesting question lives

The most interesting part is not simply whether price is above or below its structure. It is whether the relationship between the two changes depending on how the market got there.

That relationship appears asymmetric. The downside is often faster, more violent, and more distorted than the upside. After a crash, the market can remain far below its slower structure even as the conditions for recovery begin to form. Near the top, vulnerability may build gradually and remain hidden until expectations finally break.

OpenK studies whether those differences can be read from price early enough to matter.

This work remains in live testing. The question is not whether STATE and PATH can describe the past convincingly. Flexible models can almost always do that. The question is whether they contain information about the future that simpler descriptions miss.

Concept public, construction private

I am open about the concept because the concept is the honest part, and I want it examined.

STATE is where the market sits relative to its own history. PATH is the journey that produced that position. Multiple memory lengths keep the recent and the structural in view at the same time.

What stays private is the construction: how the references are defined, how memory lengths are selected and combined, how the path is represented, and how those descriptions are turned into something a model can use.

The idea is public on purpose. The implementation is the work, and the work is where the value lives.