Evidence is the chain behind the screenshot.
Curren evidence means the inspectable chain that connects a signal timestamp to its causal market context, lifecycle, result event, and final public artifact.
A visual result card is not the source of truth. It is the last presentation layer of a chain that should already be internally consistent. If the chain cannot support the entry, the market path, the terminal state, or the result endpoint, the artifact should not be treated as authenticated evidence.
This distinction matters because charts are persuasive even when the underlying methodology is weak. Curren is designed around the opposite order: establish the evidence first, then render the communication layer.
Context ends where the signal begins.
Point-in-time analysis asks a simple question: what information was actually available when the signal existed? A result can look excellent if future candles accidentally leak backward into the features, chart context, or research process. It is also no longer a causal view of the decision.
Curren therefore treats the signal timestamp as a hard evidence boundary. Historical context may extend backward from that timestamp, but future market information must not be used to improve the apparent quality of the original decision frame.
Why this is important
- Backtests remain closer to the information set that would have existed in practice.
- Research can distinguish a genuinely predictive condition from hindsight.
- Public explanations can describe what was known without quietly borrowing from what happened next.
A result should fail closed when authority is incomplete.
Curren's public result model is intentionally conservative. An authenticated result is expected to have enough authority to support the original signal, its entry, its lifecycle, its terminal state, and the event used to describe the outcome. Missing or contradictory authority is not repaired by inventing an aesthetically plausible path.
This is a fail-closed rule: uncertainty can block a public result. The presentation layer should never fill a missing fact with interpolation, a convenient timestamp, or a substituted price simply because the final image would look better.
Authenticated result charts use real market candles.
For an authenticated result artifact, candles represent market evidence rather than illustration. Curren's approved result path does not invent, smooth, or substitute candles. The displayed timeframe may adapt when the native path is too sparse for a readable chart, but the bars shown still need to resolve to real market observations.
The purpose of adaptive display density is readability, not narrative control. A finer timeframe can reveal the path between entry and result more clearly while preserving the actual extrema and causal cutoff.
What the browser demo does differently
The interactive X-result preview on the landing page uses synthetic values to demonstrate layout and interaction. It is clearly separated from authenticated production evidence and should not be interpreted as a performance record.
Price, time, marker, move, and R must describe the same event.
A result becomes ambiguous when the headline percentage is calculated from one point, the chart marker is attached to another, and the timestamp refers to a third. Curren's result grammar instead binds the visible endpoint elements to one selected, proven endpoint.
That means the endpoint price, endpoint time, favorable-move percentage, R multiple, connector, marker, and wick relationship are expected to tell one consistent story. The public artifact can simplify the presentation, but it should not mix incompatible evidence anchors.
Traceability should survive the render.
The useful unit of evidence is not only the PNG that somebody can repost. It is the connection between the normalized market evidence, the result contract, the rendered artifact, and the metadata that identifies those objects consistently.
Hash-bound evidence and artifact metadata make it possible to verify that a reviewed result corresponds to the inputs that were actually admitted. This does not make a trading signal correct by itself; it makes the path from claim to evidence more inspectable.
What this public methodology does — and does not — claim.
Curren describes a methodology for researchable market signals and evidence-backed result artifacts. It does not turn an illustrative website preview into verified performance, and it does not imply guaranteed returns, custody, brokerage, a licensed fund, or automatic account execution.
Public website
Product positioning, methodology, semantic documentation, and an illustrative renderer preview using synthetic values.
Authenticated result path
Server-side authority admission, real market evidence, endpoint verification, provenance, and review before a result is treated as publishable evidence.
The current public product overview is available at https://curren.tech. For deeper first-party notes, browse Curren Research or read why a trading signal screenshot is not evidence. For ongoing public updates, see Curren on X.
Curren turns noisy market motion into signals that can be researched, traced, and challenged before they become a decision.