Discovery & research
The builder assumes you know what you want to trade. The discovery suite is for when you do not. Six screens that scan the market for defined technical conditions and report what they found, with the reasoning behind every score.
Read this framing first
Every screen in this section names instruments and price levels. That is what a scanner does. None of it is a recommendation to trade, an order, or a statement about profitability.
The platform's own words, returned with every opportunity detail:
This is a research candidate, not investment advice, not an order, and not a profitability guarantee.
And on the analog engine:
Historical frequencies of past events, not a forecast, not a probability, and not investment advice. Past outcomes do not constrain future ones.
A score on these screens measures how completely a defined pattern is present right now. It is not a probability of profit, not statistical confidence, and not a forecast. See Compliance boundaries below.
The six screens
| Screen | Question it answers | Route |
|---|---|---|
| Opportunity Radar | Which instruments currently show a defined setup, and where would it invalidate? | /opportunity-radar |
| Universe Scanner | Across a named universe, which instruments pass my filters? | /universe-scanner |
| Trading Map | What is the whole scanned set doing, ranked by setup quality? | /trading-map |
| Cluster Density | Where in the market (region × asset class) is setup activity concentrated? | /cluster-density |
| Analog Events | When conditions looked like this before, what happened next? | /analog-events |
| Asset Explorer | What is this instrument, and what data exists for it? | /asset-explorer |
Plus an Opportunity Watchlist at
/watchlist/opportunities for tracking candidates you want to revisit, distinct from the
strategy watchlist.
The vocabulary
These terms appear across every screen and are defined nowhere else. All of them are also in the glossary.
Setup Quality
A 0-100 score for how completely a detector's conditions are met on an instrument right now. It is the sum of five components minus any penalties, clamped to 0-100:
| Component | Max points | What it measures |
|---|---|---|
| Setup Strength | 35 | How completely the detector conditions are met |
| Liquidity | 25 | Volume and turnover, spread where available, instrument eligibility |
| Regime | 20 | Alignment with trend/volatility regime and market breadth |
| Timing | 10 | Session fit and distance to trigger |
| Quality | 10 | Coverage, validity and timeliness of the required inputs |
Penalties subtract up to 30 points:
| Penalty | Fires when |
|---|---|
| High-impact event proximity | A scheduled event is close enough to distort the setup |
| Stale data | The inputs are older than the screen's freshness window |
| Extreme spread | The quoted spread is wide enough to make the setup untradeable |
| Invalid market state | The instrument is in a state where the setup cannot be read |
It measures pattern completeness and data quality. A score of 82 does not mean an 82% chance of anything. The product says this itself, on the screen.
Setup Grade
A band derived from Setup Quality. Four values:
| Grade | Score |
|---|---|
| STRONG | ≥ 80 |
| GOOD | ≥ 65 |
| WATCH | ≥ 50 |
| LOW | below 50 |
The product's own definition, worth quoting because it is exactly right: "Setup Grade is a band derived from Setup Quality. It is not statistical confidence or a probability of profit."
Opportunities below a publication floor of 50 are not surfaced at all.
Status
Where a candidate is in its own lifecycle:
| Status | Meaning |
|---|---|
FORMING | The conditions are developing but not complete |
READY | Conditions met; the trigger has not been reached |
TRIGGERED | The trigger level has been crossed |
INVALIDATED | The invalidation level was reached, the setup is void |
EXPIRED | The setup aged out without triggering |
Only READY and TRIGGERED candidates are published.
Trigger and Invalidation
Two price levels, and the pairing is the point.
- Trigger, the price at which the detector considers the setup to have activated.
- Invalidation, the price at which the setup is void. Not a suggested stop-loss: a statement that the pattern the screen identified no longer exists.
Trigger distance is how far the current price sits from the trigger, as a percentage.
Why now
The evidence list. Every opportunity carries plain-language statements explaining each component's points, computed from the same numbers the score used. For example:
Close 2,847.30 is above 20-session high 2,841.00 (+0.22%).
Volume 4.2M vs 20-session average 2.8M (1.50×).
Trigger 2,841.00; distance +0.22%.
This is the feature that separates a scanner you can reason about from a black box. If you disagree with the evidence, you have grounds to ignore the score.
Event risk
Scheduled events near the setup, earnings, policy decisions, expiry. Proximity to a high-impact event applies a scoring penalty and is listed separately, because the penalty tells you the score was reduced and the list tells you what to look up.
Liquidity band
Derived from the liquidity component: HIGH at ≥ 18 points, MEDIUM at ≥ 10, LOW below.
Freshness
FRESH, DELAYED or STALE, with the age in seconds. Stale data both applies a penalty and is
labelled. You always know how old what you are looking at is.
Snapshot
Every screen reads from a snapshot with an ID, an as-of timestamp, and a scoring version. Two consequences worth knowing: a screen is not a live tick feed, and a result you saw can be cited by its snapshot ID rather than described from memory.
The three detectors
Radar and Trading Map both run the same detector set. Each is a defined technical condition, not a proprietary model:
| Detector | Looks for |
|---|---|
| Breakout | Close at or above an N-session high, with volume confirmation |
| Momentum continuation | Multi-session return, position relative to a moving average, up-day count, volume |
| Volatility expansion | Today's range against ATR, plus range position and volume |
Each publishes its own key and version, so a result can be attributed to a specific detector revision.
From a setup to a strategy
Discovery hands off to the builder rather than to an order. See From a setup to a strategy for what carries across, what does not, and why the handoff is deliberately not a one-click trade.
Compliance boundaries
The discovery suite is the most recommendation-shaped part of the platform, so its boundaries are stated explicitly rather than assumed.
What these screens do: report a measured technical condition on an instrument, name the price levels at which the condition activates and voids, publish the formula that produced the score, and list the evidence behind every component.
What they do not do: rank instruments by expected return, estimate probability of profit, suggest position size, suggest a stop-loss, take your circumstances into account, or tell you to trade anything.
Known gap, stated rather than hidden. The service returns a disclaimer string with every opportunity detail. Trading Map, Cluster Density and Analog Events render theirs. Opportunity Radar currently does not render the disclaimer it receives, the field is present in the client and unused. Radar is also the most recommendation-shaped of the screens, since it names an instrument, a trigger and an invalidation, then offers a Build strategy button. Rendering it is a P0 fix on the compliance roadmap; the disclaimer text applies whether or not the screen shows it, and it is reproduced at the top of this page and on Opportunity Radar.