Sealed
Signal is committed before the event
The cell is assigned, but the match, team, side, and units stay hidden until the result can be audited.
DataCores Quant Lab / Signal Key
A public signal artifact: cells are sealed before events begin, updated through live state, revealed after settlement, and archived as a time-sealed record.
Key States
Sealed
The cell is assigned, but the match, team, side, and units stay hidden until the result can be audited.
Live
When coverage is available, the cell moves with the event and reflects whether the signal side is ahead, behind, or neutral.
Revealed
After settlement, the card reveals the match, signal side, odds, portfolio units, and worked or against result.
Archived
Completed keys move into the archive with their original snapshot files and final result audit.
Signal Flow
The key is not a manual pick board. It is the publication layer that sits after research, candidate scoring, snapshot sealing, and policy checks.
The research engine reads the current event slate and active contour registry. Each candidate is scored before publication.
The public layer separates strong publish candidates from observer-only candidates. Rejected candidates are still tracked for later research.
Before the event starts, the full private signal file is hashed and timestamped. The public page shows only the commitment.
When results are known, the hidden signal details are revealed and checked against the sealed snapshot.
Verification
Before an event starts, the private signal file is timestamped through OpenTimestamps. The public page can show the proof immediately without exposing the pick.
After settlement, the human-readable signal file is revealed next to the proof. Anyone can check that the revealed file matches the earlier timestamped commitment.
Public sequence
Audit
Flat
One unit per public signal. This is the cleanest public accounting layer.
Kelly Quarter
A conservative stake view derived from the signal price and portfolio sizing.
Raw diagnostic
Algorithm-weighted units used for research diagnostics, not as the public stake model.
Public
Private research
The point is not to make every signal look good. The point is to publish a record that cannot quietly rewrite itself after the result.