ADPTSettlement Oracle
BTC loading…

Calibrated on real settled outcomes

Where Bitcoin actually settles

Not a price target. The full probability distribution — every price level, every horizon — estimated from historical strike-market observations and scored against the settlement outcomes that followed.

Snapshot
Ask the oracle settlement-derived
Will Bitcoin reach within ?
Modeled intrawindow touch
Reflection-model approximation, not the nine-year empirical daily-touch statistic.
Still above it at the end
Where price actually finishes. Always the smaller of the two.
Distance
Probability Field settlement-derived spot: composite
UnlikelyLikely
Settles above
StrikeSettles aboveSettles belowDensity

The market check Kalshi market pricing

The same strikes, two independent answers when both snapshots pass their freshness gates: historical settlement frequency and the captured market midpoint. Stale or expired rows are not shown as current. Divergence is information, not advice.

How to read this. "Kalshi yes" is the midpoint of the captured bid/ask on the market that pays out if Bitcoin settles above that strike at close. "Ours" is the counted settlement-history surface, read at the event's actual time to close. The gap is shown in percentage points; a wide market spread means the midpoint itself is soft. Neither column is a recommendation — when they disagree, at least one of them is wrong, and this page does not claim to know which.

Signal convergence exchange data

The same indicators, read on three timeframes at once. They are descriptive technical readings, not an independently validated trading edge. Readings disappear when their source snapshot is stale.

Signal Snapshot

What price touches vs where it settles daily · 9-year history

Counted over years of daily bars — daily resolution, no intraday claims. Each cell is the chance price touches that level after the setup (what matters for a target or a stop) — settling there is always less likely, and the surface above answers that. Next to each figure, the edge over simply picking a random day: a big number with no edge is worth nothing, so we show both.

Model forecast experimental

The prediction engine's latest published numeric calls — shown next to what its own scored-record snapshot says calls like these have actually been worth. Stale calls are hidden. The stated number is the bot's confidence; the calibrated number maps it through resolved predictions. When calibration lands on a coin flip, the honest output is an abstention, and that is what you will see.

Loading the scored-record snapshot…

The dormancy wall node data

Every coin spent, sorted by how long it sat still first. The cool bands at the base are ordinary churn — coins moving within a day. When the hot bands at the top flare, Bitcoin that had not moved in years just did. That is the signal almost nothing else shows you block by block, and it comes from our own node.

Coins Spent by Age Held
Recently movedLong dormant

The receipts settlement truth

A forecast is worth exactly what its calibration proves. This surface was fitted on history and scored on markets it never saw — against the Black-Scholes model the industry defaults to.

HorizonFitted rowsScored rows Lognormal BrierSurface BrierBrier changeStatus
How to read this. Brier loss measures probability accuracy — lower is better. These are correlated strike rows, not independent future events. The 7-day row is experimental and is excluded from validated aggregate claims.

The chain underneath node data

Computed block by block from a sovereign Bitcoin full node — our own indexes, no third-party data vendor standing between us and the ledger. Choose what you want on screen.

Access

The historical chart is open. Current strike interactions are available only while their input snapshot passes the freshness gate. Future paid access remains disabled.

Open
Free
  • Probability field
  • One horizon at a time
  • Every on-chain metric
  • Calibration receipts
Historical surface available
Day pass
Burn to unlock
  • Every horizon at once
  • Exact strike probabilities
  • Strike alerts
  • Cone export
Awaiting token
Week pass
Burn to unlock
  • Everything in Day
  • Full surface API
  • Historical replay
  • Saved indicator sets
Awaiting token
Access burns tokens rather than gating on a held balance — you pay for what you use, and supply shrinks as the chart earns. Nothing is enforced until the token exists; until then every layer marked is simply switched off, for everyone.