About · Tollbridge
Crypto intelligence you can act on
Probabilities for crypto markets, wallets and transactions — for people and for agents.
Tollbridge turns market state, wallet activity and blockchain data into decisions you can use.
Eleven questions. One kind of answer.
Browse markets →Every probability comes with the question, the time window, the conditions, the confidence and the data age.
26 questions · 16 askable now · 7 waiting on an address · 3 not available yet (liquidation tape BTC 7/24 · ETH 4/24 · SOL 9/24) — S01 · S05 · S06 · S07 · S08 · S09 · S10 · S11 · S12 · S13 · S14 · S15 · S16 · S17 · S18 · S19 · S20 · S21 · S22 · S23 · S24 · S25 · S26 are computed from current data, and S02 · S03 · S04 are not sold on this deployment: the market-wide liquidation tape is being recorded from a named market-data provider, but it has not held 24 hourly observations yet — this question stays off the default rail until it does. Availability is read live — this page never claims a question is live on its own.
- S01availablechoice
Market regime
Probability distribution over the 60-minute market regime: trending up, trending down, range compression or range expansion — exactly one label obtains.
window 60mhourly price candles - S02not availableprobability
Liquidation risk
Probability that perp liquidation velocity reaches twice its trailing-24h median within 60 minutes, conditional on open interest, funding and book depth.
window 60mliquidations, open interest, funding, order bookNot available yet — the market-wide liquidation tape is being recorded from a named market-data provider, but it has not held 24 hourly observations yet — this question stays off the default rail until it does (hourly liquidations on the goldrush tape: BTC 7/24 · ETH 4/24 · SOL 9/24). A request still returns a clear "no decision" and refunds your credits; it is off the default pages.
- S03not availableprobability
Long squeeze
Probability that long-side liquidation notional within 60 minutes reaches twice the trailing-24h hourly median — a long-unwind cascade.
window 60msided liquidations, fundingNot available yet — the market-wide liquidation tape is being recorded from a named market-data provider, but it has not held 24 hourly observations yet — this question stays off the default rail until it does (hourly liquidations on the goldrush tape: BTC 7/24 · ETH 4/24 · SOL 9/24). A request still returns a clear "no decision" and refunds your credits; it is off the default pages.
- S04not availableprobability
Short squeeze
Probability that short-side liquidation notional within 60 minutes reaches twice the trailing-24h hourly median — a short-covering cascade.
window 60msided liquidations, fundingNot available yet — the market-wide liquidation tape is being recorded from a named market-data provider, but it has not held 24 hourly observations yet — this question stays off the default rail until it does (hourly liquidations on the goldrush tape: BTC 7/24 · ETH 4/24 · SOL 9/24). A request still returns a clear "no decision" and refunds your credits; it is off the default pages.
- S05availableprobability
Volatility expansion
Probability that realized volatility over the next 60 minutes exceeds 1.5× the trailing-24h baseline, both measured on 1-minute returns.
window 60mhourly price candles - S06availableprobability
Funding anomaly
Probability that the next funding settlement prints at least 3× the trailing-7-day median absolute rate; unresolvable — not false — if no settlement lands in the window.
window 60mfunding - S07availableprobability
Funding persistence
Probability that an extreme funding rate holds the same sign at ≥3 consecutive settlements within 24 hours — a funding regime, not a one-off print.
window 24hfunding - S08availableprobability
Open-interest anomaly
Probability that at least one 1-hour interval within 24 hours sees open-interest change exceed 3σ of its trailing-24h rhythm, standardized by its own history.
window 24hopen interest - S09availableprobability
Wallet accumulation
Probability that a tracked Hyperliquid account holds a long position in the coin at the end of a 24-hour window, with the position worth at least a quarter of the account’s own equity.
window 24hHyperliquid account positions - S10availableprobability
Wallet distribution
Probability that a tracked Hyperliquid account holds a short position in the coin at the end of a 24-hour window, with the position worth at least a quarter of the account’s own equity.
window 24hHyperliquid account positions - S11availablechoice
Transaction risk
Distribution over allow, review or block for a Hyperliquid account’s liquidation proximity: how close its nearest position sits to its liquidation price, and how leveraged the account is.
window 60mHyperliquid account positions - S12availableprobability
Momentum exhaustion
Probability that momentum is exhausted: the 14-period relative-strength index at an extreme while the most recent 1-hour move is already smaller in magnitude than the one before it.
window 60mhourly price candles - S13availableprobability
Volume drought
Probability that traded volume has fallen to under a third of its trailing-24h norm — thin participation that precedes gaps.
window 24hhourly price candles - S14availableprobability
Liquidity migration
Probability that the market’s liquidity score has shifted by at least thirty percent from its previous 24-hour level — depth moving before price.
window 24hhourly price candles - S15availableprobability
Basis stress
Probability that the perp-vs-spot basis is at least 50 basis points away from index — a structural divergence from crowded positioning.
window 24hperp basis (mark vs index) - S16availableprobability
Spread blowout
Probability that the quoted bid-ask spread has widened to 10 basis points or more.
window 60mthe order book (top-of-book spread) - S17availableprobability
Book imbalance flip
Probability that the resting-depth imbalance has flipped sign — buyers’ resting depth has become sellers’, or vice versa.
window 60mthe order book (resting-depth imbalance) - S18availableprobability
Depth withdrawal
Probability that resting depth has fallen by at least thirty percent from an already-thin book — liquidity being pulled.
window 60mthe order book (resting depth) - S19availableprobability
Liquidation burst
Probability that liquidations are clustered: a single minute carrying at least 3× the window’s hourly liquidation velocity.
window 60mthe market-wide liquidation tape (per-minute print journal) - S20availableprobability
Liquidation severity
Probability that liquidation severity reaches 15 basis points per unit liquidated — forced flow moving price more than a healthy book absorbs.
window 60mthe market-wide liquidation tape (severity per dollar) - S21availableprobability
Mark-slippage stress
Probability that liquidations are printing at least 20 basis points away from the venue mark — the mark understating stress.
window 60mthe market-wide liquidation tape (mark vs fill) - S22availableprobability
Liquidation side pressure
Probability that forced flow is strongly one-sided and hitting the side that was already crowded — the precondition for a cascade.
window 60mthe market-wide liquidation tape (sided pressure) - S23availableprobability
Wallet net outflow
Probability that a tracked wallet’s net flow is materially negative with an elevated distribution ratio — capital leaving with intent.
window 24hwallet transfer flows - S24availableprobability
Wallet concentration
Probability that a single counterparty accounts for at least half of a tracked wallet’s flow.
window 24hwallet transfer flows - S25availableprobability
Unusual wallet activity
Probability that a tracked wallet is doing at least three times its own recent normal — an unusual burst.
window 60mwallet transfer flows - S26availableprobability
Account deleverage
Probability that a tracked account’s exposure ratio has fallen by at least thirty percent from its prior reading — actively cutting risk.
window 60mHyperliquid account positions
System status
Read live from this deployment’s health sweep — dark stays dark.
- hyperliquid-resthealthyok 2026-10-17 05:29
- hyperliquid-universehealthyok 2026-10-17 05:29
- hyperliquid-accounthealthyok 2026-10-17 05:29
- base-rpchealthyok 2026-10-17 05:29
- binance-resthealthyok 2026-10-17 05:29
Embed
Drop a live view into your own page — the same records, nothing duplicated.
Question card
One question’s live claim for a market, sized for a sidebar.
<iframe src="/embed/question?asset=btc" width="360" height="480" style="border:0" title="Tollbridge question"></iframe>
Intelligence card
The market strip with the current ladder step.
<iframe src="/embed/intelligence" width="720" height="320" style="border:0" title="Tollbridge intelligence"></iframe>
Ruleset in force: version 1.0.0 · checksum 30efa65972f4f55b41c69c8d… (re-derived on this read, identical) See it on any record →
How we report results
Probabilities are calibrated against measured outcomes; calibration history is published only when sample size is sufficient.
When the data is too old or the sample is too small, the question returns no decision — it shows the reason verbatim (for example, insufficient sample) instead of a made-up number.
No performance claims: this page states what each question asks and what it is computed from — nothing about how often events occur. Paid questions show the exact price before anything is asked.