What is firing now, and how often it has happened
These 21 conditions are curated entry points into the larger EdgeDepth market microstructure search engine. Each one is joined to how often it occurred in the record and the full distribution of what followed: counts with denominators, drawdown shown next to upside, and a reproducibility key on every receipt. This reports what has happened when a condition held. It never recommends, targets, or personalizes. To define your own condition across crypto and TradFi perps, use Research, the API, or MCP.
Firing now (provisional)
The live firing read is not yet connected here. Every condition below still carries its full historical base rate and outcome distribution from the nightly receipts.
Microstructure
VPIN in the critical regime
Toxic order flow (VPIN) has entered its critical regime, the highest of the four normalized bands.
113 occurrences · reached +5% in 27 of 111 (24h)
VPIN above 0.85
The raw VPIN estimate of order-flow toxicity is reading above 0.85 on its 0 to 1 scale.
54 occurrences · reached +5% in 18 of 53 (24h)
Order book heavily bid-sided
The order book has averaged at least half of its depth on the bid side over the sampling window.
418,450 occurrences · reached +5% in 89,129 of 383,490 (24h)
Order book heavily ask-sided
The order book has averaged at least half of its depth on the ask side over the sampling window.
418,788 occurrences · reached +5% in 88,688 of 383,760 (24h)
Book imbalance above its 95th percentile
Order-book imbalance is reading above its own trailing 95th percentile for this market.
162,487 occurrences · reached +5% in 30,554 of 150,967 (24h)
Spread blown out toward its cap
The normalized spread has widened to at least 0.8 of its capped range: a spread blowout.
6,189 occurrences · reached +5% in 1,894 of 5,849 (24h)
Two-thirds of the tape on one side
Taker buy ratio over the last 15 minutes has run at two-thirds or more, a persistently one-sided tape.
241,586 occurrences · reached +5% in 41,824 of 217,335 (24h)
Whale prints carrying a third of volume
Large trades have accounted for at least a third of traded volume over the last 15 minutes.
7,632 occurrences · reached +5% in 842 of 7,396 (24h)
One-sided tape: buyers
Taker buy ratio over the last 15 minutes has run at 0.62 or more: a persistently buyer-heavy tape.
280,604 occurrences · reached +5% in 51,616 of 254,939 (24h)
One-sided tape: sellers
Taker buy ratio over the last 15 minutes has run at 0.38 or less: a persistently seller-heavy tape.
293,012 occurrences · reached +5% in 55,105 of 266,618 (24h)
Liquidations
Liquidation intensity elevated
Normalized liquidation intensity is reading above 0.5: liquidations arriving well above this market baseline.
1,955 occurrences · reached +5% in 733 of 1,884 (24h)
Liquidations arriving in cascades
The cascade clustering index is above 0.6: liquidations are arriving in bursts rather than independently.
604 occurrences · reached +5% in 280 of 587 (24h)
At least $5M liquidated in one hour
Observed liquidation notional over the trailing one-hour window is at least $5 million.
91 occurrences · reached +5% in 10 of 87 (24h)
OI accelerating into liquidations
Open-interest velocity above its own 90th percentile while at least $1M of liquidation notional printed in the trailing hour.
531 occurrences · reached +5% in 97 of 501 (24h)
Funding + open interest
Funding pinned at the extreme of its range
Normalized funding magnitude is at 0.9 or above: funding pinned near the extreme end of its range.
302 occurrences · reached +5% in 133 of 288 (24h)
Funding deeply negative into settlement
Funding is at or below -1% with 30 minutes or less to the next settlement.
61 occurrences · reached +5% in 37 of 59 (24h)
Open interest velocity above its 90th percentile
The rate of change of open interest is reading above its own trailing 90th percentile for this market.
241,006 occurrences · reached +5% in 55,906 of 231,468 (24h)
Crowd long, leaders not
Global accounts sit at least 65% long while top traders run at least 10 points less long than the crowd.
6,739 occurrences · reached +5% in 1,232 of 5,459 (24h)
Volatility + price
Realized volatility above its 90th percentile
One-minute realized volatility over the hour is above its own trailing 90th percentile for this market.
49,464 occurrences · reached +5% in 11,654 of 46,776 (24h)
Stretched 2% above the day VWAP
Price is at least 2% above the current day VWAP: stretched to the upside of its session anchor.
29,425 occurrences · reached +5% in 10,438 of 28,566 (24h)
Candle formations
Range compression on the 15-minute candle
The last completed 15-minute candle reads compressed against its recent range (compression at least 2).
86,621 occurrences · reached +5% in 18,782 of 79,337 (24h)
Swept lows and closed back above (15m)
The last completed 15-minute candle swept prior lows and closed back above them: the "bullish" sweep and reclaim formation.
79,834 occurrences · reached +5% in 16,731 of 71,769 (24h)
Hourly engulfing candle
The last completed hourly candle is an engulfing formation, either the "bullish" or "bearish" variant.
28,530 occurrences · reached +5% in 6,191 of 24,802 (24h)
Swept the lows and reclaimed (1h)
The last completed one-hour candle swept prior lows and closed back above them: the "bullish" sweep and reclaim formation on the hourly.
27,542 occurrences · reached +5% in 5,404 of 23,864 (24h)
Ordered sequences
Spread blowout, then a liquidation burst
A spread blowout followed by an elevated liquidation burst within 30 minutes.
26 occurrences · reached +5% in 17 of 24 (24h)
Liquidation cascade, then a reclaim
An elevated liquidation burst followed within 30 minutes by a 15-minute candle that swept lows and reclaimed them.
415 occurrences · reached +5% in 135 of 397 (24h)
How to read these
- Counts ship with denominators: occurrences across the eligible symbol-buckets scanned. A rate is over the occurrences with a computable horizon (present), never the total.
- Absent is absent. When a feature was not warm at a minute, that bucket leaves the denominator and is reported; it is never coerced to zero.
- This selects a setup, never its outcome. The grammar forbids filtering on what followed (OUTCOME_IN_PREDICATE), so the distribution below is over every occurrence, not a survivorship-picked subset.
Occurrences are observed examples in a selected archive, not a guarantee. Outcomes are computed forward from each anchor and are descriptive, never a filter and never a prediction. Nothing here is trading advice.