AIOUSDT fell 71% into Aug 2. The record covers the whole lead-up.
Move window: 2026-07-27 14:00 to 2026-08-02 21:00 UTC. The record for AIOUSDT currently ends 2026-08-03 00:00 UTC.
Book imbalance percentile ≤ 0.074 and Compression · 15m ≤ 0.45 and Cascade clustering ≥ 0.52
feature.book_imbalance_pctrank + feature.candle_compression_15m + feature.cascade_clustering_index
Read at 1 minute before the move.
Minutes like this appeared 87 times in the 43,201 minutes searched (0.2%). What followed was close to a coin flip: 43 rose and 43 fell over the next 30 minutes (86 of the 87 matches have a complete 30 minutes forward).
Of the 87 fires, 80 came before the move began and 7 happened while it was already under way. A fire during the move sees it; it does not telegraph it.
About 2.8 a day, on 22 of 31 days: spread through the record rather than one event.
q#d6909105 @ dataset b91f2197
Best supported: Book imbalance percentile ≤ 0.074 and Compression · 15m ≤ 0.45 and Cascade clustering ≥ 0.52, the combination with the most fires before the move began.
Opens this exact search in the research tool. Nothing runs until you confirm it there, and arming it as an alert sends a message when it next becomes true.
Picked in hindsight: the counts include every other time these readings appeared, not just this move.
Also counted
This is not a similarity search. Each count matches the exact readings above and nothing else.
Book imbalance percentile ≤ 0.095 and Compression · 15m ≤ 0.45 and Cascade clustering ≥ 0.52
feature.book_imbalance_pctrank + feature.candle_compression_15m + feature.cascade_clustering_index
Read at 15 minutes before the move.
Minutes like this appeared 100 times in the 43,201 minutes searched (0.2%). What followed was close to a coin flip: 49 rose and 49 fell over the next 30 minutes (98 of the 100 matches have a complete 30 minutes forward).
Of the 100 fires, 92 came before the move began and 8 happened while it was already under way. A fire during the move sees it; it does not telegraph it.
About 3.2 a day, on 24 of 31 days: spread through the record rather than one event.
q#604ad507 @ dataset b91f2197
Book imbalance · top levels ≥ 0.88 and Book imbalance percentile ≥ 0.91 and Cascade clustering ≥ 0.52
feature.book_imbalance + feature.book_imbalance_pctrank + feature.cascade_clustering_index
Read at 1 hour before the move.
Minutes like this are common here: 308 of the 43,201 minutes searched (0.7%), about 9.9 a day. What followed was close to a coin flip: 160 rose and 133 fell over the next 30 minutes (293 of the 308 matches have a complete 30 minutes forward).
Of the 308 fires, 294 came before the move began and 14 happened while it was already under way. A fire during the move sees it; it does not telegraph it.
About 9.9 a day, on 24 of 31 days: spread through the record rather than one event.
q#0c92ffa0 @ dataset b91f2197
Fires often arrive as clusters of consecutive minutes, and forward windows overlap. A cluster is one episode, not many independent trials.
If these counts read weaker than the move suggested, that is the tool working: the lead-up of a picked move always looks special until it is counted everywhere else.