SKYAIUSDT rose 46% into Aug 3. The record covers the whole lead-up.

Move window: 2026-07-28 02:00 to 2026-08-03 05:00 UTC. The record for SKYAIUSDT currently ends 2026-08-03 00:00 UTC.

This move ran past the end of the record, so the counts below cover the lead-up and the covered part of the move. The rest joins the record after tonight’s rebuild.

Book imbalance · top levels ≥ 0.92 and Book imbalance percentile ≤ 0.062 and Cascade clustering ≥ 0.52

feature.book_imbalance + feature.book_imbalance_pctrank + feature.cascade_clustering_index

Read at 15 minutes before the move.

Minutes like this appeared 79 times in the 43,201 minutes searched (0.2%). What followed was close to a coin flip: 40 rose and 39 fell over the next 30 minutes (79 of the 79 matches have a complete 30 minutes forward).

Of the 79 fires, 71 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 2.5 a day, on 21 of 31 days: spread through the record rather than one event.

q#94371066 @ dataset b91f2197

WATCH FOR THIS Watch the ignition replay

Best supported: Book imbalance · top levels ≥ 0.92 and Book imbalance percentile ≤ 0.062 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.

What the minutes before it looked like

Where these readings were taken

Hourly candles. The marked lines are the minutes the readings came from: 1 minute, 15 minutes, 1 hour, 4 hours and 24 hours before the move began. Every one of them sits before the move, never inside it.

A reading stands out when it sits outside the range it usually holds. The rules are fixed: the same ones the moment reader on the research page uses.

At 1 minute before the move, 2 of the 33 readings stood out. 2026-07-28 01:59 UTC

READING	BEFORE THE MOVE	SEARCHES FOR
Book imbalance · top levels feature.book_imbalance	-0.5138	≤ -0.51
Cascade clustering feature.cascade_clustering_index	0.5279	≥ 0.52

At 15 minutes before the move, 3 of the 33 readings stood out. 2026-07-28 01:45 UTC

READING	BEFORE THE MOVE	SEARCHES FOR
Book imbalance · top levels feature.book_imbalance	0.9223	≥ 0.92
Book imbalance percentile feature.book_imbalance_pctrank	0.062	≤ 0.062
Cascade clustering feature.cascade_clustering_index	0.5279	≥ 0.52

At 1 hour before the move, 4 of the 33 readings stood out. 2026-07-28 01:00 UTC

READING	BEFORE THE MOVE	SEARCHES FOR
Book imbalance percentile feature.book_imbalance_pctrank	0.9188	≥ 0.91
Compression · 15m feature.candle_compression_15m	0.3491	≤ 0.35
Cascade clustering feature.cascade_clustering_index	0.5274	≥ 0.52
Taker buy share · 15m feature.taker_buy_ratio_15m	0.6785	≥ 0.67

At 4 hours before the move, 7 of the 33 readings stood out. 2026-07-27 22:00 UTC

READING	BEFORE THE MOVE	SEARCHES FOR
Book imbalance · top levels feature.book_imbalance	0.964	≥ 0.96
Book imbalance percentile feature.book_imbalance_pctrank	0.9297	≥ 0.92
Compression · 15m feature.candle_compression_15m	2.224	≥ 2.2
Cascade clustering feature.cascade_clustering_index	0.5169	≥ 0.51
Return · 1h feature.ret_1h	-0.02301	≤ -0.023
Taker buy share · 15m feature.taker_buy_ratio_15m	0.2164	≤ 0.22
VWAP distance · day feature.vwap_dist	-0.07285	≤ -0.072

At 24 hours before the move, 4 of the 33 readings stood out. 2026-07-27 02:00 UTC

READING	BEFORE THE MOVE	SEARCHES FOR
Book imbalance · top levels feature.book_imbalance	0.6034	≥ 0.6
Book imbalance percentile feature.book_imbalance_pctrank	0.9365	≥ 0.93
Cascade clustering feature.cascade_clustering_index	0.5012	≥ 0.5
Taker buy share · 15m feature.taker_buy_ratio_15m	0.6816	≥ 0.68

Also counted

This is not a similarity search. Each count matches the exact readings above and nothing else.

Book imbalance · top levels ≤ -0.51 and Cascade clustering ≥ 0.52

feature.book_imbalance + feature.cascade_clustering_index

Read at 1 minute before the move.

Minutes like this are common here: 497 of the 43,201 minutes searched (1.2%), about 16 a day. What followed was close to a coin flip: 231 rose and 266 fell over the next 30 minutes (497 of the 497 matches have a complete 30 minutes forward).

Of the 497 fires, 339 came before the move began and 61 happened while it was already under way. A fire during the move sees it; it does not telegraph it. Timing was checked for the first 400 of the 497 fires.

About 16 a day, on 24 of 31 days: spread through the record rather than one event.

q#18d68439 @ dataset b91f2197

Book imbalance percentile ≥ 0.91 and Compression · 15m ≤ 0.35 and Cascade clustering ≥ 0.52

feature.book_imbalance_pctrank + feature.candle_compression_15m + feature.cascade_clustering_index

Read at 1 hour before the move.

Minutes like this appeared only 17 times in the 43,201 minutes searched. What followed leaned up: 10 of the 17 matches with a complete 30 minutes forward rose (59%).

Of the 17 fires, 14 came before the move began and 3 happened while it was already under way. A fire during the move sees it; it does not telegraph it.

About 0.5 a day, on 15 of 31 days: spread through the record rather than one event.

q#ceb248e6 @ 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.