Research is a work in progress. Estimated release: 1 December 2026. Our goal is to make it a search engine for trading setups in recorded crypto markets.
Binance USDT-MMARSCOINUSDT rose 335% on 3 September 2026
Picked after the rise, on one market; all matches came within 6 days.
Next stepWhat happened
Price opened the 02:00 hour at 0.062 and reached a highest trade of 0.2697 in the 18:00 hour; the tested readings were taken at 01:00.
What stood out at 01:00
- Liquidation bunching0.5311Tested: 0.53 or higher
- 15-minute price change-0.01626Tested: -0.016 or lower
- 1-hour price change-0.05505Tested: -0.055 or lower
01:00 always meets its own rules.
What the record says
About as often as any MARSCOIN minute
Price at least 2% higher after 30 minutes
30 matches in 7,995 minutes checked.
Picked after the rise, on one market.
Evidence
Does it hold outside that week?
Nothing runs until you press Run.
Exactly what was counted3 conditions · 30 of 7,995 minutes · saved text
- Liquidation bunching 0.53 or higher. How bunched in time recent liquidations were; higher means more bunched, and it stays below 1.Cascade clustering ≥ 0.53 · feature.cascade_clustering_index
- 15-minute price change -0.016 or lower. Close to close change over the last 15 minutes, as a fraction of price.Return · 15m ≤ -0.016 · feature.ret_15m
- 1-hour price change -0.055 or lower. Close to close change over the last hour, as a fraction of price.Return · 1h ≤ -0.055 · feature.ret_1h
Read at 1 hour before the move.
Read at 01:00 UTC on 3 September, one hour before the selected 02:00 start.
MARSCOIN only, 8 August to 6 September 2026 (UTC, ending 7 September 00:00); all 3 readings had values on 7,995 of those minutes.
Reference: every MARSCOIN minute over the same dates, matches included; it is not a matched control group.
Outcome: the close exactly 30 minutes later, not the highest or lowest price along the way.
The 02:00 start was picked automatically after the move.
MARSCOINUSDT stored readings end 2026-09-07 00:00 UTC (exclusive). Prices and tick replay have separate coverage.
The feature manifest spans the lead-up. Individual minutes or inputs can still be unavailable.
This is not a similarity search. Each count matches the exact readings above and nothing else.
Picked in hindsight: the counts include every other time these readings appeared, not just this move.
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.
Saved study q#899cfd1c @ dataset c5654b7b, counted 7 Sep 2026 08:34 UTC.
Saved summary text, as written 7 Sep 2026
Written automatically when this study was saved and kept as written; where a line differs from the counts above, the counts govern.
Minutes like this appeared 30 times in the 7,995 minutes searched (0.4%). What followed was more often higher: 18 of the 30 matches with a complete 30 minutes forward rose (60%).
After 1 hour this setup fell 2% or more in 8 of 30 complete windows (26.7%), against 32.8% with no condition (2,623 of 7,995): 0.81x.
This reports the time frame with the biggest gap from any minute, chosen after counting. At least 16 of its 30 windows overlap the rise.
Best point inside 1 hour reached +5% or more in 12 of 30 complete windows; worst point fell 5% or more in 12 of 30.
About 1 a day, on 6 of 31 days: spread through the record rather than one event.
These 6 days are in a row, so the matches are closer to one episode than a spread. It counts the single minute on 7 September as a day.
The other reading times, as written
From 1 minute before
Minutes like this appeared only 18 times in the 7,846 minutes searched. What followed was more often higher: 12 of the 17 matches with a complete 30 minutes forward rose (71%).
12 rose and 5 fell over 17 complete 30m windows. Too few completed windows for a directional generalization. No horizon cleared the floor of 30 complete windows on both the setup and the reference, so no lift is claimed.
About 0.6 a day, on 5 of 31 days: spread through the record rather than one event.
From 15 minutes before
Minutes like this appeared only 17 times in the 7,846 minutes searched. What followed was more often higher: 9 of the 16 matches with a complete 30 minutes forward rose (56%).
9 rose and 7 fell over 16 complete 30m windows. Too few completed windows for a directional generalization. No horizon cleared the floor of 30 complete windows on both the setup and the reference, so no lift is claimed.
About 0.5 a day, on 5 of 31 days: spread through the record rather than one event.
Picked after the rise, on one market; all 30 matches came from 1 September to 6 September, 16 during the rise.
When the matches happened6 days in a row
- 1 Sep1 match
- 2 Sep4 matches
- 3 Sep6 matches
- 4 Sep8 matches
- 5 Sep5 matches
- 6 Sep6 matches
No matches on the other 24 days.
Of the 30 matches, 6 came before the rise began, 16 during it and 8 after it.
Neighbouring minutes overlap, so these are not 30 separate trials.
Picked after the rise, on one market; all 30 matches came from 1 September to 6 September, 16 during the rise.
Other time frames30 minutes to 7 days
The answer uses 30 minutes, the time frame this study counted by default. Each row counts the same matches at a later close, in both directions. The bars show how often price was 2% or more higher at each time frame: first the matches, then any MARSCOIN minute.
- 30 minutes
7 of 30 (23.3%) were 2% or more lower; any MARSCOIN minute 2,111 of 8,025 (26.3%).
Every match has a 30-minute close.
At least 16 of these windows overlap the rise.
- 1 hour
8 of 30 (26.7%) were 2% or more lower; any MARSCOIN minute 2,623 of 7,995 (32.8%).
Every match has a 1-hour close.
At least 16 of these windows overlap the rise.
- 4 hours
12 of 30 (40.0%) were 2% or more lower; any MARSCOIN minute 3,143 of 7,815 (40.2%).
Every match has a 4-hour close.
At least 16 of these windows overlap the rise.
- 24 hours
5 of 24 were 2% or more lower; any MARSCOIN minute 1,799 of 6,615 (27.2%).
6 without a 24-hour close yet, left out.
At least 16 of these windows overlap the rise.
- 72 hours
0 of 11 were 2% or more lower; any MARSCOIN minute 0 of 3,735 (0.0%).
19 without a 72-hour close yet, left out.
At least 5 of these windows overlap the rise.
- 7 days
None of the 30 matches has a 7-day close yet, so there is no rate.
The saved summary picked, after counting, the time frame with the biggest gap from any minute: 1 hour.
After 1 hour this setup fell 2% or more in 8 of 30 complete windows (26.7%), against 32.8% with no condition (2,623 of 7,995): 0.81x.
At least 16 of those windows ran through the rise itself, so it partly counts the rise as an outcome.
Along the way (the highest and lowest price inside 1 hour, not the close):
Best point inside 1 hour reached +5% or more in 12 of 30 complete windows; worst point fell 5% or more in 12 of 30.
Picked after the rise, on one market; all 30 matches came from 1 September to 6 September, 16 during the rise.
Readings at five times before the rise24 hours to 1 minute before · 2 other times tried
Highlights follow fixed rules, not significance tests. Proposed conditions round values or use ranges; inspect them before running.
24 hours before · 2 Sep 02:00 UTC · 6 of 58 stood out
Order book tilt 0.5 or higher15-minute range against usual 2.2 or higherLiquidation bunching 0.58 or higher15-minute price change -0.069 or lower1-hour price change -0.095 or lowerDistance from the day's average price -0.059 or lowerShow the full readings
Reading Value at that minute Stored rule Book imbalance · top levels feature.book_imbalance 0.5013 ≥ 0.5 Compression · 15m feature.candle_compression_15m 2.261 ≥ 2.2 Cascade clustering feature.cascade_clustering_index 0.5836 ≥ 0.58 Return · 15m feature.ret_15m -0.06955 ≤ -0.069 Return · 1h feature.ret_1h -0.09563 ≤ -0.095 VWAP distance · day feature.vwap_dist -0.05951 ≤ -0.059 4 hours before · 2 Sep 22:00 UTC · 3 of 58 stood out
Liquidation bunching 0.52 or higher15-minute price change -0.011 or lower1-hour price change -0.051 or lowerShow the full readings
Reading Value at that minute Stored rule Cascade clustering feature.cascade_clustering_index 0.523 ≥ 0.52 Return · 15m feature.ret_15m -0.01136 ≤ -0.011 Return · 1h feature.ret_1h -0.05118 ≤ -0.051 1 hour before · 3 Sep 01:00 UTC · 3 of 58 stood out
Liquidation bunching 0.53 or higher15-minute price change -0.016 or lower1-hour price change -0.055 or lowerShow the full readings
Reading Value at that minute Stored rule Cascade clustering feature.cascade_clustering_index 0.5311 ≥ 0.53 Return · 15m feature.ret_15m -0.01626 ≤ -0.016 Return · 1h feature.ret_1h -0.05505 ≤ -0.055 15 minutes before · 3 Sep 01:45 UTC · 5 of 58 stood out
Order book tilt 0.71 or higher15-minute range against usual 2 or higherLiquidation bunching 0.53 or higher15-minute price change 0.028 or higher1-hour price change 0.021 or higherShow the full readings
Reading Value at that minute Stored rule Book imbalance · top levels feature.book_imbalance 0.7147 ≥ 0.71 Compression · 15m feature.candle_compression_15m 2.037 ≥ 2 Cascade clustering feature.cascade_clustering_index 0.5399 ≥ 0.53 Return · 15m feature.ret_15m 0.02827 ≥ 0.028 Return · 1h feature.ret_1h 0.02135 ≥ 0.021 1 minute before · 3 Sep 01:59 UTC · 5 of 58 stood out
Order book tilt -0.79 or lower15-minute range against usual 2 or higherLiquidation bunching 0.53 or higher15-minute price change -0.011 or lower1-hour price change 0.032 or higherShow the full readings
Reading Value at that minute Stored rule Book imbalance · top levels feature.book_imbalance -0.7926 ≤ -0.79 Compression · 15m feature.candle_compression_15m 2.037 ≥ 2 Cascade clustering feature.cascade_clustering_index 0.5399 ≥ 0.53 Return · 15m feature.ret_15m -0.01118 ≤ -0.011 Return · 1h feature.ret_1h 0.03253 ≥ 0.032
Two other reading times we tried 18 and 17 matches
1 minute before (01:59): order book tilt -0.79 or lower, 15-minute range against usual 2 or higher, liquidation bunching 0.53 or higher.
Book imbalance · top levels ≤ -0.79 and Compression · 15m ≥ 2 and Cascade clustering ≥ 0.53
18 matches: price was 2% or more higher after 30 minutes in 7 of 17, and 2% or more lower in 3 of 17.
Of the 18 matches, 4 came before the rise began, 10 during it and 4 after it.
15 minutes before (01:45): order book tilt 0.71 or higher, 15-minute range against usual 2 or higher, liquidation bunching 0.53 or higher.
Book imbalance · top levels ≥ 0.71 and Compression · 15m ≥ 2 and Cascade clustering ≥ 0.53
17 matches: price was 2% or more higher after 30 minutes in 6 of 16, and 2% or more lower in 6 of 16.
Of the 17 matches, 3 came before the rise began, 10 during it and 4 after it.
Fewer than 30 results, so no rate is printed.
No set had 30 matches before the rise; the saved study leads with the one that had the most.
Picked after the rise, on one market; all 30 matches came from 1 September to 6 September, 16 during the rise.
Hourly candles around the risefull context, lead-up, move
The 02:00 start was picked after the move, and the hour before was already rising.
Hourly candles, UTC. The marked lines are the minutes the readings came from: 1 minute, 15 minutes, 1 hour, 4 hours and 24 hours before the selected start.
Other ways to test itsame dates
- This market only
- The 10 large cap markets · No result for these markets has been measured.
- Every recorded market
Same 3 conditions on MARSCOIN, 9 July to 7 August 2026.
Each opens Review the test.
Running needs a free account and uses your allowance; you see the cost first.