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.
Checking readings…
A decision time that has not been read recently can take several seconds.
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.
A decision time that has not been read recently can take several seconds.
Decision time is the moment you imagine deciding. Every reading here is from the minute before it. Here it sits at the start of the move, which is only known in hindsight. Click the chart to move it.
Close-to-close change before the decision.
More open contracts; long or short side is not identified.
Exchange-reported forced trades; feed coverage is limited.
Preparing editable conditions…
Use this minute to form a question about similar conditions. The later move is known in hindsight; these readings alone do not explain its cause or show that it was predictable.
Recorded readings minute by minute, in UTC
4 Nov 14:44 to 14:59 UTC · 16 minutes
Reads each minute in turn. Past minutes need a signed-in Pro or Research plan.
7 of 7 key readings available
The automatic start is the last covered hourly candle touching the bottom 5% of the episode range (top 5% for a fall), measured using the eventual extreme. It is a hindsight anchor, not a detected causal onset. Change the start above to inspect a later acceleration. The end marks the hour containing the price extreme, not an exact tick. The feature manifest spans the lead-up. Individual minutes or inputs can still be unavailable.
The earlier snapshots are still being read.
Reads 21 sampled minutes from the selected start to one hour after the end, with extra samples in the peak hour: separate snapshots, not a continuous tape or an advance-warning test.
No additional reads have run.
Historical access applies. Missing values stay unavailable; liquidations are Binance publications, not venue totals; the OI score is an absolute USD change, not signed contract growth.
Still counting. The comparison appears here when it finishes.
Free replay covers BTCUSDT on Binance Futures, the past 3 days. Pro reaches 30 days; Research reaches 90 days. Access is checked when you open a replay.
No account? Replay a real move tick by tick, free: TUT short squeeze · 9 Aug 2026.
Reading the market 15 minutes before the decision (1 of 4) · 0s
Reading the market 1 hour before the decision (2 of 4) · 4s
Reading the market 4 hours before the decision (3 of 4) · 8s
Counting how often the standout readings held over 30 days (1 of 3) · 12s
Comparing with every other TUT minute in those 30 days · 16s
Close-to-close change before the decision.
More open contracts; long or short side is not identified.
Exchange-reported forced trades; feed coverage is limited.
The comparison measures: Price at least 2% lower after 30 minutes. It does not test a repeat of the entire selected fall.
Treat this as a candidate condition. Keep its definition fixed and check a separate period, including failures, before applying it to another market.
A rare reading and a useful predictor are different questions. No uniqueness or causal mechanism is established by this comparison.
Readings from 2026-11-04 14:59 UTC
All conditions must match. These thresholds are editable in the test.
Current recorded classification and sector tags, not point-in-time membership. Historical membership may differ. Instrument history does not guarantee every setup input or outcome is present; actual study exclusions remain authoritative.
Review conditions, markets, dates and cost before Run. Then save the result and enable supported monitoring.
Sector and volume lists are unavailable. Retry or explicitly choose another population.
2026-10-10 to 2026-11-09 UTC. Recorded membership at this dataset revision, not a claim of historical membership.
No resolved roster.
The conditions were selected while investigating this move. This is exploratory evidence, not an independent validation.
Use this minute to form a question about similar conditions. The later move is known in hindsight; these readings alone do not explain its cause or show that it was predictable.
Picked after the fall, on one market.
Free replay covers BTCUSDT on Binance Futures, the past 3 days. Pro reaches 30 days; Research reaches 90 days. Access is checked when you open a replay.
No account? Replay a real move tick by tick, free: TUT short squeeze · 9 Aug 2026.
Five separate snapshots before the decision, not a continuous tape. Green above zero, red below; unavailable is never zero.
| Reading | -24h | -4h | -1h | -15m | -1m |
|---|---|---|---|---|---|
| Distance from VWAPread at the minute itself · Highlight: ≤ -2.00% or ≥ +2.00% | +1.79% | +2.40% | +1.55% | +3.50% | +2.56% |
| Spot/perpetual close basisfrom matching closed one-minute perpetual and spot bars | Unavailable | Unavailable | Unavailable | Unavailable | Unavailable |
| realized-volatility rankmeasured over the trailing hour · Highlight: ≤ 10th pctl or ≥ 90th pctl | 58th pctl | 41st pctl | 79th pctl | 94th pctl | 98th pctl |
| share of the hour's variance that came from jumpsmeasured over the trailing hour | 0.00 | 0.34 | 0.14 | 0.00 | 0.01 |
| trailing 15-minute returnmeasured over the trailing 15 minutes | -0.23% | +0.23% | +0.17% | +1.5% | -0.73% |
| trailing one-hour rangemeasured over the trailing hour | 0.93% | 1.82% | 1.26% | 2.69% | 2.88% |
| trailing one-hour returnmeasured over the trailing hour | -0.46% | +1.04% | +0.11% | +2.23% | +1.20% |
| Trailing one-hour realized volatilitymeasured over the trailing hour | 0.12% | 0.10% | 0.14% | 0.16% | 0.18% |
| Reading | -24h | -4h | -1h | -15m | -1m |
|---|---|---|---|---|---|
| Big-trade rankagainst this coin's own last 14 days, at five-minute closes | 50th pctl | 50th pctl | 50th pctl | 50th pctl | 50th pctl |
| Dollar volume over 15 minutesmeasured over the trailing 15 minutes | $20K | $26K | $9K | $31K | $60K |
| Flow toxicity against this market's own historyread at the minute itself | 5th pctl | 62nd pctl | 88th pctl | 97th pctl | 96th pctl |
| Illiquidity against this market's own history |
| Reading | -24h | -4h | -1h | -15m | -1m |
|---|---|---|---|---|---|
| Funding against this market's own settled historythe venue value last published before that minute | 52nd pctl | 52nd pctl | 52nd pctl | 52nd pctl | 52nd pctl |
| Funding minus premium indexlatest funding minus the published premium index | +0.0050% | +0.0050% | +0.0050% | +0.0050% | +0.0050% |
| How much of the hour's volume became new positionsmeasured over the trailing hour | +2.36% | Unavailable | +11.3% | -2.05% | +3.70% |
| Reading | -24h | -4h | -1h | -15m | -1m |
|---|---|---|---|---|---|
| Cascade clusteringread at the minute itself · Highlight: ≥ 0.50 | 0.62 | 0.62 | 0.62 | 0.62 | 0.62 |
| Liquidation intensityread at the minute itself | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| Liquidations in the trailing hourmeasured over the trailing hour | $0 | $0 | $0 | $0 | $5 |
| Longs liquidated in the trailing hourmeasured over the trailing hour | $0 | $0 | $0 | $0 | $5 |
| Shorts liquidated in the trailing hourmeasured over the trailing hour | $0 | $0 |
| Reading | -24h | -4h | -1h | -15m | -1m |
|---|---|---|---|---|---|
| Ask pull rankagainst this coin's own last 14 days, at five-minute closes | 74th pctl | 37th pctl | 63rd pctl | 58th pctl | 89th pctl |
| Bid pull rankagainst this coin's own last 14 days, at five-minute closes | 54th pctl | 38th pctl | 65th pctl | 86th pctl | 99th pctl |
| Order-book imbalanceread at the minute itself · Highlight: ≤ -0.5 or ≥ 0.5 | 99% BID | 88% BID | 81% ASK | 91% BID | 88% ASK |
| book capture regime |
| Reading | -24h | -4h | -1h | -15m | -1m |
|---|---|---|---|---|---|
| last 15-minute candle range versus its baselinefrom the last completed 15-minute candle · Highlight: ≤ 0.50× or ≥ 2.00× | 0.92× | 1.45× | 0.47× | 2.23× | 2.23× |
| last four-hour candle range versus its baselinefrom the last completed four-hour candle | 1.13× | 1.26× | 0.98× | 0.98× | 0.98× |
| last one-hour candle range versus its baselinefrom the last completed one-hour candle | 0.68× | 1.33× | 0.89× | 0.89× | 0.89× |
| Reading | -24h | -4h | -1h | -15m | -1m |
|---|---|---|---|---|---|
| Time since resistance breakread at the minute itself · Highlight: 12 h to 720 h (inclusive) | 175 h | 195 h | 198 h | 198.8 h | 199 h |
| Resistance line spanread at the minute itself | Unavailable | Unavailable | Unavailable | Unavailable | Unavailable |
| Resistance touchesread at the minute itself | Unavailable | Unavailable | Unavailable | Unavailable | Unavailable |
| Touches before resistance breakread at the minute itself · Highlight: ≥ 3 | 4 | 4 | 4 | 4 | 4 |
| Line ATR contractionread at the minute itself | Unavailable | Unavailable | Unavailable |
Highlighted cells meet the printed fixed rule. This does not establish rarity, statistical significance or a cause of the move. Book imbalance and spread describe the best quotes; depth cost and pull describe recorded full-depth inputs where present; liquidation readings describe what the venue reported.
Exact setup countedOpen interest change rank 94th pctl or higher and Volatility rank 97th pctl or higher and Volume rank (own history) 99th pctl or higher · 1 minute before the decision timePrinted conditions only; edit them before testing.
The selected move lasts 32 hours. This comparison measures 0.5 hours after each historical match. Changing it uses the existing result.
Descriptive comparison only; it does not show that the setup works.
| Outcome | Matching setup | All eligible minutes | Difference |
|---|---|---|---|
| Finished at least 2% higher | 1 / 38 (2.6%)1 unavailable | 271 / 42,303 (0.6%)898 unavailable | +2.0 pp |
| Finished at least 2% lower | 2 / 38 (5.3%)1 unavailable | 282 / 42,303 (0.7%)898 unavailable | +4.6 pp |
| Reached +5% at any point | 0 / 38 (0.0%)1 unavailable | 8 / 42,466 (0.0%)735 unavailable | 0.0 pp |
| Reached -5% at any point | 0 / 38 (0.0%)1 unavailable | 23 / 42,466 (0.1%)735 unavailable | -0.1 pp |
Reference: every eligible minute in the same market and dates, the matching minutes included. It is not a control group. The 30-window reporting floor is not a significance test. Overlapping windows and shared market events can reduce independent evidence. Conditions were selected after seeing this move; confirmation needs unseen data.
Sector and volume lists are unavailable. Retry or explicitly choose another population.
Starting defaults: 2026-10-10 to 2026-11-09 UTC; finishes down 2% or more after 1 hour. Review or change the markets, dates, conditions and outcome before running. Current recorded classification and sector tags, not point-in-time membership. Historical membership may differ. Instrument history does not guarantee every setup input or outcome is present; actual study exclusions remain authoritative.
Choose the study population explicitly: This market only · All recorded crypto perpetuals
Read at 1 minute before the decision.
Of the 39 fires, 35 came before the move began and 2 happened while it was already under way. A fire during the move sees it; it does not telegraph it.
About 1.3 a day, on 24 of 31 days: spread through the record rather than one event.
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.
Read at 15 minutes before the decision.
11 rose and 8 fell over 19 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.
Of the 20 fires, 19 came before the move began and 1 happened while it was already under way. A fire during the move sees it; it does not telegraph it.
About 0.6 a day, on 18 of 31 days: spread through the record rather than one event.
Read at 1 hour before the decision.
After 30 minutes this setup rose 2% or more in 0 of 49 complete windows (0.0%), against 0.6% with no condition (271 of 42,303): 0.00x.
Best point inside 30 minutes reached +5% or more in 0 of 49 complete windows; worst point fell 5% or more in 0 of 49.
Both are true: these readings sat before this fall, and the other times they appeared, what followed more often rose. That disagreement is the finding: a lead-up that looks special is not the same as one that usually comes before falls.
| 70th pctl |
| 38th pctl |
| 58th pctl |
| 70th pctl |
| 18th pctl |
| Informed-trading pressureread at the minute itself | 0.24 | 0.30 | 0.32 | 0.35 | 0.35 |
|---|
| Net taker flow over the trailing four hoursmeasured over the trailing four hours | -0.10 | -0.06 | -0.11 | -0.05 | -0.09 |
|---|
| Net taker flow over the trailing hourmeasured over the trailing hour | -0.14 | +0.13 | -0.00 | +0.18 | +0.05 |
|---|
| Price impact per dollar tradedmeasured over the trailing hour | 0.63 per $1M | 0.643 per $1M | 0.705 per $1M | 0.752 per $1M | 0.543 per $1M |
|---|
| large-trade share of recent flowmeasured over the trailing 15 minutes | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
|---|
| persistence of taker flow over the hourmeasured over the trailing hour | -0.03 | -0.09 | +0.11 | -0.36 | -0.18 |
|---|
| taker-buy share of recent flowmeasured over the trailing 15 minutes | 0.51 | 0.60 | 0.47 | 0.56 | 0.40 |
|---|
| VPIN regimeread at the minute itself | normal | normal | normal | normal | normal |
|---|
| Turnover against its recent hoursmeasured over the trailing hour | Unavailable | Unavailable | 1.16× | 1.25× | 1.87× |
|---|
| Volume against this market's own historyread at the minute itself · Highlight: ≤ 10th pctl or ≥ 90th pctl | 42nd pctl | 73rd pctl | 23rd pctl | 77th pctl | 100th pctl |
|---|
| Minutes until funding settlesfrom the venue funding calendar |
|---|
| 60 |
| 300 |
| 120 |
| 75 |
| 61 |
| Open interest against the circulating floatread at the minute itself | 0.24 | 0.24 | 0.24 | 0.24 | 0.24 |
|---|
| Premium index change over 1hchange in the venue's published premium index over the hour | +0.0000% | +0.0000% | +0.0000% | +0.0000% | +0.0410% |
|---|
| Premium index rank over 30dthe venue's published premium index against its own 30 days | 64th pctl | 65th pctl | 65th pctl | 65th pctl | 65th pctl |
|---|
| Published premium indexthe venue's published premium index at that minute | +0.0000% | +0.0000% | +0.0000% | +0.0000% | +0.0000% |
|---|
| funding ratethe venue value last published before that minute | +0.0050% | +0.0050% | +0.0050% | +0.0050% | +0.0050% |
|---|
| global long ratiothe venue value last published before that minute | 66.7% | 67.4% | 67.5% | 67.2% | 67.2% |
|---|
| hourly change in open contractsmeasured over the trailing hour | +0.06% | Unavailable | +0.29% | -0.06% | +0.16% |
|---|
| long share of top-trader positionsthe venue value last published before that minute · Highlight: ≤ 35.0% or ≥ 65.0% | 85.4% | 85.6% | 85.6% | 85.6% | 85.5% |
|---|
| normalized funding ratethe venue value last published before that minute | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 |
|---|
| open-interest velocity rankread at the minute itself · Highlight: ≤ 10th pctl or ≥ 90th pctl | 61st pctl | 40th pctl | 48th pctl | 100th pctl | 95th pctl |
|---|
| Top-position long share minus global-account long sharethe venue value last published before that minute · Highlight: ≥ 10.0 PTS | 18.6 PTS | 18.3 PTS | 18.2 PTS | 18.4 PTS | 18.4 PTS |
|---|
| $0 |
| $0 |
| $0 |
| distance down to the nearest bright 50x to 100x liquidation shelfread at the minute itself | Unavailable | Unavailable | Unavailable | Unavailable | Unavailable |
|---|
| distance down to the nearest bright 5x to 25x liquidation shelfread at the minute itself | Unavailable | Unavailable | Unavailable | Unavailable | Unavailable |
|---|
| distance up to the nearest bright 50x to 100x liquidation shelfread at the minute itself | Unavailable | Unavailable | Unavailable | Unavailable | Unavailable |
|---|
| distance up to the nearest bright 5x to 25x liquidation shelfread at the minute itself | Unavailable | Unavailable | Unavailable | Unavailable | Unavailable |
|---|
| liquidation burst against this market's own historyread at the minute itself | 49th pctl | 49th pctl | 49th pctl | 49th pctl | 49th pctl |
|---|
| Which side the trailing hour of liquidations fell onmeasured over the trailing hour · Highlight: ≤ -0.50 or ≥ +0.50 | Unavailable | Unavailable | Unavailable | Unavailable | -1.00 |
|---|
| native |
| native |
| native |
| native |
| native |
| cost of buying 100k from the bookread at the minute itself | Unavailable | Unavailable | Unavailable | Unavailable | Unavailable |
|---|
| cost of selling 100k into the bookread at the minute itself | 489 bps | 345 bps | 455 bps | 351 bps | 393 bps |
|---|
| depth capture regimestamped once for the whole symbol-day from the capture era | live | live | live | live | live |
|---|
| normalized spreadread at the minute itself | 0.06 | 0.06 | 0.06 | 0.06 | 0.06 |
|---|
| order-book imbalance rankread at the minute itself | 71st pctl | 77th pctl | 60th pctl | 51st pctl | 30th pctl |
|---|
| quoted spread in basis pointsread at the minute itself | 5.79 bps | 5.70 bps | 5.72 bps | 5.61 bps | 5.65 bps |
|---|
| sharpest pull of resting asks in the minuteread at the minute itself | 0.09 | 0.03 | 0.07 | 0.06 | 0.13 |
|---|
| sharpest pull of resting bids in the minuteread at the minute itself | 0.04 | 0.03 | 0.05 | 0.11 | 0.35 |
|---|
| What takers paid to cross over 15 minutesmeasured over the trailing 15 minutes | 6.40 bps | 18.7 bps | 17.0 bps | 12.6 bps | 15.3 bps |
|---|
| Unavailable |
| Unavailable |
| Resistance line stateread at the minute itself | none | none | none | none | none |
|---|
| Distance to broken resistanceread at the minute itself | -2.51× | -3.01× | -3.00× | -3.32× | -3.20× |
|---|
| Distance to resistanceread at the minute itself | Unavailable | Unavailable | Unavailable | Unavailable | Unavailable |
|---|
| Support shaperead at the minute itself | Unavailable | Unavailable | Unavailable | Unavailable | Unavailable |
|---|
Of the 49 fires, 39 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 1.6 a day, on 24 of 31 days: spread through the record rather than one event.