Binance USDT-MToxic flow elevated
Historical comparison not prepared for this market.
Binance USDT-MMove
Price 6.206 · Quote 12:00 UTC · -1.46% 24h · Analysis 1 min behind
Latest analysis · 11:59 UTC
Price falling (15m) · OI rising (1h)
7 of 7 key readings recorded · 2 proposed as conditions
- Return (15m)-0.35%
- Taker buy share (15m)0.39More aggressive selling
- Open interest change (1h)+0.09%Side not identified
- Funding rate (latest)+0.0050%Longs pay shorts
- Shorts liquidated (1h)$0Reported forced buying
- Longs liquidated (1h)$0Reported forced selling
- Book imbalance (top levels)proposed98% ASKasks outweigh bids in the top levels
- VWAP distance (day)proposed-2.67%
These readings describe the minute, not what followed.
Historical evidence and records
Historical comparisons are not prepared for this market. These cached comparisons do not change with the selected minute.
Market chart, markers and liquidation layers30 days · 4-hour candles
Sep 03Sep 11Sep 18Sep 26Oct 03UTC
30 days · Binance 4H candles · diamonds replay the exact moment a rule firedUNLOCK 32 OLDER MARKERS
> What the shaded rows are
Shaded rows mark prices where leveraged positions would be forced to close if price gets there, built from this market's own price and volume. A map of fuel, not a prediction: it says nothing about whether price will go there.
MAP TO 11:55 UTC · lf.v2
Across $1.45B of real liquidations, the brightest tenth of this map captured about 22% of the dollars - roughly twice what nearness to price alone explains.
Squares open archived events; diamonds replay research detections. Research markers checked through 2026-10-03 11:59 UTC; missing readings remain gaps.
What followed the tracked conditions1 active · 6 of 6 readable at 2026-10-03 11:59 UTC
Observed 2 minutes ago. Snapshot for 2026-10-03 11:59 UTC. It does not describe price moves after that minute.
PROMUSDT: Price swept the 12-hour lows and reclaimed them.
Six defined conditions, not overall volatility. Inspect a minute to explore other readings.
Historical comparisons are not prepared. This does not mean zero matches.
Measured over the minute ending 2026-10-03 11:59 UTC. Coverage and methodology
HISTORICAL PROOF
What followed this setup before
Historical comparison not prepared. Open the Workbench to define and review a study for this market.
Readings in this snapshot
What stands out in the latest record
No reading is far from its usual range in the latest record.
Continue from this record
Investigate or monitor a defined setup
Free with an account. The live read is this minute; this page reads the record.
Older detections by weekPro
Browse older moments
Choose a recorded week. Markers are detections, not a complete history of every minute.
2026-09-27 to 2026-10-03 UTC
Historical readings and tick replay have different coverage. Replay still checks your plan and the available recording. Choose an exact move to investigate.
Full evidencereadings, conditions, research, events
State · latest recorded minute
What the latest PROMUSDT record shows
50 of 69 readings present; each absent one states its reason. P82 = higher than on 82% of this market's recorded minutes. More: percentiles
What stands out in the latest record0 OF 50 READINGS
Nothing stands out in the latest record: every present reading is close to its usual level.
A reading stands out when it sits in the top or bottom tenth of its own record, or an uncommon state is on.
Free replay covers the latest fully archived UTC day on 6 major markets. Pro reaches 90 days; Research reaches available recorded replay history. Access is checked when you open a replay. Each opens 30 minutes before the moment.
+ Full market read (69 readings)- Full market read (69 readings)
Flow10 READINGS
Toxic flow (VPIN)0.36·
VPIN regimeNORMAL·
Taker buy share · 15m0.39·
Dollar volume · 15m$39K·
Volume rank · own history30th pctl·
VPIN rank · own history31st pctl·
Amihud illiquidity · 1h0.188 per $1M·
Illiquidity rank · own history80th pctl·
Jump share · 1h0.06·
Flow persistence · 1h-0.03·
Large-trade share · 15m0.00·
Net taker flow · 1h+0.07·
Net taker flow · 4h-0.16·
Return · 15m-0.35%·
Return · 1h-0.53%·
Range · 1h1.08%·
Realized vol · 1h0.05%·
Realized vol percentile13th pctlP13
VWAP distance · day-2.67%·
Book3 READINGS
Book imbalance · top levels98% ASK·
Book imbalance percentile63rd pctlP63
Spread · normalized0.02·
Spread · bps1.61 bps·
Effective spread · 15m5.15 bps·
Ask pull · 1mABSENTNO READING THIS MINUTE
Bid pull · 1mABSENTNO READING THIS MINUTE
Cost to buy 100k · bpsABSENTNO READING THIS MINUTE
Cost to sell 100k · bpsABSENTNO READING THIS MINUTE
Book captureABSENTNO READING THIS MINUTE
Depth captureABSENTNO READING THIS MINUTE
Derivatives10 READINGS
Published premium indexABSENTNO READING THIS MINUTE
Premium index change over 1hABSENTNO READING THIS MINUTE
Premium index rank over 30dABSENTNO READING THIS MINUTE
Funding minus premium indexABSENTNO READING THIS MINUTE
Spot/perpetual close basisABSENTNO READING THIS MINUTE
Funding rate · latest observed+0.0050%·
Funding · normalized0.01·
Minutes to funding241SETTLES 16:00 UTC
Open interest change · 1h+0.087%·
OI change magnitude score52nd pctlP52
Liquidation intensity0.00·
Liq notional · 1h$0·
Longs liquidated · 1h$0·
Shorts liquidated · 1h$0·
Liquidation side skew · 1hABSENTNO READING THIS MINUTE
Cascade clustering0.47·
Top-trader long share73.7%·
Global long share31.1%·
Top vs crowd skew42.6 PTS·
Funding rank · own historyABSENTNO READING THIS MINUTE
OI conversion · 1hABSENTNO READING THIS MINUTE
OI vs circulating floatABSENTNO READING THIS MINUTE
Liquidation burst rank48th pctl·
Short shelf above · 50-100x · %ABSENTNO READING THIS MINUTE
Long shelf below · 50-100x · %ABSENTNO READING THIS MINUTE
Short shelf above · 5-25x · %ABSENTNO READING THIS MINUTE
Long shelf below · 5-25x · %ABSENTNO READING THIS MINUTE
An absent reading shrinks every count it touches; it is never treated as zero.
Candles10 READINGS
Body share · 15m-77.8%·
Body share · 1h+46.7%·
Upper wick · 15m0%·
Lower wick · 15m22.2%·
Compression · 15m0.58×·
Compression · 1h1.10×·
Compression · 4h0.70×·
Engulfing · 15mNONE·
Engulfing · 1hBULLISH·
Sweep + reclaim · 15mFALSE·
Sweep + reclaim · 1hTRUE·
Containment · 15mNONE·
Candle readings use the last completed candle, never the one forming. No completed candle: shown as absent, never guessed.
SNAPSHOT KEY · v2 · nrm.v1 · ft.v1 · rev b13ca5a6 · q#ed254833Same key, same bytes: the key is the re-run proof.> RUN THIS READ OVER THE API> WHAT THESE READINGS MEASURE
What followed · counted over the rolling window
All tracked conditions
1 of six tracked conditions is active on PROMUSDT now. Open any condition to see its exact definition and recorded history.
COUNTING WINDOW: 2026-08-07 TO 2026-09-06 (30 DAYS). DATA COVERAGE BEGINS 2025-07-15.
One-sided tape: buyers
Historical comparison not prepared for this market.
One-sided tape: sellers
Historical comparison not prepared for this market.
Large USD OI change with liquidations
Historical comparison not prepared for this market.
Crowd long, leaders not
Historical comparison not prepared for this market.
More setups, counted nightly
Beyond the six above, 11 more setups are counted nightly on PROMUSDT. 2 of the 11 measurable at the latest recorded minute are active:
- Order book heavily ask-sidedACTIVE IN LATEST RECORD WHAT FOLLOWEDPRO
- Hourly engulfing candleACTIVE IN LATEST RECORD WHAT FOLLOWEDPRO
Pro reveals for each of these what the six above already show: how often it has happened on this market, out of how many chances, and what followed each time. This page stays free; the deeper record is the paid part, and watching the page can never rebuild a past minute.
> SEE PRO PLANSResearch history · PROMUSDT
Published research and archived events
> Investigate the latest PROMUSDT moveMACHINE-READABLE INDEX: GET /api/research/reports
Investigated moves on PROMUSDT3
Published reports touching PROMUSDT9
On record: 5,951 OF 9,090 minutes matched (65.5%)Since publication: 339 OF 498 (68.1%)
On record: 84 OF 699 minutes matched (12.0%)Since publication: 11 OF 84 (13.1%)
On record: 287 OF 608 minutes matched (47.2%)Since publication: 14 OF 45 (31.1%), 1 still open at 1H
On record: 6,402 OF 60,275 minutes matched (10.6%)Since publication: 300 OF 2,687 (11.2%), 36 still open at 1H
On record: 884 OF 7,993 minutes matched (11.1%)Since publication: 704 OF 6,489 (10.8%), 114 still open at 1H
On record: 54 OF 105 minutes matched (51.4%)Since publication: 33 OF 83 (39.8%), 1 still open at 1H
On record: 23 OF 282 minutes matched (8.2%)Since publication: 27 OF 215 (12.6%)
On record: 823 OF 1,284 minutes matched (64.1%)Since publication: 951 OF 1,407 (67.6%)
On record: 45 OF 82 minutes matched (54.9%)Since publication: 1,053 OF 1,567 (67.2%)
Observed examples in a selected archive, not a guarantee. Nothing here is trading advice.
Archived events · PROMUSDT8
Terminal · API · MCPreproducible calls
Re-run this record
Fetch the recorded state with one call. Its reproducibility key returns the same canonical bytes later.
Cached re-runs do not spend research credits.
# the state block
GET /api/v1/research/snapshot?symbol=PROMUSDT&t=last-complete
→ canonical bytes · q#ed254833
GET /api/v1/research/snapshot?symbol=PROMUSDT&t=last-complete
→ canonical bytes · q#ed254833