AB ReverasalLab

How it works

Single-signal reversal tools fail because reversals are multi-condition events. ReversalLab scores a confluence stack and only prints above threshold:

  Liquidity sweep of a scored pool — the fuel event. This condition is mandatory.

  Statistical momentum divergence — price extreme versus momentum vector, t-stat filtered.

  Volume climax or absorption at the extreme, from the Effort-vs-Result module.

  Displacement confirmation — a break of the most recent minor structure with a displacement-grade candle.

  HTF location — bonus weight if the extreme lands inside a higher-timeframe supply/demand zone or ±2-sigma VWAP band.

Each reversal prints with a 0–100 conviction score, a component breakdown, and an objective invalidation level beyond the sweep. Because the sweep is mandatory, ReversalLab structurally cannot signal in the middle of a trend leg.

Marketplace description

Most reversal indicators mark every pin bar and call it an edge. ReversalLab treats a reversal as a process and demands evidence at every step: a liquidity sweep at a mapped pool (mandatory — no sweep, no signal), statistically-significant momentum divergence, climax or absorption volume at the extreme, and a displacement break of minor structure to confirm. Every signal carries a 0–100 conviction score, a component breakdown showing exactly which conditions fired, and an objective invalidation level beyond the sweep. It cannot, by design, signal in the middle of a trend leg. Staged alerts run from “sweep detected” to “reversal confirmed.” Non-repainting, EA-ready.


More from author
TrendFlow CBL Pro is a complete trend-following indicator built on the Count Back Line (CBL) method — a proven technique for identifying dynamic support and resistance levels based on actual price structure, not lagging averages. It does not just give you an arrow — it gives you a full trading framework : entry signal, stop loss placement, take profit target, phase-colored bars, filter confirmation, and a live performance dashboard — all in one indicator, right on your chart.  Why TrendFlow CBL
How it works QuantScore is the suite's flagship: a multi-factor model compressing six independent measurements into one −100…+100 edge score. –   Momentum factor — the regression t-stat from the VectorMomentum core. –   Mean-reversion factor — z-score of price versus its volume-weighted mean, contributing against stretch when the regime is mean-reverting. –   Regime factor — the RegimeLens state sets the weighting between momentum and reversion. This dynamic weighting is the secret sauce; static
How it works –   Multi-anchor engine — Session, Weekly, Monthly and Quarterly VWAPs simultaneously, plus auto-anchored VWAPs from the most recent significant swing high and low, plus unlimited manual click-to-anchor points (news, earnings, breakouts). –   True sigma bands — ±1/±2/±3 sigma computed from the volume-weighted variance of price around VWAP, not a lazy standard deviation of closes. Bands expand realistically with participation. –   Interaction signals — first-touch pullback to session
How it works –   Strict validation chain — a candidate becomes an order block only if all hold: it is the last opposing candle before a displacement move; the displacement breaks structure; the move leaves an imbalance (FVG) behind; and displacement volume exceeds its time-of-day baseline. This removes roughly 90% of what other indicators mislabel as order blocks. –   Grading (A/B/C) — from displacement magnitude, liquidity swept by the block before departing, and higher-timeframe alignment. –  
How it works –   Impulse-leg detection — identifies displacement legs (velocity + range expansion + structural break) worth continuing with. –   Zone projection — for each impulse, projects the confluence-cluster pullback zone: the overlap of the 38–62% retracement band, the breakout origin (old resistance/new support), any fair value gap inside the leg, and the auto-anchored VWAP from the leg's origin. Where two or more overlap, a graded continuation zone prints. –   Regime gate — zones only pr
How it works –   Detection with quality filters — true three-candle imbalances filtered by minimum gap size (ATR-scaled), displacement quality of the middle candle and volume, killing the micro-gap spam that makes FVG tools unusable on low timeframes. –   Fill-state tracking — each gap tracked live: Untouched, Partially filled (with fill %), Fully mitigated, Inverted. –   The headline feature — Inversion (failed FVG): when price closes entirely through a gap, the imbalance inverts. Failed bullis
How it works –   Pool detection — equal highs/lows (within an ATR-scaled tolerance), unswept swing highs/lows, session and daily/weekly extremes, and trendline liquidity (3+ touch diagonal clusters where stops accumulate). –   Pool weighting — each pool scored by touch count, age, round-number proximity and origin timeframe, then drawn with thickness/opacity proportional to score. The chart literally shows where the fuel is. –   Sweep engine — a sweep registers when price wicks through a pool an
How it works An inducement trap is a sequence, not a candle pattern. TrapRadar's state machine requires all four legs: –   Bait — an obvious level forms (minor swing, equal highs/lows, range boundary) that retail will trade the breakout of. –   Trigger — price breaks the level and actually attracts participation; relative volume rises on the breakout bar, confirming traders took the bait. –   Trap spring — within K bars a displacement candle (range greater than 1.5× ATR, closing 70%+ through its
How it works –   Base-departure detection — algorithmically finds consolidation bases (overlapping range with compressed volatility) followed by displacement departures. Patterns: Drop-Base-Rally, Rally-Base-Drop, Rally-Base-Rally, Drop-Base-Drop — detected structurally, not by candle names. –   Strength score (0–100) — from departure velocity, time-at-base, freshness (each revisit decays the score), higher-timeframe confluence and origin volume. –   Lifecycle — Fresh, Tested (decaying), Broken
FREE
ATR tells you yesterday's range. VolCore models the volatility cycle itself: –   Estimator stack — Parkinson, Garman-Klass and Yang-Zhang realized volatility (OHLC-based estimators are 5–8× more efficient than close-to-close). Select one or blend. –   Percentile ranking — current volatility ranked within its own 250-bar distribution, giving a universal 0–100 Vol Rank comparable across all symbols. –   Squeeze detection — Vol Rank below a threshold and still compressing (negative vol-of-vol slope
How it works No RSI, no stochastic. VectorMomentum measures momentum three independent ways and demands agreement: –   Regression-slope t-statistic — a linear regression on log price over N bars; the t-stat of the slope (slope divided by its standard error) gives statistically-significant momentum that self-normalizes across symbols and volatility levels. –   Volume-pressure-weighted rate of change — ROC weighted by relative tick volume, so momentum on dead volume scores lower than the same move
Filter:
No reviews
Reply to review