Wyckoff Evidence Desk v1.0.0 · Wyckoff indicator for TradingView One range.
Which story holds up?

The classical route starts with a provisional stopping candidate (SC? or BC?). Its extreme may change until a confirmed automatic reaction (AR) supplies the other boundary. The default pivot needs three bars on each side; its label appears when confirmed, not at the earlier turning point. A subsequent qualifying lower-effort secondary test starts B.

The optional balanced-range start uses the previous 30 bars, movement efficiency below 0.30, and width between 2.5 and 10 current ATR by default. It begins in B as TR? and explicitly says Phase A was not observed. Boundaries freeze once AR or bootstrap establishes them. B must develop for at least eight bars before the relevant sweep or strength transition.

A–E are the engine’s states, not proof of accumulation, distribution or institutional intent. SCAN means no active episode; a preliminary PS?/PSY? observation does not establish a range.

No. A mature B range can move directly to D on the engine’s directional-strength conditions. The optional C route starts with a breach beyond the boundary allowance and within the maximum sweep depth. A pending SWEEP? must recover by the configured deadline; this is a real later close, not future information attached to the past.

A confirmed Spring or UTAD is classified by relative volume: type 1 at least 2, type 2 at least 1, type 3 below 1. Those are this release’s thresholds. Type 1/2 require a subsequent qualifying lower-effort test. Type 3 waives that test but still needs strength, acceptance and backup. “Test confirmed” and “Type 3 · test waived” are different observations.

A candle crossing both boundaries is ambiguous because OHLC cannot reveal its intrabar order. More severe invalidation takes priority. Distribution mirrors the price-direction conditions.

A decisive breakout meeting the directional-strength rule starts the acceptance check. Two consecutive closes beyond the boundary allowance establish ACCEPT and phase E; the original breakout candle counts as the first when the next close follows through. A close back inside the allowance resets that count; later acceptance closes do not recheck the original strength rule. ACCEPT still waits for a later return, called a backup (BU).

For the accumulation side, the later candle reaches the breakout area without exceeding allowed depth, closes at or above the range top with a bullish body, and has RVOL at most 1.3 when volume is available. Existing invalidation rules run first. Distribution reverses the price conditions.

A fresh BU is checked once against the configured gates: default score at least 65, lead over the opposing score at least 10, volume available, and higher-timeframe context not opposing. A blocked BU is consumed; later improvement does not retry it. It is one opportunity per episode, not a queued order.

CONTESTED is a display rule: both hypotheses are active and their score difference is at most 10. The entry lead input is separate. At exactly 10 points, a fresh BU can satisfy the default lead gate while the display still says CONTESTED.

Each hypothesis starts from its current phase: A 12, B 25, C 40, D 55, E 65. Additions are: test requirement satisfied +8 (including type 3 waiver), volume contraction +5, close on the supportive AVWAP side +5, aligned higher timeframe +7, supportive close location +5. Opposing higher-timeframe context subtracts 8.

Bootstrap subtracts 5; unavailable volume subtracts 15; a pending reclaim subtracts 8. Footprint can add 5 when directional delta divided by total volume exceeds 0.10. Scores are bounded to 0–100 independently. They are not calibrated probabilities, and need not sum to 100.

Adaptive RVOL compares volume with the median of earlier bars. The engine also uses prior 90th-percentile volume/spread references. This is a local baseline, not an adjustment for typical activity at each time of day. Volume can mean different things across feeds, including tick volume.

AVWAP starts at the bar when an episode is observed and accumulates typical price times available volume. It is a weighted price reference, not a volume profile or POC. A failed breakout that returns to B keeps the episode’s anchor; a full reset clears it.

The context uses completed higher-timeframe bars, with a one-bar offset and lookahead enabled. Automatic context is approximately four times the chart timeframe, mapped to a supported interval. With the context switch on, the interval must be larger than the chart. Neutral context is permitted; opposing context blocks the default setup. Turning it off removes its score adjustment and gate.

Core events commit on closed chart bars. Pivot confirmation is delayed. Revised feeds, changed inputs or different loaded history can change a recalculated path. Footprint has additional historical-data limitations; absolute non-repainting is not claimed.

Before acceptance, an episode expires when its age exceeds the configured maximum. Phase E does not expire from that age setting; it ends through structural failure. The original Thai manual’s general phase-E expiry wording should be read with this source distinction.

Too many consecutive adverse closes or a wick deeper than the maximum allowed excursion fully reset an episode, even if a deep wick recovers. A failed phase-D breakout can return to B and restart its development clock within the same episode. Phase E can fail on a close through its breakout boundary or its previously established structural trail.

The trail only tightens. Its plot is not a broker stop. ATR-based boundaries can move as ATR changes; decisions use the closed bar. Ended events remain in the short event tape, not in an unlimited permanent history.

The three named conditions are Qualified bullish backup, Qualified bearish backup and Confirmed reclaim. The first two mean a new setup passed its gates; a reclaim alone does not. Create alerts with closed-bar frequency and recreate them after changing source or inputs.

Emit JSON event alerts starts off. With it enabled, create Any alert() function call to receive raw events, including INVALID. The payload includes side, phase, event, evidence and volume availability. Its barTime is the candle’s opening timestamp; the event is confirmed at its close. Only the final event per side on that bar is emitted.

This version has no Setup ID export, frozen trade-plan journal or full-history guarantee. Labels are bounded; the event tape keeps six latest events across both sides, in exchange time. Strategy order-fill alerts are a separate stream.

Base has 26 inputs, Footprint 27 and Lab 31. Shared defaults are identical. The edition label identifies the extra controls.

The Lab submits a market entry and fixed bracket after a qualifying BU while flat. The order normally fills at the next available open. Default quantity is one unit, not fixed cash risk. Stop distance is at least one signal-bar ATR; the target is two times the risk measured from the signal close.

Its bracket remains tied to that signal and does not follow a new episode. Gaps and slippage can change actual risk. The 100-bar holding limit counts from the filled entry. The date window filters the signal candle’s opening time, so a later fill can fall outside it. Market-close instructions execute on a later tick.

The default experiment includes 0.05% commission per order and two ticks of slippage. Set costs, quantity and Properties for the instrument. Keep the shipped recalculation switches off when comparing with the supplied results.

Initial capitalStarting account balance.

100,000

Additional same-side entriesNo additions; Lab enters only while flat.

0

Default sizing methodThe explicit Fixed contracts / units input controls each entry.

Fixed

Default order sizeOverridden by the entry’s quantity input.

1

Commission methodPercentage of order value.

Percentage

CommissionGeneric development assumption; adjust to the instrument.

0.05% per order

SlippageGeneric emulator adjustment, not observed execution cost.

2 ticks

Fill on the closing tickMarket instructions normally process at the next candle’s open.

Off

Recalculate every realtime tickKeep the shipped closed-bar experiment.

Off

Recalculate after a fillNo extra calculation immediately after fills.

Off

Recalculate historical ticksEnabling intrabar calculation changes the experiment.

Off

Long margin requirementFull position value required as margin.

100%

Short margin requirementFull position value required as margin.

100%

The producer’s three native Pine probe suites report 889 assertions over 25 synthetic scenarios and 129 frames. These exercise the embedded engine, not 889 trades or independent market samples. The website import checks the supplied evidence and rebuilds sources/probes; it does not claim a new run of those 889 assertions.

All three served scripts were freshly compiled with TradingView: zero errors and warnings. Compilation does not establish runtime data availability. The producer’s Basic account could not run Footprint, so its runtime remains unverified in that account despite successful server compilation.

The initial Strategy Lab sample lost money. Eight closed trades totalled −$446.03 under its stated development settings and costs. It is a small functional smoke check, not evidence of a profitable strategy or performance across markets.

Website verification and limits · Original verification record · Fresh compiler evidence