SECTION 1 — FEATURES

Everything the product is, module by module, parameter by parameter.

1. Introduction & Philosophy

1.1 What is Liquidity & Volume Nexus?

Liquidity & Volume Nexus is a professional cTrader overlay indicator that unifies three institutional trading disciplines into one coherent system — and then does what almost no indicator dares to do: it measures its own signals and shows you the verdict.

• Smart-Money Structure — fractal swings, Break of Structure (BOS), Change of Character (CHoCH), breaker blocks, fair value gaps and their inverses, premium/discount ranges with the Optimal Trade Entry zone.
• Liquidity Mapping — Equal Highs / Equal Lows clusters, single-swing Buy-Side / Sell-Side liquidity, sweep detection with the raid's exact reach, prior-day and prior-week extremes — the resting orders every session trades against.
• Volume Intelligence — a time-of-day-aware volume baseline, a per-zone Volume Score with exponential decay, a Volume Squeeze detector and a full session Volume Profile with POC / VAH / VAL.

On top of that map sits a signal engine: a Nexus Signal fires only when a strong volume-backed zone, a structural confirmation (or a freshly swept liquidity pool), a cost-aware trade plan, the session clock and the volatility regime all agree. Each signal becomes a projected trade plan (Entry / Stop / TP1 / TP2 / TP3), a corner card with a grade and trade advice, and an entry in a journal that records what actually happened — in R, on your broker feed, with profit factor, drawdown and a statistical-significance verdict.

1.2 The philosophy in one paragraph

1.3 Feature summary at a glance

2. Architecture & Guarantees

2.1 The closed-bar pipeline (why it never repaints)

cTrader calls an indicator's Calculate() on every tick, and the bar it passes is the one still forming. Nexus never lets an unfinished bar touch its engines: the whole pipeline (swings → structure → zones → liquidity → volume → signals → journal) consumes the bar that has just closed. The forming bar is used only for the live panel read-out, the provisional swing preview (explicitly marked, off by default) and the intrabar life-cycle watch (so a stop taken mid-bar is reported mid-bar — and latched: a wick that retracts can never "un-stop" a trade).

This was verified tick-by-tick: running the same series once as a live tick stream and once as closed bars produces identical journals — the same signals, the same fills, the same outcomes. What you backtest by eye on history is exactly what the live edge will do.

2.2 The guarantees

3. Swing Detection (Module 1.1)

Swings are the atoms of the whole system: structure breaks, liquidity pools and the dealing range are all built from them. A fractal swing is a bar whose high (or low) is strictly beyond its Swing Period neighbours on both sides.

• Strict publication. A swing at bar C is published only after bar C + Legs has closed. Until then it does not exist for any module — this is the anti-repaint contract.
• Tick epsilon. A neighbour must sit at least half a tick beyond the centre to count as lower/higher. Exact ties therefore cancel the swing — deliberately: equal highs/lows belong to the Liquidity Engine (Module 2), not to the swing list, so the same level is never owned twice.
• Markers. A confirmed swing high draws a filled red ▼ above the bar; a swing low draws a filled green ▲ below it (colours configurable). Optional price labels print the exact level.
• Provisional preview (off by default): a hollow △/▽ glyph on the forming bar shows a swing that would confirm if the remaining bars cooperate. It is repainting by definition, is drawn in its own grey colour, never influences any module, and vanishes the moment it becomes invalid — a confirmed swing never carries two glyphs.

4. Market Structure — BOS / CHoCH / Breakers (Module 1.2)

The Structure Engine tracks the active (last unbroken) swing high and swing low — the two levels the market is currently "answering to" — and publishes an event the moment one of them breaks:

• Break Confirmation parameter: Close (default — the bar must close beyond the level; fewer, stronger events) or Wick (any trade beyond the level; faster, noisier — one bar can then publish two events when it takes out both sides).
• Synthesised context. After every event the engine holds a context reading — BULLISH / BEARISH / RANGE — shown in the badge, the panel, and used by the signal engine (the reversal setup is a Range-context specialist). Before the first event the context is RANGE.
• Active BSL / SSL lines. Two dotted silver horizontal lines mark the current unbroken swing high (buy-side liquidity resting above it) and swing low (sell-side below). When a line breaks, the engine falls back to the previous unbroken swing of that side — the levels walk backwards through history honestly.
• Breaker blocks. When a swing breaks, the engine searches back (window: Breaker Search Window) for the last opposite candle before the broken swing — the down-candle before a broken high (bullish breaker, drawn lime), the up-candle before a broken low (bearish breaker, red). That candle's full range becomes a zone: institutions that were trapped there often defend it on the retest. Near-duplicate breakers (80%+ overlap of the smaller range) are collapsed into one. Breakers age out after Breaker Max Age bars or when fully mitigated.
• Swing separation. A liquidity sweep (Module 2, wick-based by default) never marks a swing as structurally broken (close-based by default) — so a wick raid below a low does not blind the engine to the real close-break that follows. This separation is what makes the BOS/CHoCH stream trustworthy on raid-heavy charts.

5. Fair Value Gaps, Inverse FVG, Mitigation (Module 1.3)

A Fair Value Gap is a three-bar imbalance: a bullish FVG exists when bar N's low trades strictly above bar N−2's high (the market jumped and left an unfilled range between); a bearish FVG is the mirror. The gap is drawn as a rectangle from the moment it forms, and from then on it lives a measurable life:

• Size filter. Gaps smaller than FVG Min Size (x ATR) (default 0.25) or a points minimum are ignored — sub-noise gaps never become zones.
• Partial mitigation. Every trade inside the gap updates its MitigationRatio (0→1). The fill's transparency follows the zone's Volume Score; a mitigated zone fades to a third of its alpha and its rectangle stops extending at the mitigation time (plus a few bars).
• Full mitigation → Inverse FVG. When the bar closes through the far edge of the gap, the gap is consumed and — with Enable Inverse FVG on — an inverse zone (orange) is spawned over the same range with the flipped polarity: a fully filled bullish gap becomes resistance, a filled bearish gap becomes support. Inverses never flip again — no infinite ping-pong.
• Lifetime. Unmitigated gaps deactivate after FVG Max Age bars; mitigated ones linger for Keep Mitigated Zones bars (for context) unless Hide Mitigated Zones removes them from sight. Deactivation is state, not deletion: an aged zone stops scoring and can no longer produce signals.
• Polarity. Bullish zones (bullish FVG, inverse of a bearish FVG, bullish breaker, EQL, SSL) are candidate support for longs; the mirror set serves shorts. The signal engine only ever reacts at a zone whose polarity matches the trade direction.

6. Premium / Discount / OTE (Module 1.4)

The dealing range is the last confirmed swing high / swing low pair — the battlefield institutions are currently working within. The module paints its upper half as Premium (red tint — expensive; favour selling) and its lower half as Discount (green tint — cheap; favour buying), with three dotted reference lines:

• RANGE HIGH / RANGE LOW — the range boundaries, labelled with prices.
• EQ 50% — equilibrium: the fair middle. Above it, longs are chasing; below it, shorts are chasing.
• OTE 62 / 70.5 / 79 — the Optimal Trade Entry band (yellow): the 62%–79% retracement of the range measured from the top, where a continuation long ideally enters on a pullback. Shorts mirror it: their OTE is the same band measured from the bottom (the module draws the retracement of the current range direction).

The range redraws only when a new swing confirms or on the scheduled refresh — it is cheap, stable and never flickers. With Show Premium / Discount Range off, the whole module (including its eight chart objects) is removed cleanly.

7. Liquidity Engine & Sweeps (Module 2)

Liquidity is where resting orders (stop-losses, breakout buys) sit. The engine maintains two families of levels:

7.1 Equal Highs / Equal Lows (EQH / EQL)

• Confirmed swings within Liquidity Lookback bars are clustered when their prices agree inside a tolerance (points or ATR-based). A cluster of ≥ Minimum Touches (default 2) swings becomes a pool: orange line for EQH (buy-side liquidity above), aqua for EQL (sell-side below).
• The label shows touches ("EQH x3") and the zone's Volume Score. The more touches, the thicker/denser the line style — a triple-touch EQH is a magnet with a lot of stops behind it.
• Clusters refresh in place with stable object names: a level that gains a third touch updates instead of redrawing somewhere else. A swing whose liquidity was already taken can never seed a new cluster — no phantom "fresh" levels out of consumed swings.

7.2 Single-swing BSL / SSL

• The last unbroken swing high carries a BSL line (gold — buy-side resting above), the last unbroken swing low an SSL line (turquoise — sell-side below). They follow the newest confirmed swing automatically and live until swept — age never kills them, because "the market hasn't traded there for 300 bars" does not remove the orders resting there.

7.3 Sweeps — the event everything revolves around

• A pool is swept when price trades beyond its level: by wick (Sweep Confirmation = Wick, default — raids are wicks) or by close. The engine records the sweep bar, the exact raid reach (that bar's extreme — the future reversal stop anchor), marks the source swings as swept (but not structurally broken), and repaints the level grey + dotted with a "SWEPT" label.
• Swept pools stay visible for Keep Swept Pools bars (default 60; 0 hides them at once) — they are listed on the panel with a "*" marker, because a freshly swept level is exactly what the Reversal setup fades.
• A sweep of buy-side liquidity above the highs is, in smart-money reading, fuel for a move down (stops of shorts were harvested, breakout buyers are trapped); a sweep of sell-side below the lows is fuel for a move up. The signal engine turns that reading into an actual setup — see Module 4.1.

8. Prior Day / Prior Week Levels (Module 2.5)

Four fixed lines every session trader needs: PDH / PDL (previous completed UTC day's high and low, solid) and PWH / PWL (previous completed ISO week's extremes, dotted, slightly transparent). Labels carry the prices; the panel prints all four with their signed distance from the live price in pips.

• Immutable by construction. Levels are recomputed only when the closed bar crosses into a new day/week, and the scan never includes the current period — today's spike can never leak into "yesterday's high". They cannot repaint, period.
• These are the highest-probability liquidity references on the chart: algos and desk traders key entries, stops and targets to PDH/PDL daily. A wick through PDH that closes back below it is one of the most reliable intraday fade patterns there is — and Nexus detects exactly that shape as a reversal signal when the raid coincides with a tracked pool.

9. Volume Nexus — Baseline & Zone Score (Module 3)

9.1 The honest yardstick: time-of-day baseline

cAlgo exposes tick volume only, and tick volume is strongly seasonal: the London open prints several times the volume of the late New York session. Comparing a bar against a plain rolling average therefore calls everything an anomaly at 08:00 and nothing at 20:00. Nexus maintains a per-time-bucket average (bucket size: Session Bucket Size, default 60 min) over the loaded history and blends it with the rolling average:

A bar never inflates its own yardstick (the baseline for bar N is built strictly from bars before N), and Baseline History bounds the sliding window (5000 bars default; 0 = all loaded history). The panel prints all three numbers side by side — bar volume, baseline with its %, the rolling average and the same-hour average — so you can always see which question is being asked.

9.2 Zone Volume Score (0–100)

Every active zone accumulates the volume of the bars that trade through it. The score is (average per-bar volume inside the zone) / (current baseline) × 50, clamped to 0–100: 50 means the zone trades exactly at the market's normal rate; 100 means twice the rate or more. With Zone Score Half-Life (default 200 bars) contributions decay exponentially, so a zone is scored on what it has been doing lately, not on one busy bar from three weeks ago.

The score label is printed inside each zone (optional) and the fill alpha scales with it: bright zones are busy zones.

10. Volume Squeeze Detector (Module 3.4)

A squeeze is a stretch of unusually quiet volume — the market coiling before expansion. The detector counts consecutive closed bars whose volume sits below Squeeze: Volume Below (% of the time-of-day baseline, default 60%). After Squeeze: Minimum Bars (default 5) such bars — and, when Require Zone Nearby is on, only within Zone Proximity pips of an active zone — the state fires once: a yellow SQZ marker appears under the bar, the badge and panel read "SQUEEZE", and (optionally) an alert is raised. The counter resets the first time volume returns to normal.

11. Volume Profile (Module 3.5)

A price-binned histogram of the selected period, drawn to the right of the last bar:

• Range modes. Last Bars (rolling N bars, default 200) or true calendar periods — Current Day / Current Week / Current Month in UTC (not rolling 24h/7d windows — a daily profile never mixes two trading days).
• POC (Point of Control, yellow line) — the price level with the most volume: the period's fair price and a magnet whenever price strays.
• VAH / VAL (Value Area High/Low, aqua) — the band holding Value Area% (default 70) of the period's volume around the POC. Inside the VA the market is "accepting" price; outside it is exploring, and edges of the VA become support/resistance.
• Each bar's volume is spread evenly across the bins its range covers; bins under 4% of the maximum are not drawn (the histogram stays a calm background column, not a neon block). The panel prints POC/VAH/VAL, the price's signed distance from the POC in pips, and whether price is inside or outside the VA.

12. Nexus Signal Engine (Module 4.1)

The Nexus Signal is the product's answer to one question: "is this setup worth taking right now, at this cost, in this session?" It fires on a closed bar, at most once per cooldown window, and only after passing a gauntlet of gates. Two professional setups exist; every signal is labelled with the one that fired.

12.1 The two setups (Signal Direction Mode)

Signal Direction Mode chooses what may fire: Continuation (default), Reversal, or Both (the fade is tried first — it is the more specific event — and every signal carries its setup label on the card, in the panel, the journal and the CSV).

12.2 The gate gauntlet (in evaluation order)

Every gate below must pass. Each rejection is counted by reason and shown on the panel's Rejections line.

12.3 Entry-location modes (where the close must sit)

Liquidity pools are treated as levels, not areas: their band is a fraction of the equal-level tolerance, so they are judged by a fixed 5-pip direction tolerance instead of the entry mode — a thin pool is never asked for a sub-pip close.

12.4 What a fired signal produces

1. A NEXUS marker on the signal bar (▲ lime below the bar for a long, ▼ red above for a short) plus a coloured vertical line.
2. A trade plan (Module 4.5): entry, structural stop, TP1/TP2/TP3 — validated so the stop is always on the risk side and the targets on the reward side, or the plan does not exist at all.
3. The signal card (Module 4.4) with the grade, gates, setup label and the ADVICE block.
4. A journal entry (Module 4.6) with the full plan and the live/replay flag.
5. Alerts: a sound and an [LVN ALERT] line in the cTrader log — live signals only, never during the history replay.

13. The Signal Card & Grades (Module 4.4)

The card is a corner-locked static text block (bottom-left by default) that mirrors the most recent accepted signal — it is present right after you attach the indicator, follows its plan while the plan lives, then collapses into a compact RESULT card reporting the outcome. It is the product's dashboard at the moment of decision.

13.1 Grades — a checklist, not a prophecy

The four gates on the card: (1) zone score above the signal floor; (2) structure confirmation — a fresh matching BOS/CHoCH for continuations, or the SFP close itself for reversals (a reversal is never punished for lacking the very break that contradicts it); (3) the volume anomaly, judged on the same time-of-day baseline the engine used; (4) HTF context not opposing. Below the grade the card prints the grade's own track record on your feed — Track A: 12 closed, avg +0.31R, win 58% — so the letter is never read without its history.

13.2 Card modes & lifetime

• Lifecycle (default): the card stays while the plan is actionable, carrying a live STATUS LIVE n bars left (or STATUS LIMIT PENDING @ price for an unfilled limit), then becomes the RESULT card for the configured hold.
• Timer: the classic auto-hide countdown.
• Always: never hides (not recommended — a dead plan keeps showing numbers).
• A card restored from the history replay after attach / timeframe switch is tagged [history] and prints its true age ("95 bars ago") — an old outcome can never masquerade as breaking news. Countdowns arm on the first live tick, so a deep replay can never swallow the card before you see it.

14. Trade Plan Projection & Life-Cycle (Module 4.5)

14.1 The projection

From the signal bar the plan is drawn on the chart: a solid silver ENTRY line, a dotted red STOP line (with its distance in pips), TP1 (solid-ish), TP2 and a dimmer TP3 (each labelled with price and R multiple), plus translucent risk and reward boxes and a header with direction, grade and capped R:R ("1 : 5+" above the display cap — the raw value lives in the journal). The projection lives for Trade Plan Horizon bars (default 60) and is then removed automatically — no orphans, no manual cleanup.

14.2 Stop & target construction

• Stop: for a long — the deepest of {zone bottom, active swing low, and for reversals the sweep extreme and the signal bar's low}, minus a 2-tick buffer; mirrored for shorts. If no structural reference exists on the risk side, the plan declares itself non-structural, reports R:R = unknown (0) and is rejected by the gates rather than printed with an absurd ratio.
• Targets: the nearest and second- and third-nearest live (unswept) opposite-side liquidity pools, capped at Max Target Distance (x ATR) — a pool 300 pips away is real liquidity but not a first objective. Beyond the cap, volatility-scaled fallbacks take over (2× / 4× / 6× ATR), then labelled pip fallbacks. Every rung says what it is made of ("EQH x2 pool", "2x TP1", "fallback 4x ATR").

14.3 The life-cycle — six objective rules, first hit wins, latched

The life-cycle also advances intrabar: a stop taken mid-bar is reported the moment it happens, and every outcome is latched — a wick that retracts cannot un-stop a trade. While a position is open, the plan tracks MFE / MAE (maximum favourable / adverse excursion in R from the actual fill), including the death bar of every exit type.

14.4 Market vs limit entries

• Market at close (default): the entry is the signal bar's close — filled by definition.
• Limit at zone EQ: the plan bids/asks the zone equilibrium for a pullback entry. The whole plan (stop, targets, R:R, size) is measured from the limit price. If the equilibrium sits on the wrong side of the close, the plan silently falls back to market with an honest note. While unfilled the card reads LIMIT PENDING; a bar that gaps through the limit fills at the better open price; TP1 is never booked on the fill bar itself (the intrabar order of touches is unknowable); a never-traded order ends NOT FILLED and contributes nothing to your statistics.

15. The Trade Advice Block (new in v2.3.0)

Three lines at the bottom of the card answer the questions a trader actually has at the moment of decision — every figure from live data or your own journal, never a probability:

ADVICE ▲ LONG  signal 01 May 12:00 UTC (LDN)
  hold <= 60 bars  (~5h, ~5 H4 bars) - then exit or re-assess
  adverse: max 20p = 1.0R at the stop; typical MAE -0.42R (n=12)

• (a) WHEN — the signal bar's time (UTC) and session.
• (b) HOW LONG the premise lives — the plan horizon in bars, translated into wall-clock time using your series' own median bar step, and — with MTF on — into higher-timeframe bars, because "60 bars" is five hours on M5 and ten days on H4. Past that line the plan expires by design: exit or re-assess, do not "hold and hope". This is the number to consult when price goes against you and you are deciding whether the idea is merely early or simply wrong.
• (d) HOW FAR price may travel against you — two honest figures: the contractual maximum (your stop distance in pips and R — beyond it there is no position, only history) and the empirical expectation: the average MAE of your feed's measured trades with its sample size. Adverse movement inside the typical MAE is the plan breathing; consistently beyond it means the entry class is worse than your record — that is data, not opinion.
• (c) The ladder — TP1/TP2/TP3 and the stop live in the plan section directly above, each with its R multiple.

The block is switchable (Toast: Show Trade Advice); with it off, the classic card returns, including the one-line "When" stamp.

16. Signal Journal & Performance Statistics (Module 4.6)

Every accepted signal is recorded with its full plan (entry/stop/TP1–3, gross and net R:R), grade, setup, session, spread, zone, gate count, fill state (market/limit, filled or not, fill time), MFE/MAE and — when the plan resolves — the outcome in R with its status and holding time. The panel block reports:

Journal      : 34 signals | 28 closed | 3 open | 2 not filled | 1 abandoned
               history replay: 12 entries EXCLUDED from the stats below (live record only)
               TP1 first 11 (46%) | stop 9 | other 4
               avg +0.31R  best +2.00R  worst -1.00R  win 58%
               MFE avg +1.12R   MAE avg -0.44R
               PF 1.42  maxDD 3.2R  worst streak 4  t=+1.81  edge not yet significant
               by grade: A+ +0.52R 62% (13)  A +0.14R 54% (11)
               grade check: A+ beats A by +0.38R (n=13/11)

• Populations never mix. Not-filled orders, abandoned entries (a timeframe switch destroyed their plans) and — by default — history-replay signals are counted but excluded from every R aggregate. What you read is your live record on this feed. (Journal: Count History Replay In Stats opts the replay back in.)
• Performance-risk line (v2.2.0+): Profit Factor, max drawdown of the cumulative R curve, worst losing streak, and the expectancy's t-statistic with a blunt verdict: edge not yet significant / edge significant at ~95% / NEGATIVE edge significant. This is the single most important sentence the product prints — see Section 3.
• Grade calibration: per-grade averages and win rates, plus an explicit grade check once both A+ and A hold ≥ 10 measured closes — including the sentence the letters alone never say: "A+ does NOT beat A on this feed". The card quotes the same record next to every new letter.
• CSV export (to the cTrader log — file writes would need permissions the product refuses to take): 30 columns, every row exactly the header's field count, printed every N closes: time, symbol, timeframe, direction, setup, grade, zone, session, htf_aligned, volume_score, gates, entry, stop, tp1, tp2, tp3, rr_gross, rr_net, spread_points, entry_type, filled, fill_time, source, status, result_r, held_bars, mfe_r, mae_r, outcome_time, plan_key — the last column (v2.3.5) is the join key with the AutoTrader's execution CSV.

17. Info Panel, Context Badge, Session Strip (Module 4.2)

• Info panel (top-left by default, position configurable): context and last structure event with age; active BSL/SSL with signed pip distances; PDH/PDL/PWH/PWL with distances; the nearest live zones sorted by distance (score, range or mid price, signed pip distance, STRONG flag, "*" on swept pools); the volume block (bar vs baseline %, rolling, same-hour, mode); squeeze state; profile POC/VAH/VAL and price-vs-VA; the last Nexus signal with its reason; the trade-plan block (levels, live position in R, pending-order state, or the dated outcome — the panel keeps answering "what did the last signal do?" after the projection is gone); the journal block; and the Rejections line — the top "no" reasons since attach, so "why hasn't it fired in three days?" is one glance, not guesswork. The panel refreshes at most once per Panel Refresh (ms) — reading speed, not tick speed.
• Context badge (top-right): one line — chart context arrow, the HTF context when MTF is on, SQUEEZE when active, and the last structure event with its level.
• Session strip (bottom-right): SYD / TKY / LDN / NY open-shut state with countdowns, on your configured UTC windows. The strip is the product's single session clock — the main panel deliberately no longer repeats it.

18. Multi-Timeframe Mode (Module 4.3)

With Enable Multi-Timeframe Mode on, an independent structure + zone stack runs on the chosen higher timeframe (default H4) over closed HTF bars only:

• HTF zones render semi-transparent (HTF Transparency) and carry an [H4] tag in their labels — and in signal reasons, cards and the journal when a signal reacts at an HTF level.
• The HTF context (Bullish / Bearish / Range) feeds the badge, the card's HTF line, the optional Require HTF Confirmation veto and the A+ grade, and arbitrates reversal setups.
• HTF zones share the chart registry and accumulate volume on chart bars — they carry a real Volume Score measured at your chart's resolution. Rendering is owned exclusively by the MTF pass: no delete/redraw churn, and an HTF zone that ages out loses its objects in the same pass.
• Grade A+ needs HTF alignment only when MTF is on — without a higher timeframe there is nothing to align to, and A+ stays reachable.

19. Multi-Symbol Watch (Module 4.7)

An optional board (off by default) listing up to Watch: Max Symbols symbols on a shared higher timeframe: each row shows the symbol's last structural context (BULL ▲ / BEAR ▼ / RANGE ◆ from its own swing-break scan), its current price, and the signed pip distance to its nearest resting BSL / SSL liquidity — where the neighbouring markets keep their stops. One throttled pass per refresh interval, no chart objects beyond one static text, no extra permissions. Since v2.4.2 the board sits in the bottom-right corner with a permanent two-text-row gap above the chart's bottom edge — the toast card's padding convention — and when the session strip is enabled it stacks above the strip, so the strip stays the bottom-most line and the live price area is never covered. The gap no longer depends on the strip being visible: the board breathes on every layout. A typo in the symbol list is kept as a visible "no data" row instead of silently missing.

20. Alerts, Position Sizing & the Cost Model

20.1 Alerts

• Signal alert: the cTrader log receives [LVN ALERT] Nexus LONG | EURUSD M15 :: reason | price | time and (optionally) a platform notification sound. The indicator API has no popup channel — the log line and the corner card are the popup, and the sound is your eyes-free trigger.
• Outcome alert: when a plan ends — STOPPED −1.0R after 5 bars, TP1 TAGGED +2.0R, NOT FILLED (with the honest reason, including the gap-through-stop rejection) — horizon expiry and supersede are treated as bookkeeping and stay silent.
• Squeeze alert: optional, when a squeeze fires at the live edge.
• All alerts are live-edge only: attaching the indicator to a deep history never machine-guns you with last month's signals.

20.2 Position size hint

With Risk Per Plan (% of balance) > 0 (default 1%), the card and panel print the suggested volume for the plan — computed by the platform's own proportional-risk arithmetic from the plan's stop distance, rounded to your symbol's volume step. It is display-only: the product places nothing and never reads your money amounts itself. An unavailable account simply prints no line — a wrong size would be worse than none.

20.3 The cost model

• The spread is re-read from the feed on every closed bar; when a feed publishes none (some CFD/crypto symbols), a manual assumption wins, else a conservative instrument-class estimate applies (1 pip FX; 2 pips metals, indices and crypto — including the six-character crypto pairs). Anything assumed is flagged "(assumed)" on the panel so a cost figure is never mistaken for a quote.
• Net R:R = (reward − spread) / (risk + spread): you pay the spread twice, in and out. The card prints both ratios (capped for display at Max Displayed R:R — above it you see "5+"; gates, journal and CSV keep the raw value), and the gates judge the net one.
• Micro-stop gate: a plan whose entire risk is narrower than Minimum Plan Risk (x spread) (default 3) cannot be executed as printed and is rejected before anything is drawn, alerted or journaled.
• v2.4.1 unit fix — re-run old backtests: the spread is a price difference (Ask−Bid); pre-2.4.1 builds read it as a point count and re-multiplied by TickSize, making every cost gate (net R:R, max spread, micro-stop floor) silently see a near-zero cost. If you measured plans on v2.4.0 or older, those numbers were optimistic — the gates now charge what a broker actually charges.

20.4 Signal Bridge — the robot feed (Module 4.8, new in v2.4.0)

The bridge is a one-way, file-based feed for the companion cBot Liquidity & Volume Nexus — AutoTrader: on every live closed bar the indicator writes one atomic JSON frame — schema version, heartbeat, the current plan, a journal snapshot and the rejection counters — to Documents/cAlgo/LVN/bridge/<SYMBOL>_<TF>.json (temp file + atomic move, so a polling robot never reads a half-written frame). The robot's Verify / Follow modes read that frame; the bridge is never the robot's only channel.

• Default OFF, and honest about permissions: file I/O needs AccessRights.FileSystem. The store build ships None, so switching the bridge on without rebuilding logs the exact one-line attribute change once — then stays silent instead of spamming errors.
• Measurement-neutral: with the bridge ON the engine's canonical dumps are byte-identical to OFF (proven by the rig). Publishing observes; it never decides.
• Live-edge only: history replay publishes nothing — a backtest has no indicator instance, and a replayed frame would be a lie about "now".

21. Complete Parameter Reference — 222 Parameters, 17 Groups

Every parameter with its default and a working description. Defaults in bold. Ranges shown where the dialog enforces them. All of them are cross-validated at start-up: contradictory combinations are reconciled to the safe side and reported in the log when Debug Log is on.

Group 0 — General (7)

Group 1 — Swing Detection (9)

Group 2 — Structure (22)

Group 3 — Fair Value Gaps (18)

Group 4 — Premium / Discount (9)

Group 5 — Liquidity (23)

Group 6 — Volume Nexus (17)

Group 7 — Volume Profile (13)

Group 8 — Nexus Signals (40)

Group 9 — Info Panel (15)

Group 10 — Multi-Timeframe (7)

Group 11 — Chart Clarity & Performance (3)

Group 12 — Signal Toast (11)

Group 13 — Trade Plan (11)

Group 14 — Signal Journal (8)

Group 15 — Symbol Watch (7)

Group 16 — Signal Bridge (2, new in v2.4.0)

SECTION 2 — HOW TO USE

The working trader's guide: every glyph, every line, every decision.

22. Installation & First Launch

1. Install. In cTrader: Automate → install the purchased .algo (or, for a source build: create an indicator named "Liquidity and Volume Nexus", paste the source, press F6 — expected result: 0 errors, 0 warnings), then drag it onto the chart.
2. Pick your theme. White terminal? Set Theme Preset = Light first — the whole palette re-maps.
3. Choose your timeframe. The product works on any TF. Guidance: M5–M15 for intraday signals, H1–H4 for swing. Enable Multi-Timeframe Mode on anything below H4 (default HTF: H4).
4. Set your sessions. Check the four session windows against your broker's UTC schedule (they drive the strip, the session gate and the card's session stamp).
5. Let the history pass finish. On attach the engine replays the loaded history once (seconds, not minutes) — the journal fills with [history]-tagged replay entries and the card shows the most recent signal, correctly aged. With Debug Log on you get one summary line: bars loaded, signals found, newest plan's outcome and the card's state.
6. Read the panel once. Context, spread (note "(assumed)" if your feed publishes none), baseline %, and the Rejections line — it already tells you what the engine is waiting for.

23. Reading the Chart — Every Glyph and Its Variations

This is the visual dictionary. Learn it once and the chart becomes a sentence you can read.

23.1 Swings

23.2 Structure events

23.3 Zones (FVG family)

23.4 Range & prior levels

23.5 Liquidity lines

23.6 Signals & plans

23.7 Profile

24. The Signal Card, Line by Line

▼  SHORT     Grade A+
----------------------------------
Zone   FVG- [H4]
       7639.90 - 7642.79
Score  78  (Strong)
Gates  4 / 4
Track  A+: 12 closed, avg +0.31R, win 58%
Setup  REVERSAL - faded swept BSL (2 bars ago)
Ctx    RANGE  (favours reversal)
----------------------------------
Entry  7638.10  (market @ close)
Stop   7655.80  (sweep high)
TP1    7597.60  +2.3R  (EQH x2 pool)
TP2    7530.47  +4.6R  (SSL pool)
TP3    7463.00  +6.9R  (3x TP1)
R:R    1 : 2.3   net 1 : 2.0
Size   0.42 lots  (1% risk)
ADVICE ▼ SHORT  signal 01 May 12:00 UTC (LDN)
  hold <= 60 bars  (~5h, ~5 H4 bars) - then exit or re-assess
  adverse: max 17.7p = 1.0R at the stop; typical MAE -0.44R (n=12)
----------------------------------
HTF    H4 BEARISH  ✓ aligned
STATUS  LIVE  42 bars left

The RESULT card (after the plan dies) is compact and factual: outcome status, the R, the reason in plain English, when it happened and how long ago ("held 5 bars, 2 bars ago"), the levels, and — for a limit that never became a trade — "no fill" with the honest cause ("the limit never traded" vs "the bar gapped through the entry below the stop — no openable position"). A recap rebuilt after attach carries [history].

25. The Info Panel, Line by Line

26. Life of a Signal — From Close to Record

1. Bar closes. The engine evaluates the closed bar through the ten gates. Most bars die early and are counted in Rejections.
2. Signal fires. The plan is built (stop anchored to structure, ladder from live liquidity, cost checked), validated (a plan with a non-structural stop or a ratio below the gates is rolled back — the engine behaves as if it never fired), then: marker, projection, card, journal entry, alerts (live edge only).
3. The plan lives. Every tick, the life-cycle watches: stop, TP1, premise (zone / structure / supersede) and horizon. The card counts down; the panel shows the live R state.
4. The plan dies — once, latched. The outcome tag appears on the chart, the result card recaps it, the alert reports it, the journal records status + R + holding time + MFE/MAE.
5. The record accumulates. Averages, PF, drawdown, streak, t-statistic and per-grade verdicts grow. When both A+ and A hold ≥ 10 measured closes, the grade check tells you whether the letters mean anything on your feed.

27. Entering a Trade — the Procedure

1. Read the card top-down. Grade and Track (is this letter earning its keep on this feed?), Setup (continuation or fade — they are managed differently), Gates (4/4 or not), Ctx/HTF lines (regime agreement).
2. Check the plan geometry. Stop anchored to something real (the note says what); TP1 at live liquidity; net R:R ≥ your floor; risk in pips you can stomach at the suggested size.
3. Check the ADVICE. Signal time (is this fresh or 40 bars old?), hold window vs the time you can actually manage the position, adverse guidance vs your patience.
4. Place the order manually. Market at the close price, or a limit at the plan's entry — the product places nothing, by design (AccessRights.None). Use the Size line for volume.
5. Set your broker SL at the plan's stop. The measured statistics assume the stop is real. Then let the life-cycle inform you: card STATUS, panel open-R, outcome alerts.

28. Managing a Trade

28.1 The ladder policy

• TP1 is the system's exit. The journal measures every plan as TP1-first — that is what the statistics you read describe. The simplest way to trade Nexus exactly "as measured" is: full exit at TP1.
• TP2 / TP3 are runner objectives. If you scale out (e.g., 50% at TP1, stop to entry, rest toward TP2/TP3), you are running a different policy than the journal measures. That is legitimate — but judge it with the MFE column: the CSV's mfe_r tells you, per trade, how much room runners actually had on this feed.
• The stop does not move against you. Widening a stop converts a measured −1R into an unmeasured loss. Moving it to entry after TP1 is fine — just know the journal keeps scoring the original policy.

28.2 Using the ADVICE numbers while the trade runs

• Adverse excursion. Price drifting against you inside the typical MAE (from the ADVICE line) is the plan breathing — your own feed's history says trades routinely dip that deep before winning. Consistently beyond it, or beyond ~0.7–0.8R with the hold window half spent, the premise is failing earlier than your record says winners do: that is the discretionary cut the numbers support.
• Hold window. "hold ≤ 60 bars (~5h, ~5 H4 bars) — then exit or re-assess" is the premise's shelf life. At expiry the product itself books the plan EXPIRED at the close. If you are still in, you are holding on hope, not on the setup — re-assess against fresh structure, not against your entry price.
• Premise deaths are information. ZONE GONE / STRUCTURE FLIP outcomes mean the reason for the trade dissolved before the price verdict. Taking those exits manually (as the product books them) keeps your trading identical to your statistics.

28.3 What the outcome tag tells you afterwards

"✗ STOP −1.0R" and "✓ TP1 +2.0R" need no translation. The neutral tags are the interesting ones: a cluster of SUPERSEDED outcomes means your cooldown/horizon settings overlap signals (fine — the newer signal replaced the older plan); ZONE GONE shortly after entry on continuations usually marks zones that were already mostly filled — check the entry-location mode; EXPIRED with a large positive MFE says the market moved but never quite tagged TP1 — that is the runner-policy conversation, and the MFE/MAE averages are your evidence.

29. When to Enter — and When to Wait

The engine already filters; this section is about the human decision layer: among the signals that pass, which deserve your money right now, and which should be watched instead. Read it as a scoring conversation between the card, the chart and the panel.

29.1 The green-light combinations (enter)

29.2 The yellow lights (reduce size or wait for confirmation)

• Grade B — one gate short. Usually the HTF or the structure gate. Wait for the confirming event (a BOS in your direction, or the HTF to stop opposing); if it comes while price is still at the level, the setup is alive; if price left without you, it is gone.
• Ctx disagreement: chart RANGE, HTF strongly trending, and the signal is a continuation against the HTF — the veto is off, but the tape is. Half size or skip.
• Continuation late in a leg: the 3rd–4th BOS of one directional run, price deep in premium (for longs) far from EQ — the trend is aging; entries here chase. The ADVICE "typical MAE" from your journal will show it: late-leg trades dip deeper.
• Thin targets: TP1 is a fallback ("fallback 2x ATR") rather than a real pool — the market has no obvious resting liquidity nearby; expect the R:R to be harder to deliver.
• Spread stress: the panel shows the spread near your Max Spread or flagged "(assumed)" around rollover — costs are quietly eating the net ratio; the gate protects you only if it is on.
• Session edges: signals in the last 30–45 minutes of London or New York have a truncated runway to TP1 before the session's volume disappears. Either take them smaller or only on M5–M15 where the horizon is short.

29.3 The red lights (do not enter, whatever the card says)

• The signal is stale. STATUS shows few bars left, or the card is a [history] recap, or price has travelled more than ~⅓ of the way to TP1 since the signal bar. The measured edge belongs to the entry the plan describes — not to a chase of it.
• News is minutes away (inside your blackout window even if the gate is off — the gate protects signals, your cursor needs protecting too).
• Journal says NEGATIVE: the grade's Track line is meaningfully negative with n ≥ 15–20, or the panel prints "NEGATIVE edge significant". The system is telling you, with your own data, to stop. That is the feature working, not failing.
• You cannot place the stop. If your broker's minimum stop distance or the current volatility makes the plan's stop unexecutable, the plan is unexecutable — the micro-stop gate exists for exactly this class of problem; respect its verdict manually when it is about your account, not the print.
• You are already exposed to the same idea: two longs on correlated pairs (check the Watch board — same context everywhere) is one trade with double size.

29.4 "It hasn't fired in days — is it broken?"

No. Open the panel's Rejections line. It is a histogram of the engine's patience:

30. False Moves & Breakout Clues — the Interpretation Chapter

Liquidity reading is, at bottom, the study of two questions: was that move real, or was it harvesting? and is the market about to leave this range, and in which direction? This section is the field guide.

30.1 Anatomy of a false move (the sweep / SFP)

The reliable signature, in order of appearance:

1. A level with obvious resting orders: EQH/EQL pool, PDH/PDL/PWH/PWL, the active BSL/SSL, a round number. The indicator labels all of these — the fuel is visible before the match.
2. The raid: a fast push through the level, typically a single bar or two, on elevated volume. In Wick sweep mode the pool greys to "SWEPT" immediately; the raid's extreme is recorded.
3. The failure to progress: the close comes back inside the level (the SFP). This is the tell. A breakout that is real closes beyond and holds; a harvest wicks beyond and returns.
4. The consequence: everyone who entered on the "breakout" is trapped beyond the level; their stops now fuel the opposite move. A REVERSAL signal firing here is the product trading exactly this mechanism — with the stop beyond the raid extreme, i.e. beyond the trap.

30.2 False-move variants you will meet

30.3 Breakout clues — when the range is about to end

• Squeeze + location. SQZ fired while price sits compressed just under a heavy EQH or just above an EQL (or against PDH/PDL) — the coil is loaded against a specific wall. The expansion, when volume returns, has a first direction bias: away from the wall it could not break, or through it if the release bar closes beyond with ≥ 1.5–2× baseline volume.
• Repeated defence with shrinking recoil. Each touch of a level makes a shallower pullback (visible as successive higher lows into an EQH): absorption. The level is being eaten from below — breaks under absorption tend to run.
• Value-area migration. The profile's VAH/VAL of the current session pushing against the prior session's — price is being accepted progressively higher/lower. Breaks in the direction of migration have follow-through; breaks against it are usually fades.
• Structure compression. Active BSL and SSL converging (both within a few pips on the panel) = a triangle of resting liquidity on both sides. Whichever goes first harvests the other side's stops — many desks wait for that first raid and trade the SFP if it fails, or the continuation if it holds.
• HTF context turning. An H4 CHoCH while the chart TF is still ranging against it — the higher timeframe is early, and chart-TF continuation signals in the new HTF direction are the ones to prioritise from here on.

30.4 Putting squeeze, sweeps and context together — four chart states

31. Multi-Timeframe Workflow

1. Set the pair. Rule of thumb: HTF ≈ 4–6× the chart TF. M5→M30/H1, M15→H1/H4, H1→H4, H4→D1. Enable MTF; the badge now shows both regimes.
2. Start from the HTF story. Badge: what is the H4 context? Where is price relative to the H4 active BSL/SSL and the [H4]-tagged zones? That is your permission slip: with H4 BULLISH you favour longs on the chart TF, treat shorts as scalps, and let Require HTF Confirmation veto the rest if you want machine discipline.
3. Trade the chart TF inside that story. The best signals are the ones where both lines of the badge point the same way — those are the A+ candidates (alignment is literally the A+ gate).
4. Use the ADVICE HTF line. "hold ≤ 60 bars (~5h, ~5 H4 bars)" tells you the plan spans about one H1 candle of H4 decision-time — an H4 trader would not even notice it. If the HTF count is < 1 bar, your horizon is shorter than the higher timeframe's heartbeat: either extend Trade Plan Horizon or accept scalp semantics.
5. When contexts fight, the Watch board and the rejections line explain why nothing fires — and the honest answer is usually "wait for the HTF to finish its event".

32. Working with the Journal, Statistics & CSV

32.1 The weekly five-minute review

1. Read the populations line: how many signals, how many closed, how many not-filled (limit mode), how many abandoned (you switch timeframes a lot?).
2. Read the performance-risk line: PF > 1.3 with maxDD you could have lived with = the system is paying; t ≥ 2 = the sample now vouches for it; "not yet significant" = keep the size small, keep collecting.
3. Read the grade ladder: is A+ actually beating A on your feed? If the grade check says no for weeks, your minimum-grade setting is a style choice, not an edge filter — that is fine, but know it.
4. Read MFE vs the outcome mix: avg MFE ≫ avg result with many EXPIRED/ZONE GONE = your exits leave money on the table → study the TP ladder policy (28.1). Avg MAE ≈ −1R with mostly stops = entries are late → tighten the entry-location mode or the chase cap.
5. Export the CSV periodically (or set the auto-print), and keep a folder by month. The journal ring holds up to 2000 entries; your archive is forever.

32.2 CSV columns (30)

time, symbol, timeframe, direction, setup, grade, zone, session, htf_aligned, volume_score, gates, entry, stop, tp1, tp2, tp3, rr_gross, rr_net, spread_points, entry_type, filled, fill_time, source, status, result_r, held_bars, mfe_r, mae_r, outcome_time, plan_key — every row has exactly the header's field count; plan_key (v2.3.5) joins a journal row to the robot's execution row one-to-one; unfilled orders carry empty R/MFE/MAE cells because they were never trades; source separates live from replay. This is a complete forward-test dataset: paste it into any spreadsheet and pivot by setup/grade/session — the product deliberately keeps its own aggregation simple and gives you the raw truth.

time,symbol,timeframe,direction,setup,grade,zone,session,htf_aligned,volume_score,gates,entry,stop,tp1,tp2,tp3,rr_gross,rr_net,spread_points,entry_type,filled,fill_time,source,status,result_r,held_bars,mfe_r,mae_r,outcome_time,plan_key
2026-09-28 17:00,EURUSD,h1,SHORT,CONTINUATION,B,IFVG- 1.13731,LDN+NY,true,79,5/7,1.13716,1.13912,1.13487,1.13260,1.13033,1.72,1.41,1.2,limit,true,2026-09-28 19:00,live,STOPPED,-1.0,1,0.4,-0.8,2026-09-29 02:00,lvn-e7f3a1

33. Alerts & Debug Log

• [LVN ALERT] — every live signal and every outcome (stop/TP1/premise deaths, NOT FILLED with its reason) lands in the cTrader log with this prefix, plus an optional sound. The cTrader indicator API has no popup channel; the corner card IS your popup, and the log is your paper trail.
• Live-edge discipline — attaching to a chart never replays last month's alerts at you; history is silent by design.
• Debug Log = true adds: the history-pass report (one line: bars, signals, newest plan outcome, card state and why), every card transition, per-bar rejection reasons at the live edge, and start-up settings normalisation notes. Turn it on when something seems missing; leave it off for quiet operation.
• Squeeze alerts are separate and off by default — they are informational, not trading signals.

34. Timeframe & Instrument Guide

SECTION 3 — TIPS

The habits that make the product pay for itself — and the mistakes that make traders quit it.

35. The Golden Rules

36. Recommended Presets by Trading Style

Save each as a cTrader preset (.preset) after configuring — switching styles should take one dropdown, not twenty dialogs.

37. Risk Management with Nexus

• Start from the Size line. It is the platform's own proportional-risk arithmetic on the plan's stop distance at your configured % — the single most execution-ready number on the card. 1% default is a professional starting point; 0.5% while the journal is young.
• Respect the cost gates. Minimum NET R:R ≥ 1.0 and Minimum Plan Risk ≥ 3× spread are the floor of tradability. Raising them (net 1.2–1.5 on wide-spread instruments) improves the average outcome at the price of fewer signals — usually a trade worth making.
• Use maxDD as your sizing mirror. The journal's max drawdown in R is what holding this system actually felt like on your feed. If 4R of drawdown would have made you quit, your % per trade is too high — that is arithmetic, not psychology.
• Correlated exposure. The Watch board shows neighbouring symbols' regimes at a glance: three pairs all BULL ▲ on H4 means one dollar-risk trade wearing three costumes. Count the idea, not the tickets.
• The losing streak will come. The "worst streak" statistic exists so the streak doesn't surprise you: at a 50–55% win rate, streaks of 4–6 losses are ordinary weather. Sizing is the only defence that works against weather.
• Never average down against a plan. The stop beyond the raid extreme / swing is where the idea is proven wrong. Adding size beyond it converts a measured −1R into an unmeasured disaster — the one mistake no indicator can survive.

38. What NOT to Do

39. What to Follow — the 30-Day Onboarding Plan

1. Days 1–3 — observe. Defaults on your 1–2 chosen charts, MTF on, Debug Log on. Take nothing. Read each card top-down when it appears; read the history-pass report; learn your rejections line's normal shape.
2. Days 4–10 — paper the signals. Log every card you would take (a screenshot or a CSV note). Compare against what the journal books. You are calibrating your eyes against the machine's honesty.
3. Days 11–20 — minimum size, live. 0.25–0.5% risk. Trade only green-light combinations (29.1) — expect 2–5 trades, not 20. Follow the plan mechanically: stop real, exit at TP1.
4. Day 21 — first review. Journal block + CSV: does the shape match your paper week? MFE/MAE telling you your exits are early/late? Adjust ONE thing (horizon, entry mode, or size) — never five.
5. Days 22–30 — steady state. Continue at minimum size until the measured count crosses ~20–30 closes. Only then let the t-statistic conversation decide whether to scale toward 1%.

40. Reading Grades & Track Records Correctly

• A grade is a checklist, not a forecast. A+ = very busy zone + all gates + HTF agreement. It says the setup is well-formed, not that it wins. What it wins on your feed is the Track line's job.
• Compare letters only within your feed. "A+ averages +0.52R here" and "A averages +0.14R" is actionable; someone else's A+ statistics are not your market.
• The grade check sentence is a feature, not an insult. "A+ does NOT beat A on this feed (n=13/11)" means: on this symbol/TF/broker, the HTF-alignment and score bonuses add no measurable value — trade both letters equally, or investigate why (often: your HTF pair is wrong for this instrument — try a different Higher Timeframe).
• Sample sizes rule everything. n=4 means nothing; n=15 is a hint; n=30+ is evidence. The panel prints n next to every number precisely so you never over-read a hot week.
• Per-session insight via CSV: pivot result_r by the session column — many feeds show LDN and NY behaving differently; the session filter then becomes an evidence-based choice instead of folklore.

41. Performance & Chart Hygiene Tips

• Attaching is cheap since v2.4.2: the history pass defers every draw nobody can see — structure events deeper than Render History (the surviving budget set is backfilled once at the window), intermediate zone and profile refreshes, unchanged active-level lines — measured 13× fewer chart operations on a 30 000-bar H4 replay (≈30 000 → ≈2 300), final chart state byte-identical, journal untouched. Load time barely depends on history depth any more. On weak machines you can still trim ink: Render History = 200–300, Zone Redraw Interval = 40, Panel Refresh = 2000 — signals and journal never change.
• H4 + MTF configuration note: if your chart timeframe is the "Higher Timeframe" (H4 chart with HTF = H4), the MTF engine re-analyses the very same series — double work, zero new information, and both engines share one zone budget. Set the HTF one step above the chart (H4 chart → D1), or switch MTF off on that chart.
• Clutter control without blinding the engine: Max Zones Drawn / Max Liquidity Lines Drawn cap what is painted, never what is evaluated. A clean chart with a full engine is the intended experience.
• Timeframe switching is safe: open plans are journal-closed as abandoned (excluded from every average), objects are removed, and the history pass rebuilds everything. You will see the "[history]" card recap — that is the design, not a bug.
• Baseline History: 5000 bars is the sweet spot; 0 (all loaded) is fine memory-wise since v2.0.0 and gives the most seasonal stability on symbols with months of history.
• One instance per chart. Two instances of the product on the same chart would double every object family for no new information.

42. FAQ

43. Troubleshooting

44. Glossary

Liquidity & Volume Nexus — Official User Manual · v2.4.2 · © 2026. Trading involves substantial risk of loss. Past measurements on one feed do not guarantee future results; this manual describes tooling, not investment advice.