Liquidity analysis

Skill: Liquidity Analysis | Domain: trading | Category: analysis | Level: advanced Tags: trading, liquidity, orderflow, microstructure, institutional, execution

Liquidity Analysis

Connects: ICT/SMC liquidity sweeps ← this file → market microstructure execution For BSL/SSL sweep mechanics and trap entries → see ict-smart-money.md For bid-ask spread, slippage formulas, market impact → see market-microstructure.md

Liquidity Pool Identification & Mapping

Internal vs External Liquidity

EXTERNAL LIQUIDITY (outside price range — draw on liquidity targets):
  ┌─── Previous Day High (PDH)      ← buy stops cluster above
  │    Previous Week High (PWH)      ← larger pool
  │    Previous Month High (PMH)     ← institutional target
  │
  │    [Current Price Action]
  │
  │    Previous Day Low (PDL)        ← sell stops cluster below
  │    Previous Week Low (PWL)       ← larger pool
  └─── Previous Month Low (PML)     ← institutional target

INTERNAL LIQUIDITY (inside price range — fills and rebalancing):
  ├── Fair Value Gaps (FVG)          ← unfilled orders, price returns to fill
  ├── Order Blocks (OB)             ← institutional entry zones
  ├── Breaker Blocks                ← failed OB now acts as magnet
  └── Volume Imbalances             ← one-sided orderbook zones

Liquidity Pool Formation Rules

Where pools form (strongest to weakest):
1. Equal highs/lows (EQH/EQL)     → engineered liquidity, highest probability sweep
2. Session highs/lows              → time-based, regular collection
3. Swing structure points          → structural, medium probability
4. Round numbers (00, 50 levels)   → psychological, retail clusters
5. Previous close levels           → institutional reference points

Pool strength scoring:
  +2  Equal highs/lows (multiple touches at same price)
  +2  Untested (first time price approaches)
  +1  Multiple timeframe alignment (4H pool + daily pool)
  +1  Near a key session level (PDH/PDL, PWH/PWL)
  +1  Volume profile thin zone nearby (price will move fast through)
  -1  Already swept once (diminishing returns)
  -2  Third approach (most liquidity already taken)

Score 5+ = high-probability target  |  Score 3-4 = medium  |  Score <3 = skip

Multi-TF Liquidity Alignment

Strongest setup: liquidity pools aligned across timeframes

Monthly → Weekly → Daily → 4H → 1H
  │         │        │       │     │
  └─ PMH ───┴─ PWH ──┴─ PDH ─┴─ EQH ─ All pointing to same zone
                                         = MAXIMUM draw on liquidity

Rule: If 3+ timeframes have liquidity resting at the same zone,
      that zone WILL be reached with >80% probability.

Mapping process:
1. Mark monthly highs/lows (PMH/PML) on daily chart
2. Mark weekly highs/lows (PWH/PWL)
3. Mark daily highs/lows (PDH/PDL)
4. Mark 4H equal highs/lows
5. Highlight zones where 2+ levels cluster within 0.3% price range

Institutional Liquidity Behavior

How Institutions Scale In/Out

Accumulation Phase (buying):
1. Create sell-side liquidity → push price below key lows (stop hunt)
2. Absorb selling pressure → large bid stacking on DOM
3. Mark up → aggressive buy orders, displacement candles
4. Distribute at higher prices → sell into retail FOMO

Distribution Phase (selling):
1. Create buy-side liquidity → push price above key highs
2. Absorb buying pressure → large offer stacking
3. Mark down → aggressive selling, displacement
4. Accumulate at lower prices → buy retail panic

Key tells:
- Delta divergence: price rises but cumulative delta flat/falling = distribution
- Volume spikes at extremes (not mid-range) = institutional activity
- Spread widening then tightening = large order absorbed

Block Trade & Iceberg Detection

Block trade signatures on chart:
- Single large-volume bar with minimal price impact = absorbed
- Cluster of equal-size orders at same price level = iceberg
- Volume spike with NO price movement = institutional limit order wall

DOM reading for iceberg detection:
- Bid/ask size refreshes at same price after fills = iceberg
- Large resting order that doesn't pull when tested = real
- Large order that pulls when price approaches = spoofing

Dark Pool Dynamics

When to expect dark pool activity:
- Equity CFDs during US session (most dark pool volume)
- Index futures near settlement times
- Large-cap stocks during first/last 30 min of session

Dark pool impact on price:
- Price drifts without visible volume → dark pool execution
- Delayed reporting creates "ghost" volume bars
- Price discovers large orders AFTER execution completes
- Watch for post-block momentum: direction of next move reveals institutional intent

Liquidity Dynamics by Session & Asset

Forex Session Liquidity Map

Session        | Liquidity | Spread    | Best For
───────────────┼───────────┼───────────┼─────────────────────
Asia 00-07 UTC | LOW       | WIDE      | Range identification only
London 07-12   | HIGH      | TIGHT     | Breakouts, sweeps, trend starts
NY 13:30-17    | HIGHEST   | TIGHTEST  | All strategies valid
Overlap 13:30-16| PEAK     | MINIMUM   | Maximum opportunity
NY Close 20-22 | DROPPING  | WIDENING  | Exit positions, no new entries
Weekend gap    | ZERO      | MASSIVE   | Gap risk, not trading time

Pair-specific liquidity peaks:
  EUR/USD: London + NY overlap (13:30-16:00 UTC)
  USD/JPY: Asia (00-07) + NY overlap
  GBP/USD: London open (07:00-08:00 UTC) = maximum volatility
  AUD/USD: Asia open (22:00-01:00 UTC)
  XAU/USD: NY session only (13:30-17:00 UTC) — thin outside

Indices Intraday Liquidity Pattern

Time (ET)    | Phase           | Liquidity | Action
─────────────┼─────────────────┼───────────┼─────────────────
09:30-10:00  | Opening auction | EXTREME   | ORB, initial direction
10:00-11:30  | Morning flow    | HIGH      | Trend continuation
11:30-13:30  | Lunch doldrums  | LOW       | No new trades, tighten stops
13:30-14:30  | Afternoon start | RISING    | Reversal or continuation
14:30-15:30  | Power hour      | HIGH      | Strong directional moves
15:30-16:00  | MOC imbalance   | EXTREME   | Closing auction, gap risk

Crypto Liquidity Fragmentation

Challenge: Liquidity split across 20+ exchanges
  CEX (Binance, Coinbase): 80%+ of BTC liquidity
  DEX (Uniswap, dYdX): 5-10%, higher slippage

Crypto liquidity calendar:
  Best: US business hours (higher CEX volume)
  Worst: Weekend 04:00-08:00 UTC (minimum global activity)
  Dangerous: After large liquidation cascade (thin books)

Unique crypto risk:
  - Exchange-specific flash crashes (thin book + market sell)
  - Funding rate spikes = forced liquidations = liquidity vacuum
  - Stablecoin depegs drain liquidity from ALL pairs

Micro-Level Liquidity Analysis

Tick Data Interpretation

def order_flow_imbalance(trades_df):
    """Calculate order flow imbalance from tick data.

    Args:
        trades_df: DataFrame with columns [timestamp, price, volume, side]
                   side: 'buy' (market buy hitting ask) or 'sell' (market sell hitting bid)
    """
    buy_vol = trades_df[trades_df['side'] == 'buy']['volume'].sum()
    sell_vol = trades_df[trades_df['side'] == 'sell']['volume'].sum()
    total = buy_vol + sell_vol

    if total == 0:
        return 0.0

    imbalance = (buy_vol - sell_vol) / total  # Range: -1 to +1

    # Interpretation:
    # > +0.3  = strong buying pressure (institutional accumulation?)
    # < -0.3  = strong selling pressure (institutional distribution?)
    # -0.1 to +0.1 = balanced (no edge, ranging market)

    return imbalance

Volume Profile as Liquidity Indicator

POC (Point of Control):
  = Price level with highest traded volume
  = Where most limit orders were filled
  = Strong support/resistance (liquidity magnet)

VAH/VAL (Value Area High/Low):
  = 70% of volume traded between these levels
  = Price outside VA = in thin liquidity zone → fast moves
  = Price returning to VA = seeking liquidity → mean reversion

Liquidity interpretation by VP shape:
  D-SHAPE:  Heavy center, thin edges → ranging, fade extremes
  P-SHAPE:  Top-heavy → sellers exhausting, potential reversal down
  B-SHAPE:  Bottom-heavy → buyers exhausting, potential reversal up
  THIN:     No dominant level → trending, trade with momentum
  BIMODAL:  Two POCs → market undecided, trade the space between

DOM/Level 2 Reading

Key DOM patterns:
1. BID STACKING: Large resting bids below current price
   → Support building, institutional buying interest
   → BUT: could be spoofing (pulls when tested)
   → Confirmation: bids don't pull when touched + absorption

2. OFFER STACKING: Large resting asks above current price
   → Resistance building, institutional selling interest
   → Same spoofing risk — watch for pulls

3. ABSORPTION: Large orders at a level keep getting filled
                but price doesn't move through
   → Strong institutional defense of that level
   → High-probability reversal zone

4. PULLING: Large orders disappear when price approaches
   → Spoofing / bluffing — price will move THROUGH that level
   → Often precedes a sweep

5. FLIPPING: Bid stacking switches to offer stacking (or reverse)
   → Institutional intent changing
   → Strong directional signal

Liquidity-Based Entry & Exit Rules

When NOT to Trade (Liquidity Dry-Up)

NO TRADE conditions:
□ Spread > 2x normal for this session/pair
□ Volume < 30% of 20-bar average
□ ATR last 5 bars < 40% of session ATR average
□ Within 15 min of red-folder news event
□ Bid-ask depth < 50% of normal (DOM thinning)
□ Friday after 17:00 UTC (weekend risk, thin markets)

If 2+ conditions met → NO NEW TRADES
If 3+ conditions met → CLOSE or TIGHTEN existing positions

Position Sizing for Liquidity

def liquidity_adjusted_size(base_lots, symbol, session, spread_now, spread_avg):
    """Adjust position size based on current liquidity conditions."""

    # Spread ratio: current spread vs session average
    spread_ratio = spread_now / spread_avg if spread_avg > 0 else 2.0

    # Liquidity multiplier
    if spread_ratio <= 1.0:
        mult = 1.0          # Normal liquidity — full size
    elif spread_ratio <= 1.5:
        mult = 0.75          # Slightly thin — reduce 25%
    elif spread_ratio <= 2.0:
        mult = 0.5           # Thin — half size
    elif spread_ratio <= 3.0:
        mult = 0.25          # Very thin — quarter size
    else:
        mult = 0.0           # No trade — liquidity gone

    return round(base_lots * mult, 2)

Slippage Prediction Model

Expected slippage by condition:

Condition                           | Expected Slippage
────────────────────────────────────┼──────────────────
Normal session, major pair          | 0.0-0.2 pips
Normal session, minor/exotic        | 0.5-2.0 pips
News release (first 30 sec)        | 5-50 pips
Low-liquidity session (Asia FX)    | 0.3-1.0 pips
Large position (>5 std lots)       | 0.5-3.0 pips
Market close / rollover            | 1.0-5.0 pips
Flash crash / liquidity vacuum     | 20-200+ pips

Rule: If expected slippage > 20% of stop loss distance → DO NOT TRADE
      Position too large or market too thin for this setup.

Liquidity Clusters & Confluence Scoring

Cluster Identification

A liquidity CLUSTER forms when 3+ factors converge:

Example cluster:
  PDH at 2045.50           ← external liquidity
  + Equal highs at 2045.80 ← engineered pool
  + Fib 1.618 at 2046.00   ← algorithmic orders
  + Round number 2050       ← psychological level
  = STRONG liquidity cluster (4 factors within 0.2% range)

Scoring:
  2 factors = moderate cluster (trade normally)
  3 factors = strong cluster (likely target, use as TP)
  4+ factors = extreme cluster (almost certain to be reached)

Action at clusters:
  - If price APPROACHING cluster → hold/add to position (target being reached)
  - If price AT cluster → take profits (liquidity being consumed)
  - If price SWEPT cluster → look for reversal (purpose fulfilled)

Multi-TF Liquidity Heat Map

Build a heat map by marking liquidity levels from multiple TFs:

Monthly  ━━━━━━━━━━━━━━━━━━━━━━ PMH (2100)

Weekly   ─────────────────────── PWH (2080)

Daily    ┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄ PDH (2055)
4H       ............EQH (2052)
         ............EQH (2048)  ← CLUSTER: PDH + 4H EQH within 7 pts

         [Current price: 2030]

4H       ............EQL (2015)
Daily    ┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄┄ PDL (2010)  ← CLUSTER: PDL + 4H EQL
Weekly   ─────────────────────── PWL (1995)
Monthly  ━━━━━━━━━━━━━━━━━━━━━━ PML (1950)

Trade logic:
  If TRENDING UP → target PDH/EQH cluster (2048-2055)
  If TRENDING DOWN → target PDL/EQL cluster (2010-2015)
  PWH/PWL = extended targets for swing trades

Real-Time Liquidity Monitoring

Spread Behavior as Signal

Spread tightening (converging to minimum):
  → Liquidity increasing → safe to enter
  → Typically: session open, pre-news positioning

Spread widening (expanding above average):
  → Liquidity decreasing → reduce exposure
  → Typical: news release, session close, flash event

Rapid spread spike then return:
  → Large market order absorbed → institutional activity
  → Watch price direction after absorption = institutional intent

Volume Surge Detection

def detect_volume_surge(df, lookback=20, threshold=2.5):
    """Flag bars where volume exceeds threshold × rolling average."""
    avg_vol = df['volume'].rolling(lookback).mean()
    df['vol_ratio'] = df['volume'] / avg_vol
    df['is_surge'] = df['vol_ratio'] > threshold

    # Surge + displacement = institutional breakout
    # Surge + no price movement = absorption (reversal signal)
    # Surge + wick rejection = stop hunt / liquidity sweep

    return df

Case Studies: Trap vs Real Move

Identifying False Sweeps (Bait)

FALSE SWEEP (trap — don't chase):
  - Wick pokes through level but body closes back inside
  - Volume on sweep bar is LOW (not institutional)
  - No displacement candle follows within 3 bars
  - Spread doesn't widen significantly during sweep
  → Result: Price reverses — the sweep was the move

REAL BREAKOUT (institutional — trade continuation):
  - Full body close through level with displacement
  - Volume on break bar is HIGH (2x+ average)
  - Immediate follow-through bars in same direction
  - Spread tightens after break (liquidity flowing into direction)
  → Result: Price continues — the break was institutional commitment

Key distinction: VOLUME + DISPLACEMENT + FOLLOW-THROUGH
  All three present → real move
  Any missing → likely trap

Accumulation vs Retail Trap

ACCUMULATION (institutional buying):
  Time: Days/weeks of ranging price action
  Volume: Gradually increasing on up-bars, decreasing on down-bars
  Delta: Slowly building positive cumulative delta
  DOM: Persistent bid stacking that doesn't pull
  Result: Eventually breaks up with massive displacement

RETAIL TRAP (designed to shake out weak hands):
  Time: Hours, usually during killzone transitions
  Volume: Spike on stop-hunt bar, then normalizes
  Delta: Sharp negative spike then rapid recovery
  DOM: Bids appear then immediately pull (spoofing)
  Result: Price reverses violently after shakeout

Liquidity & Risk Management

Position Size by Market Liquidity

Tier 1 (deepest liquidity) → full 1% risk:
  EUR/USD, USD/JPY, GBP/USD, XAU/USD (NY session)
  ES, NQ, SPY during RTH

Tier 2 (good liquidity) → 0.75% risk:
  AUD/USD, USD/CAD, USD/CHF
  BTC/USD (major CEX, US hours)
  Index CFDs during RTH

Tier 3 (moderate liquidity) → 0.5% risk:
  Cross pairs (EUR/GBP, GBP/JPY)
  Stock CFDs (TSLA, AAPL during RTH)
  XAG/USD, USOILm

Tier 4 (thin liquidity) → 0.25% risk:
  Exotic pairs (USD/TRY, USD/ZAR)
  Crypto altcoins
  Any instrument outside its primary session
  Pre/post-market stock CFDs

Drawdown Amplification in Thin Markets

In low-liquidity conditions:
  - Stop slippage increases 3-10x
  - Gap risk becomes significant
  - Correlation spikes (all positions move together)
  - Drawdown compounds faster than expected

Protection rules:
  □ Maximum 3% portfolio heat in thin markets (vs 6% normal)
  □ No positions held through major news in illiquid instruments
  □ Close or hedge exotic/crypto positions before weekend
  □ Use guaranteed stops where available (accept wider cost)

---

Related Skills

• Ict Smart Money
• Market Microstructure
• Technical Analysis
• Price Action
• Execution Algo Trading
• Market Making Hft
• Risk And Portfolio