Skip to main content
Cluster charts (also called footprint charts) are the primary analytical tool in Cluster Terminal. Instead of showing a single volume bar beneath each candle, clusters decompose the candle’s range into individual price levels and display exactly how much volume was traded at each level, who initiated the trades, and where significant imbalances occurred. This is the highest-resolution view of market microstructure available.

How Clusters Work

Every trade executed on Binance Futures or Spot is classified as either a buy (aggressor lifted the ask) or a sell (aggressor hit the bid). The server aggregates these trades by price level within each candle period and transmits them as cluster data. Each horizontal row in a cluster candle represents a single price level (or group of levels, depending on the tick size). The data available at each level includes:
MetricDefinition
VolumeTotal contracts traded at this level (buy + sell)
Bid VolumeContracts sold at the bid (sell aggressor)
Ask VolumeContracts bought at the ask (buy aggressor)
DeltaAsk volume minus bid volume (positive = net buying)
TradesNumber of individual trade executions at this level
Delta is calculated from aggressor-side classification. A trade that lifts the ask is counted as a buy; a trade that hits the bid is counted as a sell. This is the industry-standard method used by all professional footprint chart platforms.

Cluster Display Modes (Presets)

Cluster Terminal provides 18 cluster presets that control what data is shown in each cell and how cells are colored. These are grouped into several families:

Volume Presets

PresetCell ContentCell StyleColor Logic
VolumeTotal volumePlain textGradation by volume magnitude
Volume ProfileTotal volumeHorizontal barBar width proportional to volume
Delta Colored VolumeTotal volumePlain textCell background colored by delta direction
Delta Colored Volume ProfileTotal volumeHorizontal barBar colored by delta direction
Volume is the simplest cluster mode. Each cell shows the total number of contracts traded at that price level. Higher volumes receive more saturated background colors, making it easy to spot significant levels at a glance. Volume Profile renders cells as horizontal bars extending from the left edge, where bar length is proportional to volume relative to the candle’s maximum. The POC (highest volume) level appears as the longest bar.

Delta Presets

PresetCell ContentCell StyleColor Logic
DeltaDelta valuePlain textGreen for positive, red for negative
Delta ProfileDelta valueHorizontal barBar colored green/red by delta sign
Delta mode shows the net difference between buy and sell volume at each level. Positive delta (more buying) is colored green; negative delta (more selling) is colored red. The intensity of the color reflects the magnitude of the delta. This mode is essential for reading absorption (large volume at a level with minimal price movement) and exhaustion (delta shifting direction at price extremes).

Bid x Ask Presets

PresetCell ContentCell StyleColor Logic
Bid x AskBid left, Ask rightPlain textBid and ask in separate columns
Bid x Ask ProfileBid left, Ask rightHorizontal barsBars for each side
Bid x Ask Delta ProfileBid left, Ask rightHorizontal barsBars colored by delta
Bid x Ask ImbalanceBid left, Ask rightPlain textDiagonal imbalance highlighting
Bid x Ask is the classic footprint view. Each cell is split into two columns: the left column shows the bid (sell) volume and the right column shows the ask (buy) volume. This provides the most granular view of the order flow. Bid x Ask Imbalance compares each bid level to the ask level one tick above it (diagonal comparison). When one side exceeds the other by the configured imbalance ratio, the dominant side is highlighted. This is the standard method used in professional order flow analysis.

Trades Presets

PresetCell ContentCell StyleColor Logic
TradesTrade countPlain textGradation by trade count
Trades ProfileTrade countHorizontal barBar width proportional to trades
Delta Colored TradesTrade countPlain textBackground by delta direction
Delta Colored Trades ProfileTrade countHorizontal barBar colored by delta
Trades mode shows the number of individual executions rather than the total volume. This is useful for detecting algorithmic activity — an unusually high number of small trades at a level may indicate iceberg orders or algorithmic execution, while a small number of very large trades suggests institutional block orders.

Combined Presets

PresetLeft ColumnRight ColumnCell Style
Volume TradesVolumeTrade countPlain text
Volume Trades ProfileVolumeTrade countHorizontal bars
Volume DeltaVolumeDeltaPlain text
Volume Delta ProfileVolumeDeltaHorizontal bars
Combined presets show two metrics side by side in each cell. Volume Trades helps you quickly calculate the average trade size (volume / trades) at a glance. Volume Delta pairs total activity with directional bias.

Tick Size

Tick size determines the price resolution of each cluster row. It controls how many price increments are grouped into a single horizontal band.

What It Means

For example, if BTCUSDT has a minimum price increment of 0.10andtheticksizeissetto1,eachrowcovers0.10 and the tick size is set to **1**, each row covers 0.10. If the tick size is set to 10, each row covers 1.00,aggregatingalltradeswithinthat1.00, aggregating all trades within that 1.00 range.

Auto-Detection

Cluster Terminal automatically detects the appropriate tick size for each symbol based on the exchange API. For BTCUSDT on Binance Futures or Spot, the default is typically 1 tick ($0.10). For lower-priced altcoins, the default adjusts accordingly.

Manual Override

You can manually set the tick size in the chart toolbar. This is useful in several scenarios:
ScenarioRecommended Tick Size
Scalping on 1m chartsDefault (smallest) for maximum precision
Day trading on 5m — 15m charts2x — 5x default to reduce visual noise
Swing analysis on 1h — 4h charts10x — 50x default for readable clusters
High timeframes (1d+)Large tick sizes, or switch to candlestick mode
Very small tick sizes on lower timeframes create a large number of rows per candle, which increases memory usage and rendering load. If your chart becomes sluggish, increase the tick size or switch to a higher timeframe.

Impact on Analysis

Smaller tick sizes provide more granular data but can create visual noise. Larger tick sizes smooth out the data and make patterns more visible at the cost of precision. There is no universally correct tick size — it depends on your timeframe, the instrument’s volatility, and your trading style.
On higher timeframes (4H, 1D), increase the tick size to reduce visual noise. On lower timeframes (1m, 5m), the default tick size usually provides the best detail for order flow analysis.

POC (Point of Control)

The Point of Control is the price level with the highest traded volume inside a candle. It represents the “fairest” price of the period — where the most trading activity occurred.

Enabling POC Display

POC display is controlled by the Show POC toggle in cluster settings. When enabled, the POC level is highlighted with a distinct frame around the cell.

POC Frame Settings

SettingDescriptionDefault
POC BasisWhat metric defines the POC: volume or tradesVolume
Frame ColorThe outline color of the POC cellAmber (#f59e0b)
Frame WidthThickness of the POC highlight border in px2

Using POC in Trading

The POC serves as a natural support/resistance reference:
  • POC as support: When price pulls back to a previous candle’s POC from above, the high-volume level may attract buyers and provide support.
  • POC as resistance: When price rallies toward a previous candle’s POC from below, sellers who previously established positions there may defend the level.
  • POC migration: When the POC shifts significantly from one candle to the next, it indicates that the “fair value” is moving, often confirming a trend.

Naked POC

A Naked POC is a POC level from a previous candle that has not yet been revisited by price. Naked POCs often act as magnets — price tends to return to these untested high-volume levels. In Cluster Terminal, Naked POC levels can be extended as horizontal rays via the TPO chart settings.

Value Area

The Value Area defines the price range that contains a configurable percentage (default 70%) of the candle’s total volume. It marks the zone where most trading occurred.

VAH and VAL

  • VAH (Value Area High) — the upper boundary of the value area.
  • VAL (Value Area Low) — the lower boundary of the value area.

Adjusting the Percentage

The default Value Area is 70%, following the standard one standard deviation of a normal distribution. You can adjust this in the cluster settings under Value Area Percent. Common alternatives:
PercentageUse Case
68%One standard deviation (statistically precise)
70%Industry standard for Market Profile and footprint analysis
80%Wider value zone for lower-volatility environments

Visual Representation

When Show Value Area is enabled, the value area range is drawn with a semi-transparent overlay on each candle. Levels outside the value area appear dimmer, making the value area boundaries easy to identify.

Trading with the Value Area

The Value Area is a cornerstone of auction market theory:
  • Value acceptance: If price opens inside the previous candle’s value area, the market is accepting the established value. Expect rotation and range-bound behavior.
  • Value rejection: If price breaks above VAH and holds, buyers are asserting new higher value. If price breaks below VAL and holds, sellers are pushing value lower.
  • Value Area gap: When consecutive candles have non-overlapping value areas, it signals strong directional conviction.
  • 80% Rule: If price enters the previous value area and travels 80% through it, there is a high probability it will reach the other side.

Imbalance Highlighting

Imbalance detection identifies price levels where one side of the market (buyers or sellers) dominated aggressively.

What Imbalance Means

An imbalance occurs when the volume ratio between the buy and sell side at a price level exceeds a configurable threshold. For example, with an imbalance ratio of 3.0, a level where ask volume is 300 and bid volume is 90 would be flagged as a buy imbalance (300 / 90 = 3.33).

Imbalance Settings

SettingDescriptionDefault
Imbalance RatioMinimum ratio to trigger highlighting1.5
Use Trade ImbalanceCompare trade counts instead of volumeOff

Diagonal Comparison (Bid x Ask Imbalance)

The Bid x Ask Imbalance preset uses the standard diagonal comparison method: each bid level is compared to the ask level one tick above it. This reflects real market mechanics — sellers hitting the bid at price N are “against” buyers lifting the ask at price N+1.

Stacked Imbalances

When three or more consecutive price levels show imbalance in the same direction, this is called a stacked imbalance. Stacked imbalances are among the strongest signals in footprint analysis:
  • Buy stacked imbalance (ask dominating at 3+ consecutive levels): Strong aggressive buying. The levels where the imbalance begins often act as future support.
  • Sell stacked imbalance (bid dominating at 3+ consecutive levels): Strong aggressive selling. The top of the imbalance zone often acts as future resistance.
Imbalance detection with very low thresholds (below 2:1) can produce excessive highlights that reduce readability. A ratio between 3:1 and 4:1 works well for most instruments and timeframes. For high-volume symbols like BTCUSDT, even 1.5:1 can be meaningful due to the large absolute volumes involved.

Color Schemes and Gradation

Cluster cells use color gradation to encode the magnitude of values. Higher values receive more intense colors, allowing you to spot significant levels without reading every number.

How Gradation Works

The gradation system calculates a color intensity for each cell based on its value relative to the candle’s maximum. The cell with the highest value (the POC) receives maximum intensity, and all other cells are scaled proportionally.

Gradation Settings

SettingDescriptionDefault
Cluster Gradient EnabledEnable/disable color gradationOn
Gradient Color 2Secondary gradient endpoint colorTheme default
Disable Opacity in GradientUse only brightness changes, no transparencyOff
Min Color ValueMinimum value to begin coloring cells0

Proportion Type

The proportion type determines the reference point for the gradation scale:
TypeReferenceEffect
CandleMaximum within each candleEach candle has its own color scale (default)
OwnA fixed custom valueAll candles share the same color scale
Using Candle proportion means each candle’s gradation is independent — the POC is always the brightest cell. Using Own proportion with a fixed reference value means cells across different candles are directly comparable, but some candles may appear entirely bright or entirely dim depending on their volume.

Delta Colors

For presets that color by delta:
OverrideControlsDefault
deltaPosPositive delta (net buying)Theme green
deltaNegNegative delta (net selling)Theme red
bidBid side colorTheme bid color
askAsk side colorTheme ask color
strongBidHighlighted strong bid levelTheme strong bid
strongAskHighlighted strong ask levelTheme strong ask

Cluster Settings Panel

Open the cluster settings panel by clicking the gear icon in the toolbar when any cluster preset is active.

Text and Readability

SettingDescriptionDefault
Show ValuesDisplay numeric values in cellsOn
Minimize ValuesAbbreviate large numbers (e.g., 1.2K, 5.3M)Off
Round ValuesDecimal places for cell values0
Font SizeOverride font size (0 = auto-detect based on zoom)0 (auto)
Font WeightOverride font weight (0 = auto)0 (auto)
Text AlignHorizontal alignment within cellsCenter
Text ShadowAdd shadow behind text for readabilityOff
Text StrokeAdd outline around text charactersOff
Volume UnitDisplay in base currency or quote currencyBase

Cell Appearance

SettingDescriptionDefault
Border TypeCell border style: box, top, noneBox
Border WidthCell border thickness in px1
Corner RadiusCell corner rounding1
Min Bar WidthMinimum pixel width before cells collapse22
Direction MarkerSmall colored bar at the cell edge indicating delta directionOff

Value Filters

Value filters allow you to highlight or dim cells based on specific criteria. You can create multiple filters, each targeting a different data type:
Filter Data TypeWhat It Filters
VolumeTotal volume at the level
TradesNumber of individual trades
BidBid-side volume
AskAsk-side volume
Delta+Positive delta values
Delta-Negative delta values
Each filter specifies a minimum and maximum value, a color, and where it applies (everywhere, body only, or shadows only). Filters are processed in order, and the last matching filter determines the cell’s highlight color.
Use value filters to create a “significance threshold” — for example, highlight any level with volume above 500 contracts in bright yellow. This immediately draws your eye to the levels that matter, even on a busy chart.

Reading Cluster Charts Effectively

A practical approach to cluster analysis:
1

Identify the POC

The highest-volume level anchors your reading of the candle. This is where the market agreed on price most actively.
2

Check delta direction at the POC

Does the delta at the POC agree with the candle direction? If a bullish candle has negative delta at the POC, it may signal exhaustion — sellers are absorbing the buying pressure.
3

Scan for imbalance stacks

Three or more consecutive imbalance levels in the same direction suggest strong institutional intent. Mark these zones on your chart as potential future support or resistance.
4

Compare value areas across candles

Overlapping value areas indicate balance and range-bound conditions. Shifting value areas (each candle’s value area higher or lower than the previous) confirm a trending market.
5

Check trade count vs volume

If a level has high volume but low trade count, it means large orders are present (institutional). High trade count with low volume suggests many small participants (retail noise).

Performance Considerations

Cluster charts are the most resource-intensive rendering mode. Each candle can contain dozens to hundreds of individual cells, each requiring text rendering and color calculation.
FactorImpactRecommendation
Tick sizeSmaller = more rows per candleIncrease for higher TFs
TimeframeLower TFs = more candles on screenUse 5m+ for extended analysis
Profile styleProfile bars are more expensive than plain textUse plain for maximum performance
Number of chartsEach cluster pane uses significant GPULimit to 1-2 cluster panes in multi-chart layouts
If you need to monitor multiple symbols simultaneously, use candlestick mode for overview panes and reserve cluster mode for the one or two charts where you are actively analyzing order flow.