Bid/Ask, Delta or Volume: Which Footprint Mode Answers Which Question

A footprint - called a cluster chart on some platforms - breaks a candle into its price levels and shows what traded at each one. The underlying data is always the same: every execution with price, size, and whether it went into the bid or into the ask. What differs between the display modes is purely how that data gets condensed.

Bid/Ask: the raw form

Two numbers per price level - the volume traded into the bid and the volume traded into the ask. This is the most informative view and the most demanding one to read. It's the only mode in which ratios can be judged directly, which also makes it the only workable basis for imbalance detection: an imbalance is nothing more than a ratio between two of these numbers.

Delta: condensed, but not lossless

Delta is ask volume minus bid volume per level - one number instead of two. It reads faster, but it costs information: 500 against 400 and 5000 against 4900 both produce a delta of plus 100. A quiet level and a fiercely contested one look identical in delta. For a quick sense of direction across many candles it remains the more practical view.

Volume: where trading happened, not who did it

The volume mode shows total volume per level with no direction attached. It answers a different question than the previous two: at which prices did business actually take place, which levels did the market accept, and which did it merely pass through? That's the profile perspective inside a single candle. Direction then has to come from another mode.

Highlights are not a separate data source

Imbalance highlighting and similar colouring feel like an additional mode, but they're something else: they apply a rule to the bid and ask numbers and mark the result. Their usefulness therefore depends entirely on the thresholds chosen - the same candle can be covered in markings or completely unremarkable, depending on the settings.

The key point: display is not the calculation basis

For a custom indicator or a bot, the selected mode is irrelevant. The logic always reads the underlying bid and ask values per level, regardless of what's on screen. Problems start when a rule is described in the language of the chosen view - as a green cell, say - instead of in the numbers behind it. A rule set meant to run automatically has to be stated at the data level.

Which mode answers which question

• Who was aggressive, and in what ratio? Only the bid/ask view shows this
• Which way did the imbalance lean, compactly across many candles? That's where the delta view is strongest
• Which price levels did the market accept, and which did it pass through? The volume view answers this
• Where was aggression conspicuously one-sided? That's what imbalance highlighting does, based on the bid and ask numbers

For how those ratios turn into coherent zones, see the article on stacked imbalances in the footprint.

For how delta is accumulated beyond the individual candle, see the article on cumulative delta (CVD).

The modes don't compete with each other, they answer different questions from the same data set. Knowing which question is currently open is what picks the view - and anyone automating an evaluation ends up working one level below it anyway.