Footprint

The buy/sell volume profile of a bar, bucketed by price. Decomposes where inside a bar volume printed and which side was the aggressor there — the detail a single OHLCV bar hides.

Quick reference

Item	Value
Family	Microstructure
Input type	Trade — an executed trade with an aggressor side
Output type	FootprintOutput — variable-length list of price buckets
Output range	per-bucket volumes ≥ 0
Default parameters	tick_size required (finite, > 0)
Warmup period	1
Interpretation	Intrabar volume profile / order-flow clusters

Formula

bucket  = round(price / tick_size)
ask_vol[bucket] += size   for a buy   (lifts the ask)
bid_vol[bucket] += size   for a sell  (hits the bid)

Each update returns the full footprint accumulated since the last reset — one row per touched price bucket, sorted ascending by price. Call reset() at each bar / session boundary. See crates/wickra-core/src/indicators/footprint.rs.

Parameters

Name	Type	Default	Constraint	Description
tick_size	f64	none	finite, > 0	Width of each price bucket.

Inputs / Outputs

Indicator<Input = Trade, Output = FootprintOutput>, where FootprintOutput holds levels: Vec<FootprintLevel { price, bid_vol, ask_vol }>. This is a multi-output, variable-length indicator, so each binding returns a per-bar profile rather than one scalar:

• Python — update(price, size, is_buy) returns a (k, 3) NumPy array with columns [price, bid_vol, ask_vol]; batch(price[], size[], is_buy[]) returns a list of such arrays, one per trade.
• Node — update returns an array of { price, bidVol, askVol } rows; batch returns an array of such arrays.
• WASM — update returns an array of { price, bidVol, askVol } objects (streaming-only).

Warmup

warmup_period() == 1; every update emits the current footprint.

Edge cases

• Sub-tick prices. Trades within the same tick_size band share one bucket.
• Accumulation. The footprint grows until reset(); without resets it spans the whole stream.
• Bad tick size. A non-finite or non-positive tick_size is rejected at construction.

Examples

Rust

use wickra::{Footprint, Indicator, Side, Trade};

let mut fp = Footprint::new(1.0).unwrap();
fp.update(Trade::new(100.2, 2.0, Side::Buy, 0).unwrap());
let out = fp.update(Trade::new(100.7, 3.0, Side::Sell, 1).unwrap()).unwrap();
assert_eq!(out.levels.len(), 2);          // buckets 100 and 101
assert_eq!(out.levels[0].ask_vol, 2.0);   // bucket 100: buy 2
assert_eq!(out.levels[1].bid_vol, 3.0);   // bucket 101: sell 3

Python

import wickra as ta

fp = ta.Footprint(1.0)
fp.update(100.2, 2.0, True)
out = fp.update(100.7, 3.0, False)   # (2, 3) array: [price, bid_vol, ask_vol]
print(out)

Node

const { Footprint } = require('wickra');

const fp = new Footprint(1.0);
fp.update(100.2, 2, true);
console.log(fp.update(100.7, 3, false)); // [{price:100,bidVol:0,askVol:2}, {price:101,bidVol:3,askVol:0}]

Interpretation

A footprint exposes the intrabar auction — the buy/sell volume printed at every price inside a bar, the detail a single OHLCV candle collapses away.

• Point of control. The bucket with the most total volume is where the bar spent its conviction; it often acts as intrabar support/resistance.
• Bid/ask imbalance per level. When a bucket's ask (buy) volume dwarfs its bid (sell) volume, buyers were the aggressors there. A stack of buy-dominant buckets that fails to push price higher is absorption; the same on the lows is exhaustion — turns the open/high/low/close cannot reveal.
• Shape. A balanced, bell-shaped profile is a fair two-way auction; a thin, one-sided profile is a trend bar that ran with little opposition.

Reset per bar to get the classic bid/ask footprint chart; let it run for a session-long volume-by-price profile.

Common pitfalls

• Tick-size choice. Match tick_size to the instrument's real tick. Too fine fragments volume across empty buckets; too coarse blurs the profile into a few fat bars and hides the point of control.
• Forgetting to reset(). Without a per-bar reset the footprint accumulates over the whole stream — useful as a session profile, misleading if you wanted one bar.
• Aggressor flag required. The bid/ask split is only meaningful if trades carry a reliable aggressor side; infer it (tick / quote rule) if your feed does not provide one.

References

• J. Peter Steidlmayer and Steven B. Hawkins, Steidlmayer on Markets: Trading with Market Profile, 2003 — the price-by-volume auction lineage footprint charts extend.
• Footprint (bid/ask volume) charts are an order-flow charting convention popularised by trading platforms (e.g. MarketDelta, Bookmap) rather than a single academic result; the implementation here follows the standard bid-on-sell / ask-on-buy bucketing.

See also

• SignedVolume — the per-trade signed size each bucket accumulates.
• CumulativeVolumeDelta — the same flow as a single running line.
• TradeImbalance — rolling buy/sell imbalance, without the price dimension.
• Indicators-Overview — the full taxonomy.