Footprint chart: from the CBOT pit to crypto perpetuals
How the footprint chart was invented, who owns the trademark, why it almost died with MarketDelta in 2018, and what it shows about order flow on modern crypto perps.
A candlestick tells you four numbers. A footprint chart tells you everything that happened between them.
That is the whole pitch, and it is why the footprint has survived four decades of regime change: from open outcry at the Chicago Board of Trade, to electronic futures, to crypto perpetuals running 24/7 on Binance and Bybit. The tool has been trademarked, litigated, bankrupted, rebranded and ported across asset classes. The underlying idea — show me the bid–ask volume at every price the bar traded through — has never stopped being useful.
This article is the long version of where the footprint came from, who actually invented it, why competing software vendors had to call theirs something else, and what it does and doesn't tell you in modern crypto microstructure. If you trade order flow on perps, this is the canon.
Before the footprint: why pit traders had an unfair edge
To understand why the footprint chart matters, you have to understand the asymmetry it was built to fix.
In the open-outcry pits of the Chicago Board of Trade (CBOT) and Chicago Mercantile Exchange (CME) in the 1970s and 1980s, locals (traders standing in the pit with their own capital) could literally see who was hitting bids and who was lifting offers. They could hear the brokers, watch the body language, and clock which large filling firms were active. Off-floor traders had none of this. They had a tape of trades and a price chart. They knew where the market went, but not how it got there.
A 1985 internal CBOT memo described this gap bluntly: floor traders had access to information that retail customers literally could not buy at any price. The exchange had a problem — and an unusual member willing to fix it.
J. Peter Steidlmayer and the Market Profile (1981–1985)
J. Peter Steidlmayer joined the CBOT in 1963 and traded as a local for the next two decades. From 1981 to 1983 he served on the CBOT Board of Directors, and during that tenure he pushed the exchange to release something radical: structured intraday volume data, organised so that off-floor traders could see what the pit saw.
The result was the Market Profile, a graphical representation that grouped price-by-time activity into a bell-shaped distribution showing where the market had spent the most time during a session. It went live in CBOT educational seminars in 1983 and reached the public in 1985, packaged as a product called CBOTMP1.
Bundled with the Market Profile was the Liquidity Data Bank (LDB): end-of-day clearings that broke trading volume down by class of participant: (1) locals, (2) commercials, (3) members filling for other members, and (4) members filling orders for the public. For the first time, off-floor traders could see specific volume traded at every individual price during the session, and roughly who had been responsible for it.
The LDB never quite delivered on its biggest promises. Even contemporary reviewers in Technical Analysis of Stocks & Commodities called it "big promises, small deliveries." But conceptually, it was the unlock. Once you have seen "volume at price" broken down by participant, you cannot unsee it. Every microstructure tool that followed, including the footprint chart, is downstream of Steidlmayer's bet that disclosing this data would not destroy markets, but improve them.
The 1990s: electronic markets, broken tooling
When the CBOT and CME began their long migration to electronic trading in the 1990s, the data got better in theory and worse in practice.
In theory, every trade now had a precise timestamp, a precise price, and an aggressor side that could be inferred from whether the trade printed at the bid or the ask. No more squinting at the pit.
In practice, the charting software did not catch up. Traders moving from the pit to a screen got candlestick charts (open, high, low, close) and maybe a volume histogram at the bottom. The richest microstructure data ever produced was being thrown away as it streamed in. For nearly a decade, the question of how to visualise electronic order flow at the price level sat unanswered.
2002–2003: Trevor Harnett and the birth of the Footprint®
Trevor Harnett started his career trading at the Chicago Mercantile Exchange. By the early 2000s, he was building software, and he had a specific obsession: he wanted to see, for every bar, how much volume executed at the bid versus at the ask at each price the bar touched.
In 2002, his company coined the term "delta" to describe the difference between buy-initiated and sell-initiated volume. The language had not previously existed in retail-accessible form because nobody had built a chart that needed it.
In 2003, MarketDelta launched in Chicago at the CME, shipping software with a new chart type: the Footprint® chart. Each bar was no longer one candle; it was a column showing every price the bar traded through, with bid-side volume on the left and ask-side volume on the right, and a delta number summing the imbalance. In one image, you could see whether the move up had been driven by aggressive buyers lifting offers, or by sellers exhausting themselves and lifting price by walking out of bids.
It was the first chart designed natively for the electronic era. In 2005, Michael Burkhart, Trevor Harnett and Richard Malato filed a patent application titled "Method and apparatus for providing trading information" covering the technique. And the name "Footprint" was registered as a trademark of MarketDelta.
Why your platform calls it something else
The trademark is the reason you have never seen another vendor sell a "Footprint chart" outside MarketDelta's own product. Sierra Chart calls theirs Numbers Bars. NinjaTrader calls theirs Volumetric Bars (sometimes "Order Flow +"). ATAS calls theirs Cluster charts. Bookmap built around heatmap visualisation of resting liquidity but added cluster-style volume modes. TradingView calls them Volume Footprint charts under licence arrangements.
Functionally these are all the same animal: a bar-level breakdown of executed volume at each price, split by aggressor side. The naming fragmentation is a small but real cost: beginners often think the tool exists in five different categories instead of one.
How a footprint bar is constructed
For each candle on a chosen timeframe (commonly 1 to 15 minutes, or tick-based / volume-based bars):
- Bucket every executed trade by the price level it occurred at.
- Classify each trade as bid-initiated (sell market order hit the bid) or ask-initiated (buy market order lifted the offer).
- Sum the volume on each side at each price.
- Render the column vertically, price by price, with left/right or top/bottom layout for bid/ask volume.
- Compute delta (the net buy-minus-sell volume) for each price and for the bar as a whole.
What you end up with is a candle whose body is no longer a rectangle but a stack of paired numbers. Every micro-decision the bar made is exposed.
What the chart actually shows (and what it doesn't)
The footprint is most useful when read for patterns, not raw numbers. The vocabulary that matured in the 2010s among professional order-flow traders includes:
Absorption. Aggressive buying (or selling) hits a price level repeatedly, but price refuses to move. Someone is absorbing the aggression, passively filling at the level with size that the aggressors cannot exhaust. Absorption at a swing low after a sell-off is often the first signal that the auction has changed character.
Stacked imbalances. When the ratio of bid-side to ask-side volume at consecutive prices crosses some threshold (commonly 3:1 or 4:1) for three or more prices in a row, you have a stacked imbalance. These often mark where institutional flow is stepping in and act as a magnet for future retests.
Unfinished auctions. Healthy auction endings print volume on one side only at the extreme of the bar: buyers exhausted themselves at the top tick, no sellers met them. An unfinished auction shows volume on both sides at the extreme. The market tends to return to finish what it started. Unfinished business is a level to mark.
P-shape and b-shape distributions. A vertical volume distribution that looks like the letter "P" (heavy at the top, thin at the bottom) indicates short-covering rallies: sellers capitulated, but new buyers did not commit. A "b" shape is the mirror: long liquidation. These shapes are auction-theory artefacts that the footprint surfaces visually.
Delta divergence. Price makes a new high, but the cumulative delta of the move does not. The push was thin. This is one of the most reliable single-bar signals the footprint produces, and one of the most misused.
Iceberg orders. Large hidden orders that refill at the same price after each fill. On a footprint chart, an iceberg often shows up as a single price level with abnormally large bid (or ask) volume that should have been wiped out and wasn't. The signature is the persistence of fills at a level the book seemed to have already cleared.
No single pattern is a setup on its own. The bars that pay are the ones where two or three patterns align at the same price: absorption + stacked imbalance + a P-shape distribution at a prior unfinished auction level is a different read from any one of those signals alone.
The 2018 collapse: MarketDelta files Chapter 7
On October 18, 2018, MarketDelta, LLC filed for Chapter 7 bankruptcy in the Northern District of Illinois (case number 1:18-bk-29315). The filing followed several years of defending the company against litigation. By 2013 MarketDelta had already licensed Trading Technologies' entire 400+ patent portfolio covering electronic trading — a sign of how aggressively the IP landscape around order-flow visualisation was being contested.
The company that invented the Footprint® chart went out of business. The chart itself did not. By the time MarketDelta wound down, the technique was already a global standard, embedded in a dozen competing platforms, and migrating fast into a market segment that nobody at the CBOT in 1983 could have predicted.
Crypto perpetuals: the second life of the footprint
Bitcoin futures launched on CME in December 2017. Centralised crypto exchanges had been running margin and inverse swap products for years before that, but the perpetual futures era (Binance, Bybit, OKX, dYdX, Hyperliquid) turned crypto into the most liquid intraday playground on earth. By 2025, BTC perpetuals alone traded several hundred billion dollars of notional per day across the top venues.
Crypto perps are, in microstructure terms, futures markets with extra moving parts: a funding rate that drags the perpetual back toward spot, no settlement date, fragmented liquidity across half a dozen major exchanges, and a tape that runs 24/7 with no daily auction reset.
The footprint chart turned out to be near-perfectly suited to this environment. Platforms like ExoCharts, Bookmap, TradingLite, TensorCharts, Cignals and ATAS (originally a futures tool) all extended into crypto and shipped footprint-style visualisations on BTC, ETH, SOL and the rest of the top perps. A 2024 academic study in the Journal of Financial Markets (Anastasopoulos, Gradojevic, Liu, Maynard & Tsiakas) found that order flow has strong, economically valuable out-of-sample predictive power for cryptocurrency returns, particularly when combined with non-linear machine-learning models. Exactly the regime the footprint chart was designed for.
There is one significant nuance, though, and it matters more in crypto than it did in futures: the aggressor side classification is only as good as the exchange's tape. On a regulated futures exchange, the bid/ask classification is unambiguous. On a crypto exchange, you are dependent on the venue's WebSocket feed correctly tagging the taker side. You only see that venue. A footprint on Binance BTCUSDT is not a footprint on global BTC liquidity; it is a footprint on Binance's slice of it. Liquidation cascades frequently fire on one exchange seconds before another, and a single-venue footprint will tell you about the cascade only after it has already washed through. Multi-exchange aggregation is non-trivial and most retail tools do not do it cleanly.
MarketTrace is one implementation that does: a per-bar footprint across BTC, ETH, SOL, BNB, XRP and DOGE perps, aggregated across Binance, Bybit, OKX and Hyperliquid in the same column. It exists specifically to address the venue-specificity caveat above. Free, no paywall, with AVWAP anchors and bar sizes from 30 seconds up.
Where the footprint sits in the modern microstructure stack
The footprint is one of four lenses on the same underlying flow. None of them are sufficient alone; together they cover most of what an intraday trader can know about a perp without paying for L3 data:
- Footprint / cluster chart. Executed volume, bid vs ask, at every price the bar touched. Best for identifying absorption, stacked imbalances and exhaustion at specific levels.
- Order book imbalance (OBI). Ratio of resting bid liquidity to resting ask liquidity in the top N levels. Best for short-horizon directional pressure and detecting spoofing patterns.
- Cumulative volume delta (CVD). Running sum of aggressor-side volume across a session. Best for confirming or fading the trend with the question "is this move actually being bought, or is it short-covering on thin participation?"
- Funding rate. The periodic payment between longs and shorts that anchors a perp to spot. Best for positioning context: who is paying to hold the trade.
A trader using only the footprint is reading a single page of a four-page book. The page is the most detailed of the four, but the others provide the context that tells you whether the patterns on the footprint are operating in a normal regime or an extreme one.
What the footprint will not do
It will not give you a setup. It is a microscope, not a strategy.
It will not work in thin liquidity. In illiquid alt-perps the bars are too sparse for the patterns to mean anything; a single 5-BTC market order looks like institutional absorption when it is actually a bored retail trader.
It will not survive disconnected use. Footprint patterns are conditional probabilities: they shift behaviour depending on whether the market is trending or in balance. Reading footprints without first knowing which regime you are in is how traders convince themselves they have an edge and then donate it back.
It will not replace the order book. Executed flow is what did happen. Resting liquidity is what is about to happen. A complete order-flow read requires both.
The throughline
For forty-three years, the same idea has kept resurfacing in a new technical wrapper: show traders what the pit could see. Steidlmayer's Market Profile did it for floor data in 1985. Harnett's Footprint did it for electronic futures in 2003. Bookmap and ExoCharts and the cluster-chart crowd did it for crypto perps in the late 2010s. The naming changed every time. The chart's owner went bankrupt. The trademark fragmented across a dozen platforms. The underlying observation, that price is a summary statistic and the real information lives in the volume distribution underneath it, has not aged a day.
If you trade perps and you have never seriously read a footprint, that is the gap to close. Not because it is a holy grail. Because it is the closest a screen can get to standing in the pit.
Sources
- Background & History: Traded Volume & the Footprint® Chart — Emoji Trading
- Market Profile — Wikipedia
- V.8:3 (121-123): Liquidity Data Bank: Big Promises, Small Deliveries — Technical Analysis of Stocks & Commodities
- About Trevor Harnett & Methodica Capital
- Method and Apparatus for Providing Trading Information — USPTO Patent Application 20050080710
- MarketDelta Licenses Trading Technologies' Entire Patent Portfolio — Markets Media
- MarketDelta, LLC Bankruptcy 1:18-bk-29315 — Illinois Northern Bankruptcy Court
- Order Flow and Cryptocurrency Returns — Journal of Financial Markets, ScienceDirect
- CME Group Self-Certifies Bitcoin Futures to Launch Dec. 18, 2017