Command Palette
Search for a command to run...
Comments
Join the discussionNo comments yet. Be the first to comment.
In this article, we explore existing ways to visualize financial markets and share our ideas for how they can be improved to reflect the current landscape. We believe cryptocurrency markets have changed the playing field, making old ways outdated. New exchanges operate 24/7 and provide live public data streams for order books, market trades, and other metrics. A quick glance at CoinGecko’s list of exchanges reveals just how many are operating today including spot markets, perpetual futures (both linear and inverse), delivery futures, and options! Analyzing only price candles and volume bars from a single exchange offers a very limited perspective.
Price 📈
Candlesticks, OHLC Bars, Heikin Ashi and their alternatives are very well known. Candlesticks were a breakthrough two centuries ago, enabling traders to capture structure over longer timeframes. Today, candlestick trading patterns can be formalized and implemented into automated strategies — but they’re unlikely to be profitable. Once a strategy becomes widely known, it’s unlikely to offer any real edge.
The same applies to price-derived indicators like RSI and MACD — they offer no more information than the raw price data they’re based on - typically close values of a chosen timeframe. They probably provided consistent statistical advantage for their creators for some time in the past.
An important point: the key to successful trading is consistency. From a mathematical perspective, consistency means having a positive expected return with controlled risk. Some traders may get lucky opening their first 100x leveraged long and closing it in profit — but that means little in the long run. A win rate above 50% — after accounting for trading fees — is the bare minimum required for consistency. So let’s keep in mind next questions: What events are repeatable? What data representations help us detect them?
Plain Order Book 📚
All modern centralized exchanges operate as bidirectional auctions, featuring bids and asks (the same term bid is used for plain auctions). A dynamic order book and a continuous stream of executed trades are fundamental to how these markets operate. The common terminology includes:
L1 data - a stream of the best bid and ask prices
L2 data - order book snapshots (deltas) with specified frequency and depth
L3 data - a stream of all order book changes, including placements, updates, and cancellations.
Most crypto exchanges offer L2 data publicly, though the depth of order books may vary. Traditional exchanges (stocks, forex) often still charge for access to this data — even trade volume was considered proprietary not long ago.
Here’s the central point of this article: price moves because of what happens in the order book. Limit orders act like physical walls. If there’s a multi-million-dollar ask at 100,000 on the BTC/USD pair, price won’t move above it until that wall is either canceled or filled. Once the wall is removed, price can move freely between levels.
Price movements may be triggered by macro events (like BTC ETF approvals) or just hype on Crypto Twitter about a trending coin, whatever people or algorithms do their actions translate to placed or cancelled orders. These orders are what’s quantifiable — and they’re what ultimately drive price movements.
The standard visualization above is a direct carryover from traditional markets into crypto. But the pace of modern markets has changed dramatically. Screens are filled with flashing numbers — but processing that much data in real time is difficult for humans. Order books only show the current state — without any historical context. Market trades simply roll in one by one, offering little sense of broader activity. In traditional markets, the pace was slower and it was easier to process what was happening.
Fortunately, there are better ways to visualize order book changes and market trades. Let’s explore some of them.
Heat maps 🔥
A heat map is a common way to visualize 3D data on a 2D surface. In this context, the X-axis represents time, the Y-axis represents price, and color intensity indicates the size of limit orders at each price/time point. We believe TradingLite was the first to bring heat maps for crypto exchanges to the web — and to make them visually stunning. Heat maps make it significantly easier to interpret market data. However, over time, we’ve come to recognize several limitations:
Color intensity can highlight levels, but comparing intensities across time or price is not intuitive.
Today, traders need to observe multiple exchanges or market types (like spot and perpetual futures), but heat maps are not composable. Also, spot and perpetual futures markets have very different ranges of orders, so just summing them together doesn’t make much sense.
Heat maps don’t distinguish between canceled and executed orders.
Aggregating heat maps remains an unsolved problem. There’s been a longstanding feature request for aggregated heat maps on TradingLite’s feedback board. Technically, we think it’s feasible to implement — the real challenge is making them intuitive and useful. Several approaches have been proposed to address the lack of market order visualization on heat maps. Let’s take a look at them.
Volume Bubbles 🫧
Bookmap visualizes executed orders as bubbles, where the size of each bubble corresponds to the trade size. While it’s a step forward, correlating bubble sizes with heat map intensities isn’t simple or intuitive.
Histograms 📊
TradingLite extends the basic market trade tape with a histogram that includes filtering options. The same issue remains: limit orders appear as heat map intensities, while executed orders exist separately on the histogram.
3D 🧊
There have even been attempts to create 3D representations. In fact, we find them more convenient than heat maps — sizes are intuitive, and their visual form is naturally additive. But how can we represent data from multiple markets? Stacking columns on top of each other would just obscure the lower layers. We experimented with these ideas, but for now, we’ve concluded that it may be too complex for practical use. That said, we may revisit 3D visualizations in the future.
Depth Charts 🌊
The X-axis represents price, and the Y-axis shows the cumulative size of limit orders at each price level. These charts have two main limitations: they show no historical context and no executed orders.
It’s relatively easy to aggregate data from multiple exchanges on these charts — either by summing limit orders or layering them visually.
Okotoki has implemented this kind of aggregation brilliantly.
Tradable Depth of Markets (DOM) 🕳️
From a trade execution standpoint, Tiger.Trade offers one of the most comprehensive solutions:
Some form of historical data
Executed orders represented as bubbles, with sizes labeled
An interactive order book for placing trades directly
However, the same limitations mentioned above still apply:
Color intensity alone is insufficient — you still need to read the numbers
Executed and limit orders are shown in separate visual spaces
History is limited
Footprint charts 🦶
Footprint charts are typically offered by platforms specializing in order flow analysis. For example, the screenshot above is from ExoCharts. These charts show how trades occurred within each candlestick. They reveal important information about imbalances — which can provide a real edge in the market. However, from the perspective of our original topic, they also have limitations:
Color intensities aren’t easily measurable on their own — you still need to read the actual numbers. Horizontal bars can help, but only to a degree.
Footprint charts don’t show how executed orders interact with the limit order book.
A better way? ✨
About a year ago, we came across an article that completely changed our perspective.
As we dug deeper, we discovered that this method is also used by proprietary trading firms. The screenshot below is taken from this video.
When we started experimenting with this visualization style, we realized how naturally it ticks all the boxes.
There’s clear historical context
Sizes are intuitive and visually measurable — x pixels = y units
You can overlay multiple exchanges to compare them directly or sum their order sizes for a combined view
Executed orders can be naturally integrated into the same space (we'll show how on our screenshot)
Footprint charts appear there just naturally. Footprint bars become vertical and share the same space as order sizes
Moreover, volume bars, CVD, OI are all measurable in the same units and naturally fit into the same space
Conclusion / Vision 🎯
Here’s what we’ve built: executed orders clearly reveal how they cut through the limit order book.
Master the Markets by Tom Williams (2005) already feels like an ancient manuscript — but its core principle, “look where the volume is,” remains just as relevant today. Trade sizes, how limit orders interact with market orders is what is driving market prices.
Traditional charts use a uniform price scale, which forces trade sizes into compressed formats — like horizontal bars or bubbles. But if trade sizes are what truly drive the market, why not flip the model — visualize a space built on uniform sizes, and let prices play a secondary role?
That’s exactly what we’re building — and we already support top crypto exchanges. Many features emerge organically within this system. We’re continuously exploring the best ways to surface meaningful data. Eventually, we plan to support trading directly from the chart — similar to platforms like Tiger.Trade.
We’d love to hear your thoughts. Let’s collaborate on better ways to analyze and interact with modern financial markets.