Module 7: Perpetual Futures
Liquidation Heatmaps and Market Magnets
Chapter 10 of 14 · about 5 minutes
10.1 The bright yellow band
Mint opened a liquidation heatmap for the first time. It looked like a price chart overlaid with glowing bands of colour, some faint purple, some bright yellow. A thick yellow band sat just below the current price.
"What am I looking at?" Mint asked.
"An estimate of where leveraged traders would be forced out," said Hash. "Bright areas show where a lot of liquidations might be waiting. Many traders believe price is drawn toward those areas. Sometimes it is. But you need to understand how the map is drawn before you trust it."
10.2 What a liquidation heatmap shows
A liquidation heatmap displays estimated liquidation levels across a range of prices, over time. Brighter colours mean more estimated liquidations at that price.
| Feature | What it represents |
|---|---|
| Horizontal axis | Time |
| Vertical axis | Price |
| Colour intensity | Estimated value of positions that would be liquidated at that price |
| Bands disappearing | Positions closed, or price reached the level and liquidations occurred |
Bands below the current price mostly represent long liquidations. Bands above mostly represent short liquidations.
10.3 How heatmaps are estimated
Exchanges don't publish every trader's liquidation price. So heatmap providers estimate them. A typical approach:
- Track when and at what price open interest increases (Chapter 9), indicating new positions.
- Assume a distribution of leverage for those positions, for example some at 10x, some at 25x, some at 50x.
- Calculate where those positions would be liquidated, using formulas like those in Chapter 8.
- Remove estimated liquidations when OI falls or price passes through the level.
A simplified example. OI increases by $500 million while BTC trades around $100,000. The model assumes this is split across leverage levels.
| Assumed leverage | Share of new OI | Estimated long liquidation level |
|---|---|---|
| 10x | 40% ($200M) | ~$90,500 |
| 25x | 35% ($175M) | ~$96,500 |
| 50x | 25% ($125M) | ~$98,500 |
Each band is an estimate built on assumptions: which side opened the positions, what leverage they used, and whether they've since added margin or closed. Different providers make different assumptions, so their heatmaps differ.
"So it's a model, not data," said Node.
"It's a model built from real data," said Hash. "Open interest is real. The liquidation levels are educated guesses."
10.4 The magnet idea
Large liquidation clusters can act like magnets for price, for a few reasons:
Forced orders provide liquidity. When liquidations trigger, they create sudden market orders. Large traders who want to buy or sell size may push price toward clusters to trade against that liquidity.
Cascades accelerate moves. Once price reaches a cluster, forced selling or buying pushes it further (Chapter 8).
Stops cluster nearby. Many traders place stops near obvious levels, which are often close to liquidation clusters (Module 4, Chapter 5).
After a large cluster is cleared, the "fuel" is used up, and price often stabilises or reverses.
10.5 Reading a heatmap
Illustrative snapshot, BTC at $100,000:
| Level | Estimated liquidations | Side |
|---|---|---|
| $104,000 | ~$500 million | Shorts |
| $102,000 | ~$150 million | Shorts |
| $98,500 | ~$200 million | Longs |
| $96,500 | ~$800 million | Longs |
Chapter 10 · Reading a liquidation heatmap
A price chart with BTC trading around $100,000 over time, overlaid with horizontal bands of estimated liquidations whose brightness reflects size. Above the price: about $500 million of short liquidations at $104,000, marked as a possible magnet, and about $150 million at $102,000. Below the price: about $200 million of long liquidations at $98,500 and about $800 million at $96,500, the largest band, also marked as a possible magnet. A note says the bands are estimates built from open-interest changes and assumed leverage, not exact data.
Mint's reading: "The biggest cluster is $800 million of long liquidations at $96,500. If price drifts down, there's a risk it gets pulled toward that level and cascades through it. Above, $500 million of shorts at $104,000 could fuel a squeeze if price rises."
"Good," said Hash. "Now, what does it not tell you?"
Mint thought. "Which one gets hit first. Or whether either gets hit at all."
10.6 Limitations
| Limitation | Effect |
|---|---|
| Estimated leverage | Real leverage distribution is unknown |
| Unknown direction | Rising OI doesn't reveal which side was the aggressor |
| Hedged positions | Many positions are hedged and won't cause forced selling |
| Margin changes | Traders add margin or reduce size, moving their liquidation prices |
| Exchange coverage | Heatmaps cover only some exchanges |
| Model differences | Providers use different methods and show different pictures |
| Self-fulfilling and self-defeating | Widely watched clusters can attract price, or be avoided by traders anticipating them |
Heatmaps also cover different timeframes. A short-term model may show clusters from the last day; a longer one, from the last month. Bands that appear on several models and timeframes are generally more significant than those on one.
10.7 Using heatmaps sensibly
Hash suggested using heatmaps mainly to manage risk:
- Check where your own liquidation and stop would sit. If your stop is just beyond a major cluster, it may be swept during a cascade. Consider placing it outside the cluster, or reducing size.
- Be cautious about entering in the direction of a nearby large cluster. Entering a long just above a large long-liquidation cluster puts you in the path of a potential cascade.
- Combine with structure. A large cluster just below strong support (Module 4, Chapter 5) suggests a sweep of the level and possible fakeout is plausible.
- Combine with funding and OI. Clusters on the crowded side, with high funding, are more vulnerable.
- Watch what happens after clusters clear. Reversals often follow cascades, as forced selling ends.
"So heatmaps show where the danger zones might be," said Mint. "Not where price must go."
"That's the right way to use them," said Hash.
10.8 Try it yourself
Open a free liquidation heatmap for BTC. Identify the largest estimated clusters above and below the current price. Write down whether each is long or short liquidations. Check back after a few days: did price reach either cluster, and what happened afterwards? Then compare the same timeframe on a second provider's heatmap, if available.
Key takeaways from this chapter
- Liquidation heatmaps estimate where leveraged positions would be liquidated, shown as bright bands on a price chart.
- Bands below price are mostly long liquidations; bands above are mostly short liquidations.
- Heatmaps are models built from open interest changes and assumed leverage, not exact data.
- Large clusters can act as magnets because they provide liquidity, accelerate cascades, and sit near clustered stops.
- Limitations include unknown leverage and direction, hedging, margin changes, partial coverage, and model differences.
- Use heatmaps mainly to place stops and avoid danger zones, combined with structure, funding, and OI.