Liquidations & Liquidation Heatmaps Explained

Liquidations are the forced closing of leveraged positions when traders run out of margin. They are mechanical, automated, and directional. In crypto, where retail leverage is high and order book liquidity is thin, liquidations stack on top of each other and produce some of the sharpest, fastest moves in any market. Reading where liquidations cluster — the liquidation heatmap — is one of the most tactical reads available to a perpetuals trader.

This guide covers how liquidations work, why they cascade, how to read a heatmap, and how to use liquidation analysis as both a defensive tool and an offensive setup.

How a liquidation works

1. A trader opens a leveraged position. They post initial margin (e.g. 10% for 10x leverage). The exchange tracks the position’s mark-price-based unrealized P&L continuously.
2. As price moves against the position, unrealized P&L eats into the margin balance. The exchange computes a liquidation price: the spot or mark level at which margin falls below the maintenance threshold.
3. If price reaches the liquidation level, the exchange forcibly closes the position at market — selling for a long, buying for a short — to recover the borrowed funds.
4. The forced close is added to the order book as a market order. It pushes price further in the direction of the existing move, which can trigger the next liquidation in line.

Why liquidations cluster

Three forces conspire to make liquidations cluster at specific price levels rather than spreading evenly across price.

Round-number bias

Traders open positions and set stops at round numbers ($60,000, $65,000, $70,000). Margin requirements at common leverage ratios (5x, 10x, 25x, 50x) translate those round- number entries into clusters of liquidation prices a fixed percentage away. The result: leverage stops stack at recognizable price bands.

Common leverage choices

Most retail traders use a small number of leverage settings. 10x and 25x dominate. Each leverage ratio places its liquidation level at a roughly fixed distance from entry. Stack many traders entering near the same price at the same leverage and you get a thick cluster of stops at predictable distances from current spot.

Reactive position opening

Crypto traders tend to enter on the same kind of setups — breakouts, reversals at support, squeezes. Reactive entries mean the entry distribution itself is concentrated, which concentrates the resulting liquidation distribution.

Liquidation clustering is not a quirk of measurement. It is the structural shape of crypto leverage. Heatmaps make this visible; the clustering itself is real.

How a cascade forms

A cascade is what happens when one cluster of liquidations triggers the next. The mechanics are pure feedback:

1. Price moves enough to hit the first cluster of leveraged stops.
2. Forced market orders flood the book in one direction, consuming bids (for longs) or offers (for shorts).
3. The order book is thin enough that prices have to move significantly to fill the auto-orders, pushing past the first cluster.
4. The bigger move triggers the next cluster of stops down or up the price ladder.
5. Repeat until either the order book absorbs the flow or the next cluster is far enough away that the cascade stops.

Cascades happen in seconds to minutes. They are most violent when (a) liquidity is thin, (b) leverage is concentrated, and (c) dealer gamma positioning is amplifying rather than damping. All three conditions stacking is the recipe for the sharpest crypto flushes.

How to read a liquidation heatmap

A liquidation heatmap visualizes the leveraged stop distribution across price. The y-axis is price; bands of color show where stops cluster. Brighter bands = larger estimated stop pools.

Clusters above spot

Bands above current price represent short stops. If price moves up to those levels, shorts get squeezed. Tightly stacked bright clusters above are short-squeeze fuel.

Clusters below spot

Bands below current price represent long stops. If price moves down, longs get flushed. Tightly stacked bright clusters below are flush fuel.

Distance from spot

Clusters within 1-3% of spot are immediately tradable. They act as magnets and frequently get hit on minor moves. Clusters 5-10% away serve as multi-day targets that materialize on bigger swings.

Asymmetry

Heavy clustering on one side and thin clustering on the other indicates positioning is one-sided. The path of least resistance is toward the heavier side, because the larger pool of stops has more potential energy.

How traders use the heatmap

• Setting targets. A long entry can target the next short cluster above as a clean exit. The cluster is a known liquidity pool; price tends to magnet toward it.
• Setting stops. Stops should sit just outside meaningful clusters, not just inside. Stops just inside a cluster get run with the cluster.
• Anticipating cascades. When stacked clusters sit close together below spot in negative gamma, the asymmetry suggests a downside flush is the higher-probability path.
• Fading flushes. After a confirmed cascade has run, leveraged longs are out of the book. Combined with extremely negative funding, this often produces an oversold bounce.
• Sizing. Trade smaller into thick clusters because cascades are non-orderly and stop placement is harder to enforce in a fast move.

Liquidation behaviour by regime

• Positive gamma + thin clusters: Quiet. Dealer hedging absorbs flow. Liquidations stay localized.
• Positive gamma + thick clusters: Pinning behaviour. Price drifts toward but rarely breaches the cluster.
• Negative gamma + thin clusters: Trending. Moves extend further than expected but without the violent flush.
• Negative gamma + thick clusters: Cascade conditions. Largest single-day moves in crypto happen here.

Common misconceptions

“Liquidations cause crashes.” They amplify them. The initial move comes from somewhere else — macro, news, dealer flow, exchange exploits. Liquidations turn a normal correction into a flush, but rarely start one from nothing.

“Heatmaps predict the future.” They show where stops sit right now. Stops are dynamic; traders open and close positions continuously. A snapshot is stale within hours during active markets. Use live data.

“Liquidations are manipulation.” Cascades happen mechanically. Manipulation may exist on individual moves, but the structural clustering of stops at common leverage ratios is just retail behaviour, not coordinated action.

Frequently asked questions