Liquidation Clusters Explained: How to Spot Liquidity Magnets

If you trade crypto futures, you have probably watched price drift toward a level, accelerate into it, and then snap back. Often that level is a liquidation cluster — a price where a lot of leveraged positions are set to get forced out. This guide explains what clusters are, how they form, and how to read them on a liquidation map without fooling yourself about what the data actually shows.

What a liquidation cluster actually is

A single liquidation level is the price at which one leveraged position can no longer cover its margin and gets force-closed by the exchange. On its own, one level means nothing. A cluster is what you get when many liquidation levels stack up at or near the same price.

Think of it as a pile of pending forced orders waiting at a price. If 42,000 is where hundreds of long positions would be liquidated, then a move down to 42,000 triggers a wave of forced selling all at once. That concentration is what makes clusters worth watching — not the individual levels, but the density of them at one spot.

On MarginPad's BTC liquidation map, clusters show up as bright, tall bands. The taller and brighter the band, the more leverage is estimated to be sitting at that price.

How clusters form from open interest

Clusters are a side effect of leverage piling up. When open interest rises, it means new leveraged positions are being opened — fresh contracts that did not exist before. Every one of those positions carries an implied liquidation price that depends on entry price and leverage.

Here is the key idea: traders do not all use the same leverage. A model spreads new positions across a leverage distribution — some at 5x, some at 10x, 25x, 50x and higher. Each tier implies a different liquidation distance from the entry price:

• A 10x long liquidates roughly 10% below entry.
• A 25x long liquidates roughly 4% below entry.
• A 50x long liquidates roughly 2% below entry.

When lots of positions open around the same entry zone, their liquidation prices bunch together at predictable distances. The high-leverage liquidations sit close to current price; the lower-leverage ones sit further away. Stack enough of them and you get a cluster. If you are fuzzy on the mechanics, our leverage explained post walks through the math.

Why clusters act as liquidity magnets

A liquidation is a forced market order. A long liquidation is forced selling; a short liquidation is forced buying. A dense cluster is therefore a pool of guaranteed flow sitting at a known price — and markets tend to gravitate toward known pools of liquidity.

There are two honest reasons clusters pull price:

• Liquidity needs liquidity. Large players who want to fill big orders need counterparties. A cluster is exactly that — a stack of forced orders waiting to be matched.
• Cascades are self-reinforcing. When price reaches a cluster, the forced orders push price further in the same direction, which triggers the next set of liquidations, and so on.

This is why traders describe clusters as magnets. It is not magic and it is not guaranteed — it is the simple gravity of concentrated forced flow.

Long clusters below, short clusters above

Direction matters, and it follows directly from how leverage works:

• Long clusters sit below current price. Longs get liquidated when price falls, so their stacked liquidation levels form a band underneath the market. A drop into that band can trigger a cascade of forced selling.
• Short clusters sit above current price. Shorts get liquidated when price rises, so their levels form a band overhead. A rally into that band can trigger a short squeeze — forced buying that fuels more upside.

Because of this, clusters often behave like soft support and resistance. A thick long cluster below can act as a downside magnet that, once hit and consumed, becomes a springboard. A thick short cluster above can cap a move until price punches through it.

How clusters get consumed

A cluster is not permanent. When price trades through it, the positions there get liquidated and the cluster is consumed — the leverage that was stacked at that price is gone. On the map you will see a bright band fade or disappear after price sweeps it.

This is the single most useful read. A level that was a magnet on the way in is often empty on the way back, because the fuel has been spent. Watching clusters appear, get hit, and vanish tells you where the leverage has already been flushed and where it is still loaded. Our guide on how to read a liquidation heatmap covers this consumption pattern in more depth.

A practical sequence looks like this: open interest climbs, a cluster builds at 42,000, price grinds toward it, the band lights up as liquidations fire, and then it goes dark as the cluster clears. After that, support at 42,000 is far weaker because the forced-buy flow that defended it is gone.

The honest caveat: estimated clusters are a model

This matters, so be clear-eyed about it. MarginPad shows two different things, and they are not the same kind of data:

• Real liquidations are actual events reported by Bybit, OKX and Binance. They happened. There is no guesswork.
• Estimated clusters are a model. They are derived from open-interest changes plus an assumed leverage distribution. They are not exchange order-book data, and no exchange publishes the exact liquidation price of every position.

In other words, the cluster layer is an educated estimate of where leverage is probably stacked, not a guaranteed list of where stops sit. Treat it as a probability map, not a price oracle. It is genuinely useful for spotting zones of concentrated risk — but anyone who tells you a cluster will be hit is overselling a model.

Reading MarginPad's Clusters layer

On the live map, toggle the Clusters layer on top of the real-liquidation feed. A few practical habits:

• Read brightness as density. Brighter, taller bands mean more estimated leverage at that price. Faint bands are thin.
• Check both sides. Note the nearest long cluster below and short cluster above — those are your two most likely magnets.
• Watch open interest alongside it. Rising OI means clusters are building; falling OI after a sweep means they are being flushed.
• Confirm with real liquidations. When the real-liquidation feed lights up exactly where the model predicted a cluster, your read is on solid ground.

Used this way, clusters become a context tool: they tell you where the market's leverage is loaded, which directions carry cascade risk, and which levels have already been cleared.

Liquidation clusters are not a crystal ball — they are a map of where leverage is stacked and where forced flow is likely to fire. Read them as liquidity magnets that can pull price, act as soft support and resistance, and then vanish once consumed. Pair the estimated liquidation map with the real-liquidation feed, keep the model's limits in mind, and you will have a far better sense of where the next move is most likely to accelerate.