MidasFlow
AI
MidasFlow agent hub · Updated June 11, 2026
The Liquidation Tracker maps where leveraged positions die. It estimates clusters of liquidation levels above and below the current price and flags the likely cascade direction — because dense clusters act as magnets (forced orders are guaranteed liquidity) and as fuel (each liquidation pushes price toward the next one). In the consolidated report this becomes the key-levels context for entries and stops.
Every leveraged position has a price where the exchange force-closes it. Those prices are predictable from entry zones and typical leverage brackets, so they cluster — usually just past obvious support/resistance, where most traders place entries and stops. When price reaches a cluster, forced market orders fire regardless of anyone's opinion. That makes liquidation maps one of the few forward-looking structures in the market: the orders are already committed, only the trigger is pending.
Input: derivatives positioning data and observed liquidation events from major venues. Output: nearest significant clusters above/below price and a cascade-risk note, merged into the report's level map. It pairs naturally with the Whale Detector: a wall defending a level just before a liquidation cluster is a very different setup from a naked cluster.
The homepage card shows 85% — a backtest estimate, not audited live results.
TODO (owner): add one anonymized liquidation line from a real bot report.
A chain reaction: price hits a cluster of liquidation levels, forced market orders fire, those orders push price into the next cluster, and so on. Cascades explain why some moves accelerate exactly where leverage was hiding.
Forced orders are guaranteed liquidity. Large players need liquidity to fill size, so dense liquidation zones act as magnets — the market often visits them before the "real" move.
Whichever side is crowded. Stacked long liquidations below price make downside moves fast and springy; stacked shorts above fuel squeezes. The asymmetry matters more than the absolute numbers.
No. Exchanges do not publish position maps, so every heatmap — ours included — is an estimate built from positioning data and observed liquidation events. Treat clusters as zones, not lines.
Avoid parking stops inside obvious clusters — you are volunteering to be someone's liquidity. The risk-management guide covers placement: midasflow.ai/knowledge-base/risk-management.
Run a free analysis — liquidation context included