Every liquidation level we publish is graded against what actually happened, using independently verified cross-exchange data. No competitor in this category publishes its own accuracy. This page is the measured historical record — not a forecast.
| Symbol | Status | Grade | precision@N (all) | last 7d | Wilson 95% low | × vs random | windows (n) | events (n) | span |
|---|---|---|---|---|---|---|---|---|---|
| Loading… | |||||||||
Price zones where leveraged positions are forced to liquidate — the "magnets" shown in the product.
On a recurring cycle the model's top-10%-ranked zones are replayed against the liquidation events that actually occurred over the trailing 7 days. Picking zones at random would score ~10% by construction — that is the baseline every number on this page is compared against. Scoring windows with fewer than 10 events are excluded as noise.
The liquidation event feed itself is reconciled every 5 minutes against sources we don't control: open-interest signatures on Bybit and Binance, direct REST liquidation records on OKX, and a structural expected-zero check on Deribit. Verification ran for 7 days in shadow before any badge went live, and has run continuously since 2026-05-10.
When a symbol's live precision drops below 0.25, its predictions are switched off automatically and it appears on this page as auto-disabled until it re-qualifies. Nothing is removed from the ledger; there is no survivorship editing.
A historical measurement, not a forecast. Past precision does not guarantee future precision, and nothing here is financial advice.