A raw liquidation print is a bad signal by itself. Aegium reads forced long and short liquidations against each market's own baseline, then folds that into a cross-sectional view of where pressure is actually stretched.
Bitcoin can print a huge dollar liquidation number on an ordinary day. A smaller coin can print a much lower number and still be in a true local flush. Aegium normalizes the event by symbol, so the question changes from 'is this number big?' to 'is this abnormal for this coin right now?'
The dashboard separates long-flush and short-squeeze pressure, compares the event with recent baseline behavior, and checks whether broader flow confirms or contradicts the liquidation read. That keeps liquidation spikes in context instead of turning them into automatic fade signals.
Liquidation exhaustion is useful when it lines up with relative flow. Aegium is built around that spread: which coins are still sponsored by taker flow, funding and order-book pressure, and which names are being left behind.
No. A liquidation spike can mark exhaustion, but it can also be continuation. Aegium treats it as context and checks whether cross-sectional flow supports the read.
No. Liquidations are one input. The core edge is the relative ranking of coins by flow, funding, order-book pressure and validated book behavior.