PARKED · data
Hypothesis
Hidden iceberg orders reveal themselves through repeated same-size refills at a level; detecting them locates strong support/resistance to lean on.
Math
$$ \text{flag level } L \text{ if } \#\{\text{refills}(L)\} \ge k \text{ with size } \approx \text{const} $$
Method
Reconstruct per-level refill patterns from L2 updates; tag persistent hidden liquidity. Limited by public depth-stream snapshotting.
Detection works when full L2 deltas are available, but the public depth stream is snapshotted, not event-by-event, so refill patterns are aliased. Parked pending a true incremental book feed. Promising as a context layer, not standalone.
Reconstructing hidden liquidity needs event-level book updates. Snapshot feeds blur exactly the micro-pattern you are trying to detect.