The EdgeDepth blog
Market ideas,
checked against the data.
What we tested, what we found and where the evidence stops. Read the takeaway first, then explore the method and results.
Do liquidation cascades rebound? The chart says yes. The count says no.
We wrote the question down before looking, compared cascades with ordinary drops, and caught two results that would have fooled us: an 'eventually' framing and a single market-wide day. Read the articleEarlier articles
Explore research tools →-
Why Binance liquidation counts understate activity
Binance liquidation messages do not count every liquidation. Our capture and cross-venue simulations show why the distinction matters.
Read the article -
Do crowded longs precede a sell-off?
Across 8,637 crowded-long episodes, the 24-hour rate of a 2% fall was close to the comparison rate. Larger falls showed a different pattern.
Read the article -
Does extreme negative funding mark a market bottom?
In this study, 82 of 103 complete episodes closed at least 2% lower after 24 hours. The finding applies to a narrowly defined funding condition.
Read the article
Looking for the full results?Published research reports hold the study definitions, counts and supporting examples.Browse research reports →
Have a market idea to test?
Define the condition, compare outcomes and inspect the evidence.
REVIEW THE DEFINITION BEFORE RUNNING