Glossary

Crypto liquidation and derivatives glossary

Every term we use on the Money Map and in our guides, defined in plain English and written for traders. The validation terms have their own pages too, because how a number is measured matters as much as the number: our method is described on the proof page.

A

B

  • Basis

    The gap between a futures or perp price and spot for the same asset.

C

  • Cascade fuel

    How much force-closable leverage sits within reach of price, ready to accelerate a move.

  • Contango and backwardation

    Futures curve above spot is contango; below spot is backwardation.

  • Crowding

    Leverage concentrated on one side, so that side has something to unwind.

D

  • Delta

    How much a position's value moves per unit move in the underlying.

F

  • Forced liquidation

    A position closed by the exchange risk engine, not the trader, as a market order.

  • Funding rate

    A recurring payment between longs and shorts that tethers a perp to spot price.

G

H

  • HVN and LVN

    Volume-profile areas of heavy trade (HVN) versus areas price passed through fast (LVN).

I

  • Insurance fund

    An exchange reserve that covers liquidation shortfalls before ADL is needed.

  • Isolated vs cross margin

    Isolated risks one position's margin; cross backs positions with the whole balance.

L

  • Leverage

    Controlling a position larger than posted margin, which shortens the distance to liquidation.

  • Liquidation

    Forced closure of a leveraged position when margin can no longer cover its losses.

  • Liquidation cascade

    A chain reaction where forced closures push price into the next cluster and trigger more.

  • Liquidation heatmap

    A shaded view of the price levels where leveraged positions would be liquidated.

  • Liquidation map

    A plot of estimated liquidation levels across price, above and below spot.

  • Liquidation price

    The price at which a position fails maintenance margin and gets force-closed.

  • Long/short ratio

    The balance of long versus short exposure on a venue, by accounts or by size.

M

  • Magnet level

    A dense liquidation cluster that can attract price because of the forced flow waiting there.

  • Maintenance margin

    The minimum equity a position must hold before the exchange liquidates it.

  • Mark price vs last price

    Last price is the latest trade; mark price is the index valuation that liquidates you.

  • Max pain

    The strike where the most open option value would expire worthless.

N

  • Negative funding

    Shorts paying longs, because the perp trades at a discount to spot.

O

  • OI-weighted

    An average weighted by each venue's open interest, so bigger books count more.

  • Open interest

    The total contracts currently open, measuring committed capital rather than volume.

  • Open interest divergence

    Price and open interest moving inconsistently, separating new positioning from unwinds.

  • Order book depth

    Resting bid and offer size at each price, which sets how far size moves price.

  • Out of sample

    Data that played no part in building or tuning the model, so results on it mean something.

  • Overfitting

    A model fitting noise instead of signal, which looks great in a backtest and fails live.

P

  • Perpetual swap

    A futures contract with no expiry, kept near spot by periodic funding payments.

  • Point of control (POC)

    The price with the most traded volume in a period, the profile's fattest level.

  • Precision@N

    How often the top N ranked predictions are correct, rather than scoring everything.

R

  • Reach rate

    How often price actually reaches a published level within a stated horizon.

S

  • Socialized loss

    An unrecoverable liquidation deficit spread across profitable traders.

  • Spoofing

    Placing orders with no intent to fill, to fake supply or demand.

  • Stop hunt

    A push through an obvious stop and liquidation cluster to trigger the flow, then a reverse.

  • Survivorship bias

    Measuring only what survived, and mistaking that filtered sample for the whole population.

V

W

  • Walk-forward validation

    Testing a model only on data that came after the data used to build it, rolled forward.

  • Wick

    The candle line showing highs and lows that traded but did not hold.

  • Wilson confidence interval

    A confidence interval for a proportion that stays sensible at small sample sizes.