Market and liquidity risk: price, depth and cascades
Price risk and the risk of being unable to transact near that price are separate problems, and the second one arrives during stress.
Market risk is the risk that the price moves against a position. Liquidity risk is the risk that the price shown cannot be obtained for the size that needs to trade. They are usually discussed as one thing, and they behave differently: market risk is present continuously and measurable from a price series, while liquidity risk is close to invisible in calm conditions and dominates during stress, which is exactly when a measured estimate taken earlier is least applicable.
Measuring price risk without assuming a bell curve
Realized volatility over a window, shown here as 30-day volatility, describes the dispersion of recent returns. It is a summary of a sample and inherits every property of that sample. Return distributions in this asset class have heavy tails, so the largest daily moves are far larger than a normal distribution fitted to the same data implies, and they cluster: quiet periods follow quiet periods and violent days arrive together. Two complementary readings help. Drawdown, as current drawdown and maximum drawdown, describes the worst realized path rather than the average dispersion. Downside deviation isolates the variability of negative returns, since an asset that rises sharply and falls gently produces the same volatility figure as its mirror image. None of these describes future dispersion; they describe what has already been survived.
A market with no closing bell
These markets trade continuously with no scheduled halt, no circuit breaker on most venues, and no auction to reopen after a disorderly move. That removes overnight gap risk in the traditional sense and replaces it with something less forgiving: there is no interval during which a position cannot be liquidated, and no pause in which liquidity can regroup. Stress therefore expresses itself as a continuous decline through thinning depth rather than as a gap between a close and an open. Fragmentation compounds this, because liquidity sits across many venues in many jurisdictions with different access, so aggregate figures overstate what any single participant can actually reach.
The banking system around the market does keep hours, which produces a particular pattern. Transfers of national currency into and out of venues settle on business days, so the ability to add collateral or move cash between venues is constrained at exactly the moments when the asset market continues trading. Weekends and public holidays therefore combine a fully open market with a partially closed funding channel, and stablecoins circulate as the substitute precisely because they do not stop. This is one reason large moves have repeatedly begun outside business hours in the major financial centers.
Liquidity is a regime rather than a level
Depth is measured as resting size within a band of the mid price, and it is revocable at any instant, since quotes can be withdrawn faster than an order can arrive. Depth is thickest when volatility is low and thins precisely when it rises, because market makers widen quotes and reduce size as inventory risk increases. The practical consequences are slippage that exceeds any historical average and an execution cost that scales with size in a way calm-period measurements do not predict. Turnover, as 24-hour turnover measured against size, is the standard proxy for tradability relative to size, with the caveat that its numerator can be inflated by wash trading on venues with an incentive to appear active. A high turnover figure sourced from unverifiable venues is not evidence of anything.
The leverage feedback loop
Derivatives convert a price move into forced selling. A perpetual future has no expiry and holds its price near spot through a funding rate paid periodically between longs and shorts, which means crowded positioning is observable as persistent one-sided funding. When price moves against leveraged positions, the venue's engine closes them automatically, and those closures are themselves market orders that push the price further in the same direction, triggering the next tier. This is a liquidation cascade, and it produces the characteristic pattern of a fast move far beyond where any fundamental repricing would place it, followed by a partial recovery once forced flow is exhausted. The severity depends on how much leverage is outstanding, how concentrated liquidation prices are, and how thin the book is when the sequence begins. Leverage in the system is therefore a liquidity variable as much as a positioning one.
Correlation rises when it is least helpful
Cross-asset correlation to bitcoin is typically high and tends toward one during stress, so the diversification apparently available in normal conditions is smallest at the moment it would matter. Beta to bitcoin adds the magnitude dimension, describing how much an asset tends to move for a given move in the reference. The reason is structural as much as behavioral: positions across many assets are financed by the same lenders, collateralized in the same stablecoins, and liquidated by the same engines, so forced flow arrives in everything simultaneously regardless of what each asset does.
Where liquidity risk hides on-chain
On-chain venues have their own version of the same problem. A pool with concentrated liquidity offers deep pricing inside the range providers chose and very little outside it, so a move beyond the range converts a liquid pair into a thin one instantly. Providers experience impermanent loss as the relative prices diverge, and are most likely to withdraw during the volatility that causes it. Exit mechanics matter too: staked positions can face an unbonding period, some wrapped or receipt assets redeem only through a queue, and a pool that appears to offer an exit is really offering the other asset in the pair at whatever price the curve gives after the trade. Liquidity risk in these cases is a property of a specific mechanism rather than of the asset in general.
Concentration in supply, stake and validators is examined next, and it is one of the structural reasons liquidity can be thinner than headline figures suggest. The screener allows filtering on turnover and volatility, and compare places these measures side by side.