A trader opens a leveraged perpetual futures position on Hyperliquid with $10,000 in notional exposure at 5x leverage using $2,000 in collateral. The market moves 3% against the position. Under full liquidation mechanics, the trader loses the entire $2,000 immediately and the position closes. Under partial liquidation, the platform closes only enough of the position to restore the account above minimum margin requirements—perhaps 60% of the trade—and the remaining 40% stays open. The trader keeps $800 of their collateral instead of zero. This sounds like survival. In practice, it is the beginning of a cascade that combines leverage compounding, forced rebalancing at the worst time, and a tax bill that traders rarely anticipate.
Partial liquidation is a design choice, not a miracle. Hyperliquid uses it to reduce catastrophic losses and prevent situations where entire positions evaporate in a single block. But partial liquidation creates its own category of losses that sit between “walk away breakeven” and “complete wipeout.” A trader watching 15% to 40% of their position automatically closed at market price, combined with the forced exit of their best-execution opportunity and the remaining position’s changed risk profile, often experiences a worse outcome psychologically and financially than a clean full liquidation. Understanding why requires examining the mechanics of how partial liquidation thresholds trigger, what liquidation costs actually extract from survivors, and why position sizing becomes even more critical when partial protection appears to reduce risk.

How partial liquidation thresholds actually work on Hyperliquid
Hyperliquid operates on its own blockchain with an on-chain order book, which means every liquidation is transparent and executed at observable prices rather than handled through internal calculations. The platform’s liquidation mechanics involve a specific triggering event: when an account’s maintenance margin ratio falls below the required threshold for that position size and leverage, the liquidation engine activates. The maintenance margin is typically lower than the initial margin required to open a position—this gap is where partial liquidation lives.
When a position is opened, the user sets initial margin (let’s say 20% for 5x leverage). This is the collateral percentage required to enter the trade. Maintenance margin is lower, perhaps 10%. The liquidation threshold is triggered when collateral value drops to 10% of the position’s notional value. At this point, the liquidation system does not close the entire position immediately. Instead, it closes enough of the position so that the account’s margin ratio returns to a healthier level—often around 50% above the maintenance minimum. This partial closure preserves some exposure while extracting cost.
The timing of partial liquidation matters because Hyperliquid’s on-chain order book means liquidations execute at whatever price is available on the book at that moment. If the market is moving quickly in the loss direction, the liquidation may execute at a worse price than the position’s entry. The account pays spreads, market impact, and opportunity cost all at once. A trader who was underwater 3% may suddenly be underwater 5% on the remaining position after the forced exit of a portion at unfavorable execution.
The liquidation system on hyperliquid also features an auto-deleveraging (ADL) mechanism and HLP (Hyperliquid LP) backstop for cases where liquidation proceeds insufficient to cover losses. This protects the system from cascading failures but does not protect individual traders from the partial closure itself. The trader whose position is 40% closed has lost not only the liquidated portion but also gained an involuntary reduction in exposure at precisely the moment when reducing exposure feels worst.
The hidden tax burden of partial liquidation
Partial liquidation creates a tax event that many traders misunderstand or overlook entirely. In most jurisdictions, closing a portion of a position—even an involuntary closure—triggers a taxable gain or loss. A trader with a $10,000 position that is 60% liquidated at a $500 loss has realized a $500 loss on the closed portion and must report it. The remaining 40% of the position is still open and unrealized, but the closed portion is locked in for tax purposes.
The complexity deepens when traders operate across multiple positions or use margin accounts in different tax regimes. Some jurisdictions require marking-to-market daily on derivatives positions. Others allow deferral until actual closure. A partial liquidation forces realization regardless of the trader’s intent, potentially triggering tax liability in a year where they would have preferred to realize losses in a different period or offset them against different gains. A day trader running multiple perpetual positions across different altcoins, each with a mix of realized and unrealized losses, can face a sudden tax bill from a single partial liquidation event on one coin.
The tax bill is often substantially higher than the nominal loss on the liquidated portion because it interacts with wash-sale rules, jurisdiction-specific treatment of derivatives, and the timing of other trades. A trader who enters multiple positions with the intent to close them as profit-taking opportunities emerge may find that a partial liquidation on one position prevents them from later offsetting that realized loss against realized gains on another position in the same year. The tax cost is invisible in the moment but surfaces at filing time, and it is permanent.
Why position sizing prevents partial liquidation from feeling like salvation
The most common error is treating partial liquidation as a second chance and re-risking the preserved collateral immediately. A trader’s position is partially liquidated, leaving them with $1,200 of the original $2,000 collateral. They perceive this as a warning and should reduce exposure. Instead, they re-lever to the same notional position on the remaining collateral, effectively doubling down on the same thesis at a higher leverage ratio. The market continues to move against them, and the second liquidation occurs at a higher percentage of remaining capital.
This behavior is common because partial liquidation feels like a reprieve rather than a failure signal. The account still has buying power. The position can still move favorably. The trader has time to wait for a reversal. All of these thoughts are true in the narrow sense and false in the strategic sense. A position that was undersized before liquidation was undersized for a reason: the trader’s account could not afford the adverse move that just happened. Partial liquidation proved that hypothesis. Rebuilding the same position at higher leverage is a decision to ignore the evidence.
Risk management best practice says that position size should be determined before entry based on the amount of collateral you are willing to lose on that position. If you are willing to risk 5% of your account on a trade, and the trade against you is 10%, then your position is twice as large as it should be. Liquidation, whether partial or full, is a market’s way of saying you sized incorrectly. Partial liquidation is not a discount on that lesson; it is a second lesson delivered with less time to react to the first.
Liquidation mechanics interact with margin mode and leverage limits
Hyperliquid offers three margin modes: cross margin, isolated margin, and strict isolated margin. Each changes how liquidation works and how quickly a single bad position can affect your entire account. In cross margin, all positions share collateral, so a liquidation on one perpetual position can pull from the collateral you set aside for another. In isolated margin, each position has its own collateral bucket, so liquidation on one position does not immediately affect others—but the position itself liquidates faster because only that position’s collateral is available.
Strict isolated margin is the most conservative: it locks in a specific amount of collateral for a position and prevents liquidation of that position from affecting other positions or vice versa. The trade-off is that a position in strict isolated mode can be liquidated more suddenly because no cross-collateral transfers can slow the decline. A trader using cross margin with three open positions, one of which is 15x leveraged, is essentially stacking leverage across all three. If one position moves 2% against the trader, cross margin uses all available collateral to maintain all three positions. A partial liquidation on the 15x position cascades because the collateral used to support it is no longer available for the others.
Maximum leverage on Hyperliquid varies by asset. Major pairs such as BTC and ETH may allow up to 50x, while smaller altcoins are capped at lower ratios. This is a risk control mechanism, but it does not prevent partial liquidation. A trader on BTC at 20x leverage is still exposed to a 5% adverse move triggering liquidation. The platform’s liquidation system operates identically across all leverage levels; the difference is how quickly a move of a given size becomes liquidation-triggering.
The rebalancing problem after partial closure
Partial liquidation changes the composition of a trader’s positions and collateral ratio in ways that cascade beyond the immediate trade. A trader with a portfolio consisting of a long BTC position and a long ETH position at 3x combined leverage, with 70% notional weight in BTC, experiences a partial liquidation on BTC. The liquidation engine closes 40% of the BTC position. The trader’s portfolio is now 55% BTC and 45% ETH, if no additional action is taken. The leverage on the remaining positions is no longer what the trader intended.
Some traders respond by immediately rebalancing: selling part of the ETH position to restore the original ratio. This creates another set of execution costs and tax events. Other traders leave the portfolio unbalanced and discover that it now carries a different risk profile than they modeled. If the trader’s thesis was that BTC would outperform ETH, the partial liquidation on BTC due to volatility, not fundamental weakness, has now made the portfolio underweight the intended core position. Rebalancing to fix this feels like “buying the dip,” which is a description that leads traders to ignore that they are already underwater and now are adding fresh capital to a position with a reduced profit margin.
The rebalancing problem is especially acute for day traders who run multiple positions on Hyperliquid with correlated but distinct timing. A trader scaling into a short position on a specific altcoin, with staggered orders and different entry prices, may find that a partial liquidation on the first scaled entry closes a portion of that position while the later entries remain open at worse average prices. The portfolio rebalances involuntarily in a direction that increases the loss on the remaining entries.
Fee extraction happens during liquidation, not after
Hyperliquid’s base taker fee is 0.045% and maker fee is 0.015% for perpetuals. During liquidation, the position is closed as a market taker order, meaning the full taker fee applies. On a $10,000 notional position being partially liquidated at $4,000 notional size, the trader pays a $1.80 taker fee on top of the slippage from executing at market prices. This seems small until you consider that the position is already underwater. The fee is extracted from an account that is already below its initial collateral value.
The fee is not a cost borne by the platform or the insurance fund. It is paid by the liquidated trader to the market makers and liquidity providers who filled the liquidation order. In a healthy market with tight spreads, this fee is barely noticeable. In a market with sparse liquidity on the book or during volatile period when the spread widens, the fee compounds with the slippage and creates a material additional cost. A trader being liquidated on a low-liquidity altcoin pair may face a 0.2% to 0.5% execution cost, which combined with the fee means the liquidation itself costs 0.25% to 0.55% of the liquidated position’s value.
This extraction happens automatically and is not negotiable. A trader cannot submit a limit order during liquidation or wait for a better price. The platform closes the position at the best available price on the on-chain order book, which is immediately market price. Traders should therefore model their liquidation risk not as the simple notional loss from price movement but as the price movement plus fees plus execution slippage on the portion being closed. A 3% adverse move on a position with 0.045% taker fee, 0.15% slippage, and a 0.25% cost from wider spreads is really a 3.45% economic loss by the time the partial liquidation is done.
Funding rates compound the damage after partial liquidation
Perpetual futures on Hyperliquid carry hourly funding rates, a mechanism where longs pay shorts (or vice versa) based on how much the market price diverges from the index price. When the market is extremely bullish, longs pay shorts 0.05% or more per hour. When bearish, shorts pay longs. A trader who is partially liquidated and retains a position in that direction continues to pay funding costs on the remaining notional value, but now at a reduced capital base and with the position moved closer to the liquidation threshold again.
A trader long BTC with $2,000 collateral on a $10,000 position at a +0.04% hourly funding rate is paying $4 per hour in funding costs on the full position. After partial liquidation, the remaining position might be $4,000 notional on $1,000 collateral. The funding rate continues to accrue on the $4,000 notional, so the trader is still paying $1.60 per hour. Over an 8-hour period during which the market moves sideways, the trader loses $12.80 to funding alone. This is a small amount, but it is extraction without any price movement, and it brings the account closer to liquidation again.
Funding rate regimes change based on market sentiment. A sustained bull run increases long funding costs. A sustained bear run increases short funding costs. A trader caught in a partial liquidation and forced to hold the remaining position cannot easily exit without recognizing the loss and must decide whether to hold through continued funding payments or close the remaining position manually at a loss. Neither option is pleasant, and both are more attractive if the position was sized correctly to begin with.
Smart contract and bridge risks apply to collateral and exits
Hyperliquid is built on its own L1 blockchain with HyperCore infrastructure and HyperEVM capabilities, which means smart contract vulnerabilities or operational failures in those systems directly affect the ability to trade and liquidate. The Ethereum Foundation publishes frameworks for smart contract security, and Hyperliquid’s infrastructure must meet equivalent standards of auditability and transparent code review. Traders deposit USDC via Arbitrum and other chains using a bridge. If the bridge is exploited or malfunctions, collateral can be lost or locked up, preventing normal liquidation mechanics from functioning and causing cascading liquidations across the platform.
A trader with $50,000 in collateral on Hyperliquid might face a scenario where a bridge vulnerability temporarily prevents collateral withdrawals. If the market moves against their positions, they cannot reduce exposure by withdrawing and closing positions on other platforms. They are locked in, and liquidation proceeds according to Hyperliquid’s own mechanics and liquidity, which may be degraded if other users are also attempting to exit due to the same bridge issue. The regulatory environment around crypto derivatives, as outlined by CFTC regulations on leveraged derivatives, does not provide customer asset protection for decentralized platforms, so bridge failures are trader risk, not insurance-covered events.
Regulatory restrictions create hidden liquidity and execution risk
Hyperliquid does not operate in the United States, Ontario, or sanctioned regions. Traders in these jurisdictions who attempt to access the platform using VPNs or residential proxies are taking additional risk beyond the normal partial liquidation and leverage mechanics. If a trader’s account is flagged as potentially US-based, their collateral can be frozen, and their positions may be force-closed at prices determined by the platform’s liquidation algorithm. This is not a normal liquidation with market-price execution; it is a punitive closure in a degraded liquidity environment.
Traders in the European Union face additional restrictions due to ESMA restrictions on cryptocurrency derivatives, including leverage caps and retail investor protections that may prevent account creation or restrict position sizes. The FCA in the UK has issued consumer warnings about crypto derivatives and restricts retail access to leveraged crypto trading. These regulatory barriers mean that traders in high-income countries with the most capital often have the least legal access to platforms like Hyperliquid, while traders in less-regulated jurisdictions may have more access but fewer recourse mechanisms if something goes wrong.
How to survive partial liquidation: discipline over luck
The practical defense against partial liquidation is position sizing before entry, not recovery after liquidation. A trader should calculate the maximum loss they can afford on each position—typically 1% to 3% of total account capital—and then derive the position size from that maximum loss, not the reverse. If a BTC position is 8% volatile intraday, and the trader is willing to risk 2% of account capital, then the position size is fixed. Leverage follows from position size, not the other way around.
The second defense is choosing margin mode carefully. For day trading multiple positions on Hyperliquid, cross margin allows efficient use of collateral but couples risk across positions. Isolated margin on key positions prevents a liquidation on one pair from cascading to others. The trade-off is that isolated positions liquidate faster, so position sizing becomes even more critical. A trader using strict isolated margin on a high-conviction position has reduced their ability to use cross-collateral for margin calls, so the position must be sized to survive a normal adverse move without relying on flexibility from other positions.
The third defense is monitoring liquidation prices continuously and treating them as a hard stop, not a theoretical limit. Most trading platforms provide liquidation price estimates, and Hyperliquid’s tools include liquidation price calculators. A trader should know their liquidation price before entering, monitor it throughout the trade, and exit the position manually well before liquidation becomes a realistic scenario. Exiting at a 2% loss to avoid a partial liquidation at a 5% loss is an excellent trade, even though it feels like a failure in the moment.
The fourth defense is avoiding the fallacy that partial liquidation is a reprieve. It is a warning that the position was sized too aggressively for the available collateral. Adjusting position size downward and reducing leverage after a partial liquidation is the correct response, not re-leveraging the remaining collateral. A trader who experiences partial liquidation twice on the same strategy in a month should stop that strategy and re-examine the position-sizing methodology.
Frequently asked questions
What exactly is partial liquidation and how does it differ from a full liquidation on Hyperliquid?
Partial liquidation closes only enough of a position to restore the account’s margin ratio above the maintenance minimum, while full liquidation closes the entire position. On Hyperliquid, partial liquidation is triggered when maintenance margin is breached but executes incrementally to preserve some exposure. The trader retains part of the position but suffers forced exit at market prices during a loss, plus taker fees and execution slippage, followed by continued funding costs on the remaining notional exposure.
Why does a partial liquidation feel worse than a full liquidation?
Partial liquidation is worse because it combines multiple costs: loss on the closed portion, fees and slippage on forced execution, a realized tax event, forced rebalancing of the remaining portfolio, and continued funding rate exposure on the remaining position at a reduced capital base. Full liquidation is catastrophic but final. Partial liquidation extends the pain and often triggers poor decision-making such as re-leveraging the remaining capital. Psychologically, it feels like a reprieve when it is actually a cascading failure.
Can I avoid partial liquidation by using different margin modes on Hyperliquid?
Margin mode selection (cross, isolated, or strict isolated) changes how quickly liquidation occurs and whether liquidation on one position affects others, but it does not prevent partial liquidation. Cross margin uses all available collateral to maintain all positions, slowing liquidation but coupling risk. Isolated margin prevents spillover but liquidates faster because only that position’s collateral is available. Strict isolated margin is most protective of other positions but liquidates most quickly. The real defense is position sizing; margin mode selection is a secondary control.
What is the total cost of partial liquidation beyond the headline loss?
A partial liquidation costs the trader the notional loss from price movement, plus taker fees (0.045% on perpetuals), plus execution slippage (typically 0.15% to 0.5% depending on liquidity), plus a realized tax event, plus continued funding costs on the remaining position at a reduced capital base. In total, a 3% adverse market move on a position being partially liquidated can cost 3.5% to 4.2% economically. This total cost is rarely visible in the moment but dominates the actual impact of the liquidation event.
Should I re-enter a position immediately after it is partially liquidated on Hyperliquid?
No. Partial liquidation is evidence that the position was sized incorrectly for the available collateral. Re-leveraging the remaining capital to restore the same notional exposure compounds the error and doubles the risk at a higher leverage ratio on a reduced capital base. The only correct response is to reduce exposure and position size, then re-examine the strategy’s assumptions before entering any new position.