Liquidation Price and Stop-Loss: Five Differences That Matter

A stop-loss instruction and a liquidation mechanism have different jobs and can use different price references. Comparing their displayed numbers alone does not establish which event occurs first. Read the contract, collateral scope, trigger reference and actual fills before interpreting an apparent failure of an exit plan.

A stop-loss instruction and a liquidation mechanism have different jobs and can use different price references. Comparing their displayed numbers alone does not establish which event occurs first. Read the contract, collateral scope, trigger reference and actual fills before interpreting an apparent failure of an exit plan.

A hypothetical reference-price mismatch
95 markDisplayed liquidation level
96 lastUser’s sell-stop trigger
94.9 / 96.2Current mark / current last price

The example illustrates different references, not a real liquidation calculation or a safe distance between levels.

BTKWIN.COM · EDUCATIONAL EXAMPLE

Difference one: who initiates the action and why

A user sets a stop condition to initiate an attempted exit under the product’s supported options. A liquidation system acts when the product’s collateral and risk conditions require it. It can operate without a user stop and does not treat the existence of a stop as a substitute for maintenance requirements.

An open stop entry establishes an instruction with a particular current status, not that every position unit will exit before risk action. Verify whether the stop is linked to the position and whether its quantity still matches. An independent order can become outdated after the position changes.

The Kraken derivatives margin overview describes maintenance requirements within that product family. Check the actual contract and account rather than transferring liquidation terminology from another market that happens to use the same word.

Difference two: the two conditions may observe different prices

The last price records the latest match in a specified market. A mark price is a calculated reference under the product’s methodology. Both can exist simultaneously with different values. The Kraken reference-price explanation distinguishes last, index and mark prices and their roles.

Assume a hypothetical long position displays liquidation at a mark price of 95 and has a sell stop triggered by last price reaching 96 or lower. At one moment the mark is 94.9 while last price is 96.2. Under the assumed conditions, the liquidation reference has crossed its level while the last-price stop condition has not.

Illustrative numberReferenceRole
95MarkDisplayed liquidation condition
96Last tradeUser stop condition
94.9Current markCompared with liquidation level
96.2Current lastCompared with stop level

The Bybit execution and liquidation FAQ discusses reference differences in its product context. Selecting the same reference could address this particular mismatch, but does not guarantee processing order, sufficient liquidity or complete execution during fast movement. This example provides no universal safe distance.

Difference three: a displayed liquidation estimate can change

Collateral additions, withdrawals, fees, other positions’ results and collateral-asset valuations can change the resources available against requirements. Position size or maintenance requirements can also change. A saved liquidation estimate should therefore not be treated as an immutable number for the entire life of the position.

Inspect the estimate’s update time and assumptions, especially in shared collateral. If a collateral asset itself loses value, support can weaken independently of the contract’s price movement. Record collateral valuation and position valuation separately rather than assigning every account change to the contract.

The isolated and cross margin guide explains the boundary questions. Do not derive a reliable liquidation price merely by taking the reciprocal of leverage or comparing the initial deposit with entry value. Required product-specific inputs and charges are missing from that shortcut.

Difference four: triggering and filling are separate events

A stop-market instruction can execute at available prices that differ from the trigger, subject to its product rules. A stop-limit adds a price constraint but can remain partly or fully unfilled. Neither turns its trigger into a guaranteed sale receipt.

Assume a hypothetical stop-market sale of ten units triggered at 96. During execution, four can sell at 95.8 and six at 95.2. Gross value is 383.2 + 571.2 = 954.4, averaging 95.44. The difference from 96 is a price-execution effect in this model, not automatically a commission.

In a separate stop-limit scenario, a minimum acceptable sale price of 95.5 permits the four at 95.8 but not the six at 95.2. It produces partial proceeds of 383.2 and leaves six units, subject to the remaining conditions. Price control and completing all quantity are distinct requirements.

Additional protections can affect triggering. The Bybit TP/SL price-protection explanation describes a setting that can block certain last-price triggers when reference divergence exceeds its threshold. Read the applicable scope. The word protection does not guarantee that every requested exit will trigger in every market condition.

Difference five: liquidation is a product risk process

A liquidation mechanism may cancel orders, reduce positions or act across a defined collateral scope through several stages. Review the resulting record for quantities, actual closing values and fees. Crossing a displayed level alone does not explain every final account movement.

Bankruptcy prices and other risk references can have functions different from liquidation estimates and actual average closure prices. Do not collapse all those labels into a single supposed executable quote. Read what each field represents before using it to calculate the final result.

If a product describes insurance or negative-balance treatment, inspect its conditions and scope rather than treating the label as a universal guarantee covering every account obligation. This article distinguishes mechanisms and records; it does not determine the legal consequences of a particular contract.

Check how much of the position the stop covers

Suppose a position originally has ten units and an independent ten-unit stop. An additional purchase of five raises the position to fifteen without necessarily updating that instruction. Even if the stop triggers and fills completely, five units can remain. Successful execution of one order is not proof of a full position exit.

The reverse can occur after a partial manual exit, leaving an older instruction larger than the position. Its effect depends on the product and options such as reduce-only. Reconcile linkage, size and active status after changes instead of assuming every independent instruction adjusts automatically.

Use event records rather than one candle

A chart can display last price while risk action uses another reference. Its candle also aggregates events and may not establish their exact order. A notification’s arrival time can lag server processing. Collect trigger, order, fill, cancellation and liquidation timestamps from the relevant source.

Before relying on an exit arrangement, document the contract, settlement currency, margin scope, current requirements, liquidation reference, stop reference, resulting order type, size and protection settings. After an unexpected event, reconcile those facts and request an official explanation if necessary. Do not reopen a live position merely to recreate the event. For a later withdrawal problem, use the transfer helper as a separate investigation.

Continue reading

English content and examples reviewed: 27 October 2026. Read the Arabic edition.

تامر طعمه
تامر طعمه
Articles: 484