How Liquidation Price Works: Why 10x Isn't a 10% Buffer

First, the myth: 10x doesn't mean "a 10% drop liquidates you"

This is the most common — and most expensive — gut-level mistake new futures traders make. The logic sounds clean: at 10x leverage my margin is one-tenth of the position, the price moves 10% against me, my loss equals my margin, so it takes a 10% drop to zero me out and liquidate me.

The problem is the word "takes." The exchange won't wait until you've actually lost every last cent. It force-closes the position while you still have a little margin left — enough to cover the cost of closing — because if the market keeps running and the loss eats through your principal, the exchange ends up footing the bill. That "a little left" floor is the maintenance margin. So the real liquidation point always arrives earlier than "you lose 100%."

On top of that, three more things pull the distance back in: the maintenance margin rate takes a slice, the open/close fees get counted in, and slippage at liquidation leaves you with even less. Stack those together and the real distance to liquidation at 10x is usually well under 10%. Below we'll turn it back into numbers, one piece at a time.

How liquidation actually gets triggered

Set the formula aside and liquidation comes down to one sentence: when your position's margin, after subtracting the unrealized loss, falls below the "maintenance margin" the exchange requires, the system force-closes the position.

Break it into three quantities and it clicks:

• Initial margin: the principal you put up when you opened. In isolated margin mode, it's the money you allocated to this position alone.
• Unrealized PnL: what the position is worth right now. When the price moves against you this is a negative number, eating into your margin in real time.
• Maintenance margin: the minimum margin the exchange requires to keep the position from being closed. It equals the position's notional value times the "maintenance margin rate."

When initial margin + unrealized PnL ≤ maintenance margin, liquidation triggers. In other words, your unrealized loss doesn't have to equal your entire margin — it only needs to push "the margin you have left" below the maintenance line. Solve that inequality for price and you get the liquidation price.

Isolated long: deriving the formula step by step

Let's derive the liquidation price for the most common setup: isolated margin, going long, on a USDT-margined linear contract. Notation:

• P₀ = entry price; L = leverage; m = maintenance margin rate (as a decimal — e.g. a tier of 0.5% is written 0.005); P_liq = the liquidation price we're solving for.

Step one: write down the relationship between principal and position. You use your principal as initial margin at L times leverage, which means in every 1 unit of notional value your own margin is 1/L. In other words, your margin ratio at entry is 1/L.

Step two: see how much you lose when the price moves against you. Going long, as the price falls from P₀ to P, the drop is (P₀ − P) / P₀. That drop comes straight out of your margin ratio — because the notional value is moving, the loss relative to your principal is amplified L times, but relative to the notional value the loss ratio is simply the drop itself.

Step three: write the trigger condition. Liquidation hits when the "remaining margin ratio" falls to the maintenance margin rate m:

Step four: solve for P_liq. Move the drop term across and tidy up:

The meaning is plain: open an L-times long and the price has to drop by roughly (1/L − m) to trigger liquidation. Note the minus sign — the maintenance margin rate m is "subtracted" from your buffer, which pulls the liquidation point closer to the entry price, not farther away.

An example with no specific exchange numbers, just to build intuition: assume 10x leverage and a maintenance margin rate of 0.5%, so 1/L = 0.1, and the liquidating drop ≈ 0.1 − 0.005 = 0.095 — about a 9.5% drop, already not 10%. Once you add the fees and slippage we cover below, it's closer still. The higher the leverage, the smaller 1/L gets, the more weight the maintenance margin rate carries, and the more glaring that "stolen distance" becomes.

Isolated short: the same logic, flipped

A short is perfectly symmetric — you only lose when the price goes up. You open a short, and as the price rises from P₀ to P_liq, the gain (P_liq − P₀) / P₀ eats into your margin ratio. It triggers the same way, when the remaining margin ratio falls to m:

Solve it:

A short's liquidation price sits above the entry price, triggering when the price rises about (1/L − m). One extra warning here: a short has no "floor" to its theoretical loss — however far the price can climb, your loss can be amplified that much, so when a short runs into a short squeeze, liquidation comes fast and hard. That's why a lot of experienced traders are especially wary of high-leverage shorts.

Maintenance margin rate: the key number that moves

The m in the formula is the most easily overlooked quantity in the whole mechanism — and the one you most must not memorize. It's the minimum margin, as a share of notional value, that the exchange requires to keep the position open. It isn't a fixed number; it changes for several reasons:

• The contract differs: major coins and small-caps often carry different maintenance margin rates, and less liquid contracts usually face stricter requirements.
• The position tier differs: this is the crucial one. The larger your position's notional value, the higher the "risk tier" it falls into, the higher the required maintenance margin rate, and the lower the maximum leverage allowed. In other words, the bigger you go, the harsher the system gets, and the liquidation price sits closer to your entry than it would on a small position.
• The exchange adjusts it: in extreme conditions or when a contract's risk profile shifts, the exchange may change the parameters.

So any claim that "the maintenance margin rate is just 0.5%" can only be treated as illustrative. For the real numbers, go by Gate's contract rules page, its risk-limit table, and what's shown on your order page, and note when you checked (we verified this in June 2026). When your position size crosses a tier, your liquidation price can "jump" — that's worth watching especially closely.

Why liquidation tracks the mark price

Plenty of people calculate a liquidation price by hand, watch the last traded price on their charts, and then get closed out before the last price ever touches their level. The reason: liquidation isn't triggered by the last traded price — it's triggered by the mark price.

The mark price is a "fairer" reference price the exchange computes from a spot index and multiple sources, designed to stop someone from using a single abnormal order to spike one exchange's price for an instant and maliciously trigger a wave of liquidations. It's usually smoother than any single exchange's last price. The result is that your unrealized PnL and your liquidation check both run on the mark price, not on the jumpy last price your eyes are glued to.

That has two practical consequences. First, the liquidation price you calculate by hand and the level that actually triggers will differ, because the "current price" you fed in is measured differently. Second, sometimes the last price wicks deep but the mark price doesn't follow, and you don't get liquidated — and the reverse holds too. This difference is worth understanding on its own; we cover it in mark price vs last price, and we strongly recommend reading them back to back.

Fees and slippage: the distance everyone forgets

Up to here we've been computing the "theoretical" liquidation price. But at the actual moment of liquidation, two more costs tilt the result further against you:

• Fees: you pay them on both opening and closing. Liquidation is itself a close, and that fee comes out of your margin. If you ignore fees entirely when estimating your distance to liquidation, you'll overstate your buffer. The actual maker/taker rates change — go by Gate's fee page — and for the mechanics see how contract fees are calculated.
• Slippage: liquidation tends to happen exactly when the market is violent and liquidity is thin. When the system takes over your position and closes it into the market, the fill can be worse than the mark price, and that loss is on you too. In extreme conditions slippage can be large.

The takeaway is simple: once you count in fees and slippage, your real liquidation point is a bit closer than the bare formula says. That's why we keep saying "don't trade right up against the liquidation price" — the little room you leave can easily get eaten by these two costs.

How far each leverage sits from liquidation

Lay out 1/L — the "ceiling on your reverse buffer" — and you'll feel completely differently about what those leverage numbers really mean (the table below shows only that 1/L ceiling, before subtracting maintenance margin, fees and slippage, so the real figure is closer):

See it clearly: from 10x to 100x, your buffer is cut from about 10% to about 1% — a tenfold shrink. And in crypto, a 1% move against you happens almost every few minutes. So high leverage was never "more opportunity"; it's "a buffer shrunk to almost nothing." We wrote a whole piece on why this is a trap for most people: why 100x leverage is a trap.

Why cross margin has no single fixed liquidation price

Every formula above is built on isolated margin — the margin is a specific sum you allocated to this one position, the boundary is clear, so you can compute a relatively definite liquidation price. Cross margin is different: your whole account's available balance backs the position, and your "margin" changes dynamically.

That has two consequences. First, cross margin is harder to liquidate on a single position, because the rest of the money in the account is propping it up. Second, and precisely because of that, cross margin has no single liquidation price you can pin down at a glance — the PnL of your other positions, and any funds you deposit or withdraw, all make that level drift. In an extreme case, one position's liquidation can drag the whole account down with it. For how to choose between the two modes and where each one bites, see isolated vs cross margin.

For people just starting out, isolated margin is usually easier to grasp — you know in advance that "this position can lose at most this much margin," and the liquidation price is clear to compute. Treat it as a hard boundary you set for yourself.

One more thing moves that level: adding margin to a position as price closes in. Put money in and the point gets pushed out a stretch. But it buys time without changing direction, and whether to top up or take the stop is a separate question — see adding margin, and whether you should.

Before you open: lay these numbers out

In the end, calculating the liquidation price isn't about "trading right up to the edge" — quite the opposite. It's about knowing where the cliff is so you can stay away from it. For turning "how far away" into concrete stop and position rules, read on to futures risk management.

Nobody can sell you a way to "never get liquidated"

The whole point of the derivation above is that the liquidation price is set by the exchange from margin, maintenance-margin rate and the mark price — it isn't a number a third party can quietly move for you. Keep that in mind, because the most common pitch aimed at English-speaking futures beginners is exactly the promise that someone can.

Now the specific scams that cluster around liquidation. Watch for a paid "anti-liquidation" bot or signal service that claims it will "protect your position" or "guarantee you never get liquidated" — no external tool can override the exchange's maintenance-margin math, so the promise is false on its face. Watch for "insurance" or "compensation" offers from strangers who say they'll refund your losses if you get liquidated, or who ask for an upfront "premium" to cover you; that's just a way to take a second payment. And be wary of anyone selling a "secret formula" or paid course promising a liquidation price that behaves differently from what the exchange shows — the only figures that count are the estimated liquidation price on your own order page and the mark price. The honest version of "avoiding liquidation" isn't a product you buy; it's smaller size, lower leverage and a stop set well before the liquidation level, which is exactly what risk management and the position size calculator are for.

FAQ