Leverage on a perpetual position expresses how much notional exposure a fixed amount of collateral controls. Choosing 5x, 10x or 25x does not change the dollar exposure to the underlying asset's price once the notional is fixed; it changes how little collateral is required to hold that notional, and how close a liquidation sits to the entry price.
Leverage equals notional divided by margin posted, so initial margin equals notional divided by leverage. A fixed collateral amount used at higher leverage opens a bigger notional position; a fixed notional position held at higher leverage frees up collateral for other use. Both descriptions describe the same ratio viewed from opposite ends, and neither one changes how much profit or loss a given percentage move in price produces on the full notional; only the collateral required, and the resulting distance to liquidation, changes.
It helps to keep two separate questions apart when thinking about leverage: how big is the bet, and how much of it is funded by collateral versus by the leverage mechanism itself. A trader who wants a large directional bet can reach it either by posting a large amount of margin at low leverage or a small amount of margin at high leverage, and the two routes produce an identical position size and an identical dollar sensitivity to price. What they do not produce identically is risk of ruin: the low-leverage route has a much larger cushion before that same position gets liquidated.
The table below is purely illustrative arithmetic, showing how a fixed $10,000 notional position, or a fixed $1,000 of margin, scales with leverage.
Reading the table either way makes the same point: leverage is a ratio applied to the same underlying price move; it does not manufacture a separate source of profit on its own. A 5% move in the underlying asset produces the same dollar profit or loss on a $10,000 notional position whether it was opened with $2,000 of margin at 5x or $400 of margin at 25x; what changes between those two is that the 25x position loses its entire margin, and gets liquidated, on a much smaller adverse move.
Liquid, lower-volatility markets such as BTC or ETH can typically support higher maximum leverage than a thinly traded or high-volatility token, because the venue needs enough order-book depth and enough time to close out a position before a price move outruns the posted margin. This is why maximum leverage is published market by market instead of as one account-wide number, and why the ceiling on any given asset can move as that asset's own liquidity and volatility change over time, tightening in periods of thin trading and loosening again once depth returns.
The leverage ratio chosen when opening a single position is not always the same number as a trader's overall exposure relative to their total account balance. A trader might open one position at 10x leverage while holding the rest of their collateral unused, in which case the account's effective leverage, total notional across every open position divided by total account equity, is much lower than 10x. Effective leverage is the more useful number for understanding overall account risk, while the per-position ratio is the more useful number for understanding a single position's liquidation distance. A trader running several positions at once, each at a modest per-position ratio, can still end up with a high effective account leverage if the positions are opened on correlated assets that tend to move together, since a single adverse market-wide move then hits every position at once. Two positions on unrelated assets, by contrast, are less likely to be liquidated by the same market event, which is part of why effective leverage across a diversified set of positions is not simply the average of the individual per-position ratios.
Higher leverage places the liquidation price closer to entry, because there is less margin buffer between entry and the maintenance margin floor. As a rough illustration, at 5x leverage price needs to move roughly a fifth of the way against the position, before fees and maintenance margin, to erase the posted margin; at 50x that move shrinks to roughly a fiftieth. The exact liquidation price on any real position also depends on the venue's specific maintenance margin requirement and fee schedule, so this is a useful approximation, and the same leverage ratio can therefore produce a slightly different liquidation distance from one venue to the next even on the same asset. It is also worth noting that a fee charged on the full notional becomes a larger percentage of margin as leverage rises, since the same dollar fee is measured against a smaller collateral base.
The leverage ratio chosen interacts with the margin mode behind it. In isolated margin, the ratio applies only to the collateral allocated to that single position. In cross margin, the effective buffer behind any one position also depends on how much of the shared account balance sits free elsewhere. How collateral itself is selected and discounted across these modes is covered in collateral types for perps , and how leverage interacts with position direction is covered in perp position types .
Leverage trading means controlling a position larger than the collateral posted to open it, using borrowed buying power built into the contract's margin design. It magnifies both gains and losses relative to the margin posted, and it brings liquidation risk that a fully collateralized spot position does not carry.
A leverage ratio is the multiple between a position's notional size and the margin posted to open it, such as 10x meaning $1,000 of margin controls $10,000 of notional. The ratio determines both how much collateral a given position requires and how close price has to move to trigger liquidation.
Maximum leverage is the highest ratio a venue allows on a given market, set per asset. It is generally set lower for thinly traded or high-volatility assets, since the venue needs enough liquidity and time to close a position before an adverse move outpaces the posted margin.