The result at a glance
Assume a BTC entry at 50,000 USDT, a quantity of 0.2 BTC (10,000 USDT notional), an isolated long, and the teaching inputs used across this site: a 0.50% maintenance-margin rate plus a 0.05% liquidation-fee reserve, so r = 0.0055. Initial margin is 2,000, 1,000, 500 and 200 USDT at 5x, 10x, 20x and 50x respectively.
The simplified liquidation estimates are about 40,221.22, 45,248.87, 47,762.69 and 49,270.99 USDT, which are adverse distances of roughly 19.56%, 9.50%, 4.47% and 1.46%. Moving from 5x to 50x reduces the price buffer to about one thirteenth; it does not make the setup ten times more likely to win.
Entering the same figures into the DZPTO liquidation calculator reproduces these numbers exactly. If your result differs, check that the liquidation-fee reserve was filled in.
How the estimate is produced
Simplified long liquidation price = (entry × quantity − margin) / [quantity × (1 − r)], where r is the maintenance-margin rate plus the liquidation-fee reserve. At 10x: (50,000 × 0.2 − 1,000) / [0.2 × (1 − 0.0055)] = 9,000 / 0.1989 = 45,248.87 USDT, an adverse distance of (50,000 − 45,248.87) / 50,000 = 9.50%.
A short flips both signs: (entry × quantity + margin) / [quantity × (1 + r)]. The same 10x short gives 54,699.15 USDT, a 9.40% adverse distance — long and short are not symmetric, because the denominator moves the other way.
This is the point where margin balance equals maintenance margin plus the fee reserve. It is a comparison model: it excludes maintenance tiers, added margin, other positions and cross-margin equity. The exchange mark price and account risk engine remain authoritative.
The full comparison across four leverage levels
The same inputs, four leverage levels side by side. Every row reproduces line by line in the liquidation calculator.
Margin falls tenfold, from 2,000 to 200 USDT. The adverse distance is compressed thirteenfold, from 19.56% to 1.46%.
The two do not scale together, and the gap comes from r sitting in the denominator: the higher the leverage, the more that constant weighs on the result.
| Leverage | Initial margin (USDT) | Simplified liquidation (USDT) | Adverse distance |
|---|---|---|---|
| 5× | 2,000 | 40,221.22 | 19.56% |
| 10× | 1,000 | 45,248.87 | 9.50% |
| 20× | 500 | 47,762.69 | 4.47% |
| 50× | 200 | 49,270.99 | 1.46% |
Run it backwards: let the stop distance cap the leverage
Most people pick leverage first and read the liquidation price second. The useful order is the reverse. If a strategy needs d of price room, the highest leverage that still fits is L = 1 / [1 − (1 − d) × (1 − r)].
That cap only guarantees the liquidation level sits beyond the stop; it contains no safety margin. In practice step down one or two notches, because mark price, spread and funding deductions all eat into what is left, and the maintenance-margin rate itself rises with the notional tier.
Read the cap against your actual stop: needing 6% of room while running 20× means liquidation arrives before the stop does. That is not a matter of risk appetite, it is a structural mismatch.
Maximum leverage L = 1 / [ 1 − (1 − stop distance) × (1 − r) ]| Stop distance the strategy needs | Theoretical maximum leverage | Adverse distance after rounding down |
|---|---|---|
| 2% | 39.39× | 39× → 2.03% |
| 4% | 22.08× | 22× → 4.02% |
| 6% | 15.34× | 15× → 6.15% |
| 10% | 9.53× | 9× → 10.62% |
Why high leverage is fragile
With only a 1.46% buffer, short-lived volatility, spread and mark-price movement can consume most of the available room. A trade can be liquidated on the path even if the later directional thesis proves correct.
Leverage also does not proportionally reduce trading fees. Fees are generally calculated from notional exposure, so the same 10,000 USDT position can incur similar base fees at 5x and 50x even though the posted margin differs.
A practical pre-trade check
Work backward from a defensible stop and loss budget instead of selecting the highest leverage first. A strategy needing 4% of price room is structurally incompatible with a simplified 1.46% liquidation buffer.
Then verify margin mode, maintenance bracket, mark price, displayed liquidation price and account fee tier in the exchange order preview. If it differs from DZPTO, use the account value.
Official sources and calculation boundary
The liquidation condition comes from OKX's documentation and the leverage tiers from Binance's contract endpoint. Both describe venue rules, not your account's current tier: r = 0.0055 is fixed here so the four leverage levels stay comparable to each other.
Next checks in this series
Liquidation and margin risk
How adding margin changes a BTC long liquidation estimate
Reproducible scenario guide
Liquidation and margin risk
Why DZPTO, Binance and OKX can show different liquidation prices
Reproducible scenario guide
Venue comparison
Binance versus OKX margin models: what must be aligned
Reproducible scenario guide
Frequently asked questions
Is 20x exactly twice as risky as 10x?
Real risk cannot be expressed as a simple multiple, but for equal notional exposure 20x posts less initial margin and has a materially shorter simplified adverse distance.
Should the liquidation price be my stop?
No. A stop should reflect strategy invalidation, slippage and loss budget, and should normally sit before the liquidation-risk zone.
Why does this differ from what the exchange shows?
This article fixes r = 0.0055 so the four leverage levels are comparable to each other. An exchange uses your account's current maintenance tier, liquidation fee, margin mode and mark price. Tiers move with notional value, so the gap widens for larger positions. The account order preview is authoritative.
What does a 1.46% buffer at 50× mean in practice?
After spread and mark-price drift are taken out, little more than a percentage point may remain. This is not simply 'somewhat riskier': ordinary volatility is enough to reach the threshold, so the chance of being liquidated on the path while the directional view is eventually correct rises sharply.