Where each leverage level gets liquidated, from 3x to 100x
Leverage decides how much adverse movement a position can take before the exchange closes it. At 10x it takes a drop of close to 10%; at 100x a little over half a percent is enough. This guide works out where each leverage level liquidates, shows how a liquidation map groups them and explains why the leverage a trader declares rarely matches the leverage they actually carry.
Leverage and the distance to the liquidation price
At leverage L, the initial margin is a 1/L fraction of the position's value. If price moves against the position by that same fraction, the margin is gone. In practice the exchange steps in slightly earlier, because it requires a maintenance margin, a percentage of value known as the MMR. Rearranging the exact formula gives the relative distance between entry and liquidation:
distance_long = (1/L − MMR) / (1 − MMR)
distance_short = (1/L − MMR) / (1 + MMR)As a rule of thumb, the distance is roughly 1/L minus the MMR. Doubling leverage cuts the room for error by a bit more than half, because the MMR does not shrink along with it.
Liquidation levels for 10x, 25x, 50x and 100x leverage
For an entry at 100,000 USDT with the 0.4% MMR of Binance's first BTC bracket, these are the liquidation prices at each level:
| Leverage | Long liquidation | Distance | Short liquidation | Distance |
|---|---|---|---|---|
| 10x | 90,361.45 | −9.64% | 109,561.75 | +9.56% |
| 25x | 96,385.54 | −3.61% | 103,585.66 | +3.59% |
| 50x | 98,393.57 | −1.61% | 101,593.63 | +1.59% |
| 100x | 99,397.59 | −0.60% | 100,597.61 | +0.60% |
A 10x long liquidates at 90,361.45 and a 25x long at 96,385.54. At 50x the level climbs to 98,393.57, and at 100x to 99,397.59. On the short side the same leverages liquidate at 109,561.75, 103,585.66, 101,593.63 and 100,597.61. Longs always sit below entry and shorts above, at nearly symmetrical distances.
How a liquidation heatmap groups leverage tiers
A liquidation map cannot track every possible leverage, so it groups money into four tiers. liqlin keeps two maps per pair: one for short timeframes, built for moves that play out over hours, and one for long timeframes, built for days and weeks. Each has its own four tiers:
| Tier | Scalp (5m, 15m, 1h) | Swing (4h, 1d) |
|---|---|---|
| L1, conservative | 10x | 3x |
| L2, moderate | 25x | 5x |
| L3, aggressive | 50x | 10x |
| L4, maximum | 100x | 25x |
The scalp map uses 0.05% price cells and the swing map 0.20% cells. Swing levels land much further from entry, as the same calculation with their leverages shows:
| Leverage | Long liquidation | Distance | Short liquidation | Distance |
|---|---|---|---|---|
| 3x | 66,934.40 | −33.07% | 132,802.12 | +32.80% |
| 5x | 80,321.29 | −19.68% | 119,521.91 | +19.52% |
| 10x | 90,361.45 | −9.64% | 109,561.75 | +9.56% |
| 25x | 96,385.54 | −3.61% | 103,585.66 | +3.59% |
Each tier's leverage is fixed. What gets tuned is how much new notional each tier receives: the starting split comes from real positions, and calibration against observed liquidations corrects it.
The exchange's maximum leverage caps each tier
Not every contract allows 100x. Each margin bracket on an exchange has a maximum leverage, and on liqlin a tier never exceeds it: if the tier is 100x and the contract only allows 40x, the tier is computed at 40x. This is not cosmetic. Once leverage goes above 1/MMR, the formula puts the liquidation on the wrong side of the entry.
Hyperliquid BTC shows it clearly: a 40x maximum and a 1.25% MMR. With an entry at 100,000, a long computed at 100x would liquidate at 100,253.16, above the price it was opened at, which cannot happen. Capped at 40x, the level sits at 98,734.18, where it belongs. Gold and silver perpetuals in Hyperliquid's xyz section (25x with a 2% MMR) and oil (20x with 2.5%) behave the same way.
Declared vs effective leverage in cross margin
The tables above describe isolated margin, where a position only has the margin assigned to it. In cross margin the cushion is the whole account balance, so a position declared at 25x can survive a far bigger move than the 3.6% its leverage implies.
Hyperliquid makes this measurable because its positions are public. On 2026-10-05, 31,640 real positions were analysed: 90.6% of notional was in cross margin. In isolated margin the formula matched; in cross margin the median real distance to liquidation was 5 to 12 times the nominal one.
So the map sorts money by the distance at which it actually liquidates, not by the leverage the trader declares. Each tier is a band of distances: on the scalp map, the 100x tier holds what liquidates within 1.41% of entry; 50x, between 1.41% and 2.83%; 25x, between 2.83% and 6.32%; and 10x, between 6.32% and 14.14%. It is effective leverage, not declared leverage.
Why high-leverage liquidation levels cluster near price
A 100x position lives 0.6% from its entry. A routine few hours of BTC price action can reach it, so on a liquidation heatmap the 50x and 100x levels show up as thin lines hugging the price that appear and vanish quickly: the mark price crosses them and they drop off the map. Levels at 10x or 3x sit much further out and can last for weeks.
In the real-position measurement, what liquidated within 1.41% of entry was a small share of the total, and on the scalp map the 100x tier starts with 6.11% of new notional. A nearby cluster is rarely the largest, but it is the first one price meets, and its liquidations are market orders that push in the direction of the move. Reading that is part of the analysis; what to do with it is each reader's own decision, and this guide is not investment advice.
Frequently asked questions
At what percentage does a 100x position get liquidated?
With a 0.4% MMR, like Binance's first BTC bracket, about 0.6% from entry on either side. A higher MMR makes the distance even shorter.
Why does the liquidation map show no real 100x levels on some coins?
Because the exchange does not allow that leverage on that contract. liqlin caps each tier at the bracket's maximum, so a 100x tier is computed at the real maximum, for example 40x on Hyperliquid BTC.
Does cross margin change the liquidation price?
Yes, a lot. In cross margin the position uses the whole account balance as its cushion. In the real Hyperliquid positions measured, the median distance to liquidation was 5 to 12 times the nominal one.
Does more leverage mean more liquidations on the map?
Not necessarily. Most of the measured notional uses low effective leverage. High-leverage levels stand out because they sit close to price, not because they are large.
Related maps
Guides to read the map
- Liquidation heatmaps explained: what they show and how to read oneA liquidation heatmap estimates where leveraged positions would be force-closed. See what public data it uses, why longs sit below price and how to read it.
- The exact liquidation price formula, derived and explainedThe exact liquidation price formula for Binance USDⓈ-M perpetuals, derived step by step: MMR, margin brackets, cum and why the shortcut puts levels wrong.
- Gold (XAUUSD), silver and oil liquidation mapWhere leveraged money gets liquidated in gold, silver, oil, EURUSD and USDJPY perpetuals on Binance, Bybit, OKX and Hyperliquid, shown with MT4 names.
- Accuracy: how we measure whether the map is rightHow we check whether a liquidation map is right: observed liquidations, hit rate by percentile, out-of-sample walk-forward tests and real Hyperliquid positions.