---
title: How often would 10x or 20x leverage get liquidated?
url: https://comparand.xyz/blog/how-much-leverage-liquidation-odds
description: On Binance Bitcoin, a 10% move against you came within a day 1.94% of the time and within a week 16.21%. Odds for 5x to 100x, with the limits.
updated: '2026-10-21'
category: Data
---

# How often would 10x or 20x leverage get liquidated?

On Binance Bitcoin, a 10% move against you came within a day 1.94% of the time and within a week 16.21%. Odds for 5x to 100x, with the limits.

On Binance Bitcoin perps from September 2019 to October 2026, price fell 10% below an hour's close within the next 24 hours in 1.94% of cases. That 10% is roughly how far a 10x long can fall before it is closed. Within 7 days, it happened in 16.21% of cases. At 20x, the distance shrinks to about 5%, and the figures rise to 9.91% within a day and 42.94% within a week.

These are past cases, not a forecast, and not a recommendation of any leverage. They also use a simple rule for the liquidation price, so real trades would differ.

For how liquidation works and how to work out your liquidation price, see [How does liquidation work in crypto?](https://comparand.xyz/blog/how-liquidation-works-crypto).

## What does leverage have to do with distance?

**Leverage** means a trade is bigger than the money you put into it. At 10x, $10 of your own money (called **margin**) controls a $100 position. If the position loses $10, your margin is gone and the exchange closes the trade. That closing is a **liquidation**.

Losing $10 on a $100 position is a 10% move against you. So a rough rule is: the distance to liquidation is 1 divided by the leverage. At 10x, it is about 10%. At 20x, about 5%. At 100x, about 1%.

For a long, the move against you is price falling. For a short, it is price rising. We measured both. A **perp** is a futures contract that never expires, and Binance's Bitcoin perp is called BTCUSDT.

*Diagram: A $100 position with $10 margin at 10x and $5 margin at 20x, the price fall (10% and 5%) that uses up each margin, and how often Bitcoin longs reached it within 24 hours (1.94% and 9.91%).*
## How did we count the odds?

Every hourly closing price on Binance BTCUSDT, from September 2019 to October 2026, is a starting point. For each one, we looked at the next 24 hours and the next 7 days. Using each hour's lowest and highest price, we asked: did price touch the distance against a long, or against a short? The table shows the share of starting points where it did.

There were 62,333 starting points for the 24-hour window and 62,189 for the 7-day window.

| Leverage | Distance | Long, 24 hours | Short, 24 hours | Long, 7 days | Short, 7 days |
| --- | --- | --- | --- | --- | --- |
| 5x | 20% | 0.18% | 0.15% | 3.86% | 4.88% |
| 10x | 10% | 1.94% | 1.77% | 16.21% | 19.91% |
| 20x | 5% | 9.91% | 10.19% | 42.94% | 46.08% |
| 50x | 2% | 39.29% | 39.05% | 74.44% | 76.40% |
| 100x | 1% | 63.54% | 64.15% | 86.42% | 88.20% |

The odds climb fast as leverage rises. At 100x, a 1% move against the trade came within a day in about 64 of every 100 starting points. At 5x, a 20% move came within a day in fewer than 1 of every 100.

Longs and shorts were close to each other over 24 hours. Over 7 days, the short side was reached somewhat more often at every leverage level.

## Does holding longer change the odds?

Yes, a lot. This table is for Bitcoin longs on Binance, from the same hourly starting points.

| Leverage | Within 1 hour | Within 4 hours | Within 24 hours | Within 7 days |
| --- | --- | --- | --- | --- |
| 10x | 0.03% | 0.16% | 1.94% | 16.21% |
| 20x | 0.19% | 1.25% | 9.91% | 42.94% |
| 50x | 2.16% | 10.18% | 39.29% | 74.44% |

A 10x long held for a week had its distance reached more often (16.21%) than a 20x long held for a day (9.91%). So "how much leverage" has no answer without "for how long".

## Does the answer change from coin to coin?

Yes, a lot. This table uses longs only, on Binance perps. Each coin's history starts between September 2019 and September 2020, so the periods are not identical.

| Coin | 10x, 24 hours | 20x, 24 hours | 10x, 7 days |
| --- | --- | --- | --- |
| BTC | 1.94% | 9.91% | 16.21% |
| ETH | 3.97% | 17.35% | 26.47% |
| TRX | 3.55% | 12.46% | 17.59% |
| XRP | 5.89% | 21.10% | 32.34% |
| DOGE | 7.01% | 25.51% | 36.83% |
| SOL | 7.92% | 29.60% | 42.26% |
| AVAX | 8.57% | 30.39% | 42.95% |

A 10x long on Avalanche saw its 10% distance reached within a day 8.57% of the time. For Bitcoin, it was 1.94%, more than four times less often. A coin that swings more reaches the same distance more often.

## Did the odds change by year?

They did. The year table is for Bitcoin only, within 24 hours. 2019 starts in September and 2026 runs to October 19, so those two are partial years.

| Year | 10x long | 10x short | 20x long | 20x short |
| --- | --- | --- | --- | --- |
| 2019 | 2.34% | 2.19% | 9.38% | 8.65% |
| 2020 | 2.00% | 3.10% | 11.02% | 13.63% |
| 2021 | 5.74% | 3.97% | 22.47% | 20.66% |
| 2022 | 2.58% | 1.87% | 13.11% | 10.99% |
| 2023 | 0.30% | 1.62% | 3.49% | 7.71% |
| 2024 | 1.40% | 0.75% | 8.39% | 9.04% |
| 2025 | 0.60% | 0.30% | 5.75% | 3.62% |
| 2026 | 0.52% | 0.37% | 4.17% | 5.11% |

In 2021, a 5% move against a long came within a day from 22.47% of starting points. In 2023, it was 3.49%. The typical ATR, which is the average size of recent hourly high-to-low ranges, was 1.17% of price in 2021 and 0.56% in 2023.

So one overall number mixes wild years like 2021 with quiet years like 2023. Read the year rows, not only the total.

## What do these numbers not show?

- **The distance is a rough rule.** Fees and the exchange's safety buffer, called maintenance margin, usually move the real liquidation price closer than 1 divided by the leverage. The real shares would be somewhat higher. We did not model any exchange's own formula.
- **A touch inside an hour counts.** We use each hour's high and low, so a quick spike counts, even if you could not know its order.
- **Exchanges often use a different price.** We used last traded prices, not the mark price many exchanges use to decide liquidations.
- **Starting points overlap.** Every hour is a start, so nearby hours share most of their path. The real number of separate cases is far smaller than 62,333.
- **These are selected coins.** The ten Binance perps are large coins that survived. They are not a random sample of all coins.
- **It is not a trade plan.** Stops, adding margin, or closing early change what happens. Funding payments and fees are left out too.

For how exchanges set the real liquidation price, see [how liquidation works](https://comparand.xyz/blog/how-liquidation-works-crypto). For the same kind of numbers used to place a stop, see [stop-loss distance in crypto](https://comparand.xyz/blog/stop-loss-distance-crypto-data). For normal daily swings, see [how much Bitcoin moves in a day](https://comparand.xyz/blog/how-much-does-bitcoin-move-in-a-day).

## How we measured this

We used hourly bars for ten Binance USDT perps from September 2019 to October 20, 2026. For each hourly close, we looked at the following 1, 4, 24 and 168 hours of bars. Windows missing any bar were dropped. A long counted as hit if the window's lowest low was at or below the close times (1 minus the distance). A short counted as hit if the highest high was at or above the close times (1 plus the distance). We applied the distance of 1 divided by the leverage and did no fee, margin or funding math.

## Try this in Comparand

With Comparand connected to your AI app, you can ask:

- "Show completed 1-hour bars for Binance BTCUSDT for the past 30 days." venue_bars returns open, high, low, close and volume for each completed bar. You can then ask your AI app to count how often the low reached a distance below an hour's close within the next 24 hours. The bars tool is free on every key.
- "Show hourly liquidation totals for one Hyperliquid coin around a big drop." hl_liquidations_series returns one coin's liquidation totals per complete UTC hour or day. It covers Hyperliquid only, not Binance. On pay as you go it costs $0.005 a call.

The first question lets you repeat this study on your own coin and dates. The second shows how many trades were actually closed on one exchange. Those are different from the touches we counted.

You can [try 3 coins free](https://comparand.xyz/try), read the [data tool docs](https://comparand.xyz/docs), or check [pricing](https://comparand.xyz/pricing).

## The short version

- A rough distance to liquidation is 1 divided by the leverage: about 10% at 10x and 5% at 20x.
- On Binance Bitcoin, a 10% move against a long came within 24 hours in 1.94% of cases, and within 7 days in 16.21%. At 20x, the 5% distance came within a day in 9.91% of cases and within a week in 42.94%.
- Holding time matters a lot: a 10x long held a week was hit more often than a 20x long held a day.
- Coins and years differ a lot. Solana's 10x figure for a day was 7.92% against Bitcoin's 1.94%.
- These are past cases, not a forecast, and real liquidation prices differ from this simple rule.

## Questions people ask

**How often would 10x leverage get liquidated on Bitcoin?**

A 10% move against the position is the rough distance for 10x. On Binance BTCUSDT, price reached 10% below an hourly close within 24 hours in 1.94% of cases, and within 7 days in 16.21%. Real liquidation prices differ a little, so these are rough.

**How often would 20x leverage get liquidated on Bitcoin?**

A 5% move against the position is the rough distance for 20x. On Binance BTCUSDT, price reached 5% below an hourly close within 24 hours in 9.91% of cases and within 7 days in 42.94%. For a short, the matching figures were 10.19% and 46.08%.

**How much leverage is safe in crypto?**

The data cannot name a safe amount. It shows how often price traveled a given distance against a position in the past, and that this differed by coin and by year. Fees, margin rules, stops and your own choices change what would happen to a real trade.

**Is liquidation risk the same for every coin?**

No. Within 24 hours, a 10% move against a long was reached 1.94% of the time on Bitcoin, 3.97% on Ethereum and 7.92% on Solana, on Binance. Coins that move more reach a given distance more often.