Research · October 28, 2026A liquidation cascade is not the bottom. What Bitcoin did after the 20 worst flushes
"Longs got wiped, so that's the bottom." That's what the crowd thinks while catching the knife right in the cascade. I pulled the 20 worst Bitcoin cascades since March 2025 and checked. In 11 of 20 cases price was pushed even lower within the next 24 hours.
What a cascade is
A cascade is an hour when BTC dropped the hardest and long liquidations in that or the next hour were among the largest of their period. I took hourly candles of BTC futures on Binance from Mar 18, 2025 to Oct 23, 2026 and picked the 20 strongest drops.
The top 20 flushes
The king of all cascades came on Oct 10, 2025 at 21:00 UTC. Bitcoin dropped from 114,200 to 101,500 within the hour, minus 11.1%.
The nasty part: that hour closed at 113,300. Whoever sold at market at the bottom sold the very wick. And 7 days later the price was still 5.3% below the close of that hour.
The rest of the top 20 dropped 2.6% to 11.1% within the hour. The whole list is in the table below.
| Cascade hour (UTC) | Price at the open | Low of the hour | Drop | 24h later | 7 days later | Back to the start |
|---|---|---|---|---|---|---|
| Apr 7, 2025, 06:00 | 76,900 | 74,500 | −3.1% | +6.2% | +13.1% | in 4h |
| Apr 10, 2025, 15:00 | 81,200 | 78,600 | −3.3% | +4.8% | +7.3% | in 15h |
| Aug 14, 2025, 12:00 | 120,900 | 117,400 | −2.9% | −0.2% | −4.9% | not within 30 days |
| Aug 24, 2025, 19:00 | 114,600 | 110,500 | −3.6% | −1.7% | −3.3% | in 18 days |
| Oct 10, 2025, 21:00 | 114,200 | 101,500 | −11.1% | −2.1% | −5.3% | in 2h |
| Oct 30, 2025, 04:00 | 110,700 | 107,800 | −2.6% | +1.4% | −4.2% | in 2h |
| Nov 21, 2025, 07:00 | 85,500 | 81,100 | −5.2% | +0.6% | +8.2% | in 10h |
| Dec 1, 2025, 00:00 | 90,300 | 86,900 | −3.7% | −0.6% | +3.8% | in 39h |
| Dec 5, 2025, 16:00 | 90,400 | 88,000 | −2.7% | +1.0% | +1.4% | in 2 days |
| Dec 12, 2025, 15:00 | 92,400 | 89,700 | −2.9% | +0.2% | −2.2% | in 23 days |
| Dec 18, 2025, 17:00 | 88,000 | 85,400 | −2.9% | +0.5% | +1.9% | in 14h |
| Jan 21, 2026, 16:00 | 90,200 | 87,500 | −3.0% | +1.6% | +2.3% | in 3h |
| Jan 30, 2026, 01:00 | 84,200 | 81,000 | −3.8% | +1.9% | −20.3% | in 17h |
| Jan 31, 2026, 18:00 | 78,900 | 75,500 | −4.3% | 0.0% | −10.9% | in 5h |
| Feb 6, 2026, 00:00 | 62,900 | 59,800 | −4.9% | +10.2% | +3.8% | in 1h |
| Feb 23, 2026, 01:00 | 66,700 | 64,200 | −3.7% | −0.4% | +3.3% | in 3 days |
| Feb 28, 2026, 06:00 | 65,600 | 63,000 | −3.9% | +5.9% | +7.1% | in 11h |
| Jun 5, 2026, 18:00 | 60,800 | 59,100 | −2.8% | +2.1% | +7.2% | in 2h |
| Jun 25, 2026, 13:00 | 61,200 | 58,000 | −5.2% | +3.3% | +6.3% | in 6 days |
| Sep 15, 2026, 18:00 | 76,900 | 74,900 | −2.6% | −0.5% | +13.8% | in 42h |
Don't market-sell at the bottom of a cascade candle: the hour often closes far above its low.
Caught the knife, cut the hand
In 11 of 20 cases the price went below the cascade low within the next 24 hours. More than half of the "bottom" buyers got pushed lower.
My take: a cascade is not a buy signal. It's a signal to wait until the last ones get flushed.
Don't catch the knife mid-cascade: let the market finish the flush.
Three days later it's a different story
24 hours later the price was above the cascade hour's close in 13 of 20 cases. 3 days later, in 15 of 20, with a median move of +2.6%. 7 and 30 days later, in 13 of 20.
Then I went further and checked all drops of 2% or more within an hour: 20 before 2026 and 19 in 2026. 3 days later the price was higher in 15 of 20 and 12 of 19. Panic sells, patience collects.
Looking to enter after a cascade? Think 3 days ahead, not the next hour.
Price goes back to where it fell from
The juicy part. The price came back to the level where the cascade started within 24 hours in 12 of 20 cases, within 7 days in 17 and within 30 days in 19.
Not a lucky streak either: across all drops of 2% or more, it was back within a week in 16 of 20 before 2026 and 15 of 19 in 2026. A cascade flushes leverage, not the trend. Leverage gone, price back.
The rebound target after a cascade is the price where it started.
Shorts get flushed the same way
The worst short squeeze came on Aug 19, 2026 at 15:00 UTC: BTC ripped 6.9% within the hour, from 65,900 to 70,400.
After the 20 biggest short squeezes the price went above that hour's high within the next 24 hours in 17 of 20 cases. Across all squeezes of 2% or more: 14 of 15 before 2026 and 17 of 21 in 2026.
So shorting right after a squeeze means lining up for the next flush.
Don't short right after a short squeeze: the high usually gets taken out.
My forecast for the next cascade
The next cascade will come, the only question is when. And I'm sure about the script: first they push it lower, then within a week they bring it back to where it all started.
So don't buy at market in the panic. Put limit orders lower and let the crowd burn out. And a short after a cascade is just chasing.
So what will you do in the next cascade: catch the knife or wait until they finish the job?
Questions and answers
What happens to the Bitcoin price after long liquidations?
Across the 20 worst cascades, 3 days later the price was above the cascade hour's close in 15 of 20 cases and 7 days later in 13.
Is a liquidation cascade the bottom?
No. In 11 of 20 cases the price was pushed below the cascade low within the next 24 hours.
How fast does price recover after a cascade?
Back to the start level within 24 hours in 12 of 20 cases, within 7 days in 17 and within 30 days in 19.
What happens after a short squeeze?
In 17 of 20 cases the price went above the squeeze hour's high within the next 24 hours. Shorting right after it is a bad idea.
Prices are rounded to hundreds, shares to tenths of a percent. Times are UTC. · Published October 28, 2026
Watch liquidations, whales and the order book of all exchanges live.Free account · no card