How often does crypto crash? The historical frequency of 50% falls

How do you count a 50% fall in crypto history?

Fix the rule before you count. A common rule measures from a high to the lowest price that follows it, and it counts the episode when that drop reaches 50%. The episode ends when the price returns to the old high, so one long decline counts once and not once per month. Write the rule down, and any reader can repeat your count.

Choose the price series as well. Daily closing prices smooth out sudden spikes, while intraday prices catch brief dips that may reverse within minutes. Each choice is reasonable, and each gives a slightly different tally. State yours, and the frequency you report becomes a fact about the rule and not a matter of opinion.

Which Bitcoin falls of 50% or more are documented since 2017?

Three episodes stand out in the record of Wikipedia’s History of bitcoin (consulted 23 October 2026). After a high of $19,783.06 on 17 December 2017, the price dropped 50% in 16 days by 5 February 2018, and it dipped below $3,300 on 7 December 2018, a 76% drop from the previous year. In early 2020 the price dropped 50%, including a 25% loss in 24 hours.

The third episode runs from late 2021 into 2022. Bitcoin reached $66,975 on 20 October 2021, fell almost 50% from its highs to below $35,000 by 22 January 2022, and stood at $16,216 on 28 November 2022 after the collapse of FTX. Together these show at least 3 falls of 50% or more in about 5 years.

Why does the crypto crash count change with the rule?

The threshold matters most. At 30% the list grows, since ordinary monthly swings in Bitcoin can touch that level. At 75% the list shrinks to the deepest episodes. A frequency quoted without its threshold is not comparable with any other.

The window matters next. A fall of 50% that unfolds over 16 days differs from one that unfolds over 12 months, even if the end price is the same. Faster falls force more liquidations and leave less time to react. When you read a claim about frequency, ask for the threshold, the window, and the price series.

How long are the gaps between crypto crashes?

The documented gaps are uneven. About 2 years passed between the fall that reached its first 50% mark in February 2018 and the drop of early 2020. The next large decline began after the October 2021 high. Uneven gaps are the usual pattern in markets driven by leverage and sentiment, and they make a calendar rule unreliable.

A planner should therefore think in terms of exposure to a possible fall, with no expected date attached. A holder who commits funds for 5 years has met at least 3 deep falls in the recent record, and a holder who commits for 3 months may meet none. The length of your horizon decides how much of the history applies to you.

Do other crypto assets crash more often than Bitcoin?

Smaller tokens usually have thinner markets, so the same selling pressure moves them further. A shock that cuts Bitcoin in half can cut a small token far more. The frequency of deep falls therefore rises as you move down the list by size, and the count for Bitcoin serves as a lower bound for the market.

Records for thousands of tokens are harder to verify, so this page limits its count to Bitcoin. If you hold other assets, repeat the counting rule on their own price history, using the same threshold and window. A like-for-like count tells you whether your holdings fall more often than the benchmark.

What does a short crypto history say about crash frequency?

Bitcoin has a short price history, so a handful of episodes dominate the count. With 3 episodes in about 5 years, a calculated frequency can swing widely if one more event arrives or a single one is reclassified. The sound response is a wide range, such as 1 fall every 1 to 3 years, with a note that the range comes from a small sample.

A small sample also limits what you can claim about causes. Each episode had its own trigger, such as the end of a speculative run or the failure of a large exchange. Treat the pattern as evidence that deep falls occur in this asset class, and treat the date of the next one as unknown. The pillar page on crypto crash gathers the causes and the history.

How can you plan around how often crypto crashes?

Use the frequency to size what you can carry. If falls of 50% have occurred at least 3 times since 2017, a plan that assumes none will fail under ordinary history. Set the position so a fall of that size leaves your obligations intact. This page is general information, not investment advice.

The EQOS Asia Λ simulator gives a first structural reading for your own case, and the pricing page lists the diagnostic options for teams that want a full review. A 30-minute session through the contact page applies the same method to your exposure, with a ranked set of options as the result.

Frequently asked questions

About the author

Stéphane Jambu

President, EQOS Forecast Asia

Stéphane Jambu leads EQOS Forecast Asia and works as a consultant in e-commerce, search visibility, and AI. LinkedIn profile