What is a Bitcoin drawdown?
A drawdown measures how far Bitcoin has fallen from its previous all-time high.
If Bitcoin reaches an ATH of $100,000 and later trades at $70,000, the drawdown is 30%.
This is different from measuring a decline from a recent local high. The reference point stays at the highest price Bitcoin had ever reached up to that point.
For this research, we calculated the historical ATH using the highest daily price reached so far and measured the drawdown using each day's closing price.
How often did Bitcoin reach different drawdown levels?
We identified 18 separate drawdown cycles of at least 10% during the observed period.
Every one of them gave a 10% dip-buying opportunity. But deeper levels became much less common.
| Drawdown level | Cycles reaching level | Missed cycles | Opportunity rate |
|---|---|---|---|
| −10% | 18 / 18 | 0 | 100% |
| −20% | 10 / 18 | 8 | 55.6% |
| −30% | 6 / 18 | 12 | 33.3% |
| −40% | 4 / 18 | 14 | 22.2% |
| −50% | 4 / 18 | 14 | 22.2% |
Waiting deeper means missing more corrections
A 30% drawdown occurred in only 6 of the 18 cycles we identified. That means a strategy waiting exclusively for −30% would have received no entry at all in two thirds of the observed drawdowns.
Historical buying opportunities by drawdown level
How deep did Bitcoin drawdowns actually go?
Not every Bitcoin correction turned into a deep crash. In fact, most of the drawdown cycles we observed stopped well before reaching the 40–50% range.
Of the 18 drawdown cycles of at least 10% since August 2017, almost half bottomed between 10% and 20% below the previous ATH.
| Maximum drawdown in cycle | Cycles | Share |
|---|---|---|
| 10–20% | 8 | 44.4% |
| 20–30% | 4 | 22.2% |
| 30–40% | 2 | 11.1% |
| 40–50% | 0 | 0% |
| 50%+ | 4 | 22.2% |
Most corrections never became deep crashes
Two thirds of the observed drawdown cycles bottomed before reaching −30%. At the same time, 4 of the 18 cycles went beyond −50%, creating a clear split between relatively shallow corrections and much deeper bear-market declines.
Distribution of Bitcoin drawdown depth
The largest Bitcoin drawdowns since 2017
The deepest drawdowns were not short corrections. Some lasted for months, and recovery to the previous ATH took years.
| ATH date | ATH price | Bottom date | Bottom price | Max drawdown | Recovery | Days to bottom | Days to recovery |
|---|---|---|---|---|---|---|---|
| Dec 17, 2017 | $19,798.68 | Dec 15, 2018 | $3,211.72 | −83.78% | Nov 30, 2020 | 363 | 1,079 |
| Nov 10, 2021 | $69,000 | Nov 21, 2022 | $15,781.29 | −77.13% | Mar 5, 2024 | 376 | 846 |
| Apr 14, 2021 | $64,854 | Jul 20, 2021 | $29,790.35 | −54.07% | Oct 20, 2021 | 97 | 189 |
| Oct 6, 2025 | $126,195.50 | Jun 30, 2026 | $58,631 | −53.54% | Not recovered | 267 | — |
| Sep 2, 2017 | $4,939.19 | Sep 14, 2017 | $3,189.02 | −35.43% | Oct 10, 2017 | 12 | 38 |
| Jan 20, 2025 | $109,900 | Apr 8, 2025 | $76,322.10 | −30.55% | May 21, 2025 | 78 | 121 |
Deep drawdowns can last much longer than they look on a price chart
The 2017–2018 drawdown took 363 days to reach its lowest daily close and 1,079 days to regain the previous ATH. The 2021–2022 cycle took 376 days to bottom and 846 days to recover.
We analyzed recovery separately to see how long Bitcoin historically took to return to its previous all-time high after different crashes. See the full Bitcoin recovery after a crash analysis .
Bitcoin drawdown from ATH over time
What happens after a dip level is reached?
Reaching a deep drawdown does not necessarily mean Bitcoin is close to the bottom.
For every drawdown cycle that reached a given threshold, we measured how much further Bitcoin's daily closing price fell before the cycle reached its eventual bottom.
| Buy threshold | Historical cases | Median further decline | Worst further decline | Median days to bottom |
|---|---|---|---|---|
| −10% | 18 | −9.22% | −81.43% | 10 |
| −20% | 10 | −14.26% | −79.27% | 65 |
| −30% | 6 | −31.95% | −75.90% | 143 |
| −40% | 4 | −40.53% | −70.53% | 225.5 |
| −50% | 4 | −30.21% | −65.18% | 171 |
How much further did Bitcoin fall after reaching each level?
A deep dip was not necessarily close to the bottom
Waiting for a larger drawdown reduced the number of buying opportunities, but it did not eliminate the risk of buying too early. In the four historical cycles that reached −40%, Bitcoin still fell a median 40.53% from the first qualifying daily close to the eventual bottom.
What happened after Bitcoin fell 40%?
A 40% drawdown may sound like an unusually deep discount. Historically, however, reaching that level did not necessarily mean that Bitcoin was close to its bottom.
Since August 2017, four drawdown cycles in our dataset reached at least 40% below the previous all-time high. In every case, Bitcoin continued falling after the first daily close that crossed that level.
| ATH | First close below −40% | Entry price | Bottom | Further decline | Days to bottom |
|---|---|---|---|---|---|
| Dec 2017 | Jan 16, 2018 | $10,900 | $3,211.72 | −70.53% | 333 |
| Apr 2021 | May 19, 2021 | $36,690.09 | $29,790.35 | −18.81% | 62 |
| Nov 2021 | Jan 20, 2022 | $40,680.91 | $15,781.29 | −61.21% | 305 |
| Oct 2025 | Feb 4, 2026 | $73,158.60 | $58,631 | −19.86% | 146 |
Entry price is the closing price of the first daily candle that crossed the −40% threshold. Bottom is the lowest daily closing price reached before Bitcoin established a new all-time high.
Waiting longer did not mean buying near the bottom
The trade-off becomes clear: deeper thresholds produced fewer opportunities, but they did not reliably identify market bottoms.
A −40% threshold was reached in only 4 of the 18 drawdown cycles of at least 10% in our dataset. And when it was reached, Bitcoin subsequently fell another 18.81% to 70.53% before reaching the cycle bottom.
So which Bitcoin dip level should you wait for?
The historical data does not point to one universally optimal drawdown level.
Instead, choosing a threshold means accepting a trade-off. Shallower levels occur much more often, but buying there leaves more room for the market to fall. Deeper levels offer a larger discount from the previous ATH, but they occur far less frequently and can leave capital waiting on the sidelines for long periods.
| Threshold | Cycles reaching it | Opportunity rate | What it means |
|---|---|---|---|
| −10% | 18 / 18 | 100% | Frequent opportunities, but relatively shallow entry. |
| −20% | 10 / 18 | 55.6% | More selective, while still appearing in over half of historical drawdown cycles. |
| −30% | 6 / 18 | 33.3% | A substantially deeper discount, but two-thirds of drawdown cycles never reached it. |
| −40% | 4 / 18 | 22.2% | Rare — and historically still no guarantee of being close to the bottom. |
| −50% | 4 / 18 | 22.2% | Extreme drawdown territory with very few historical observations. |
Why use multiple dip levels instead of one?
The historical trade-off suggests another approach: instead of trying to predict one perfect entry point, capital can be deployed progressively as the drawdown becomes deeper.
For example, an investor might make a smaller purchase at −20%, another at −30%, and deploy more capital if Bitcoin eventually reaches −40% or −50%.
This does not solve the problem of timing the bottom — nothing in this historical data suggests that the bottom can be identified reliably in advance. Instead, it reduces the need to make one all-or-nothing decision about where the bottom will be.
What progressive dip buying looks like
Instead of choosing a single drawdown level and waiting for it, an investor can split available capital across several levels.
One simple example could look like this:
| Bitcoin drawdown from ATH | Purchase | Total deployed |
|---|---|---|
| −20% | $10 | $10 |
| −30% | $20 | $30 |
| −40% | $30 | $60 |
| −50% | $40 | $100 |
If Bitcoin only falls 20%, only the first portion of capital is deployed. If the decline becomes deeper, progressively more capital is invested at lower prices.
The exact levels and amounts are not the point of the example. They can be adjusted depending on how aggressively or conservatively someone wants to deploy capital.
The goal is not to predict the bottom
Progressive buying accepts that nobody knows whether a −20% decline will recover immediately or eventually become a −50%, −70%, or even deeper drawdown. Capital is deployed gradually instead of depending on one prediction.
How Cryptera applies this idea
Cryptera is built around this drawdown-based approach. Instead of buying Bitcoin on a fixed calendar schedule, it monitors the distance from the previous all-time high and can place purchases when configured drawdown levels are reached.
A strategy can use one level or several progressive levels, with different purchase amounts assigned to deeper drawdowns.
This makes the process systematic: the levels and purchase amounts are defined in advance rather than decided emotionally while the market is falling.
We have also backtested different Bitcoin drawdown thresholds to see how they would have behaved historically. See our Bitcoin dip-buying level backtest .
How is this different from DCA?
Dollar-cost averaging invests on a fixed schedule regardless of Bitcoin's price. Drawdown-based buying waits for predefined declines from the all-time high instead.
We tested both approaches using comparable amounts of invested capital. See our Bitcoin dip buying vs DCA comparison .
Methodology
This analysis uses daily BTC/USDT market data from August 17, 2017 through August 21, 2026.
For each daily candle, we calculated Bitcoin's drawdown using the daily closing price relative to the highest price reached before that day. A drawdown cycle ends when Bitcoin's daily high reaches or exceeds the previous all-time high.
We analyzed drawdown thresholds of 10%, 20%, 30%, 40%, and 50%. A cycle is counted as reaching a threshold when Bitcoin's daily closing price falls at least that far below the preceding all-time high.
When measuring what happened after a threshold was reached, we used the first daily close that crossed the threshold as the hypothetical entry price. The subsequent bottom is the lowest daily closing price before Bitcoin reached a new all-time high.
Historical results describe what happened in the available dataset and do not predict future Bitcoin price behavior. The current drawdown cycle may also be incomplete.
Frequently asked questions
What was Bitcoin's largest drawdown?
In our BTC/USDT dataset since August 2017, the largest drawdown was −83.78%. Bitcoin fell from its December 2017 all-time high of $19,798.68 to a daily closing price of $3,211.72 in December 2018.
How often does Bitcoin fall 20% from its all-time high?
In our daily BTC/USDT data from August 2017 through August 2026, 10 of the 18 drawdown cycles that reached at least −10% eventually reached −20%. That means a 20% drawdown occurred in about 56% of the drawdown cycles we analyzed.
How often does Bitcoin fall 30% from its all-time high?
Six of the 18 drawdown cycles in our dataset reached at least −30%, or about 33% of the cycles analyzed. Waiting for a 30% decline therefore produced substantially fewer historical buying opportunities than waiting for a 10% or 20% decline.
Is a 40% Bitcoin drop necessarily close to the bottom?
No. Four historical drawdown cycles in our dataset reached at least −40%, and Bitcoin continued falling after the first daily close beyond that threshold in every one of them. In the most severe case, the price later fell about 71% further from the hypothetical entry price.
Should I wait for Bitcoin to drop before buying?
There is no drawdown level that guarantees a better entry. Shallower thresholds historically created more opportunities, while deeper thresholds deployed capital less often and still did not reliably identify market bottoms. One alternative is to spread purchases across several drawdown levels rather than depending on a single threshold.
What Bitcoin's drawdown history tells us
Bitcoin's history does not reveal a single ideal percentage at which to buy the dip.
Shallow drawdowns happened frequently, but buying early sometimes meant sitting through much deeper declines. Waiting for larger drawdowns reduced the number of opportunities, yet even a 40% or 50% decline did not reliably mark the bottom.
In our dataset, every drawdown cycle that reached at least −10% offered a 10% entry opportunity. Only about 56% reached −20%, 33% reached −30%, and 22% reached −40% or −50%.
This is the trade-off behind drawdown-based investing: buying earlier gives more opportunities but exposes capital sooner, while waiting for deeper declines may provide lower prices but can leave capital undeployed for long periods.
Rather than trying to identify one perfect entry point, a progressive approach can distribute capital across several drawdown levels. It does not predict the bottom — it defines in advance how much capital to deploy if the decline becomes deeper.
Automate a drawdown-based Bitcoin strategy
Cryptera monitors Bitcoin's drawdown from its all-time high and can automatically place purchases at the levels you configure.
You choose the drawdown levels, purchase amounts, and strategy parameters. Cryptera follows those rules when the market reaches them.
Follow Cryptera on TelegramFurther reading
This article is a historical backtest for research and educational purposes only. Past performance does not guarantee future results. It does not include trading fees, slippage, taxes or other execution costs and should not be considered financial advice.