Published Jun 2026·By Bitcoin Verdict Editorial Team

Dollar-Cost Averaging Into Bitcoin During a Pullback: What Past Dips Actually Returned

A historical backtest of DCAing through Bitcoin's major drawdowns. We walk six real dips from 2017 to 2025 and show, with stated assumptions, how a disciplined buyer's cost basis and outcome compared to a lump-sum buyer at the top.

The hardest moment to keep buying Bitcoin is the moment the math says you should. Price is down 30%, 50%, sometimes 80%, the headlines are grim, and your existing position is underwater. This is precisely when dollar-cost averaging (DCA) does its best work, and precisely when most people stop doing it.

We have a full guide to how DCA works and why it beats market timing. This post is the companion backtest: instead of arguing the theory, we walk Bitcoin's six biggest drawdowns since 2017 and look at what a disciplined DCA buyer would have actually experienced versus someone who put a lump sum in at the peak. The point is not to sell DCA as a magic trick. It is to show, with real drawdown magnitudes and honest assumptions, what averaging through a dip has and has not done historically.

How to Read These Backtests

Everything below is an illustrative backtest, not a guarantee and not a precise track record. The assumptions are deliberately simple and stated up front so you can see exactly what is and is not being claimed:

Fixed-amount DCA: the same dollar amount is bought on a regular cadence (weekly or monthly) across a defined window. A fixed dollar buy purchases more Bitcoin when the price is low and less when it is high, which is the entire mechanism.
Defined windows: each example uses the documented peak, the documented trough, and a stated recovery point. We are not cherry-picking the perfect entry. The DCA buyer starts buying at or near the peak, the worst possible time to start, and keeps going.
The comparison buyer deploys an identical total amount as a single lump sum at the peak.
Approximate prices. Bitcoin's peaks and troughs are public record but vary by exchange and by the hour. We use widely reported round figures and label them approximate. We do not quote precise return percentages to the decimal, because that precision would be false.

The drawdown magnitudes themselves are real and sourced. The "what DCA would have returned" framing is the illustration.

The Six Dips

Peak	Peak Price (approx)	Trough (approx)	Drawdown	Type
Dec 2017	$19,700	~$3,200 (Dec 2018)	~84%	Cycle top
Feb-Mar 2020	~$10,000	~$3,800 (Mar 2020)	~55%	Macro shock
Apr-May 2021	~$64,800	~$30,000 (Jul 2021)	~50%	Mid-cycle
Nov 2021	~$69,000	~$15,500 (Nov 2022)	~77%	Cycle top
Mar 2024	~$73,800	~$52,500 (Aug 2024)	~29%	Mid-cycle
Oct 2025	~$126,000	TBD	~20-40% so far	Ongoing

These six cover the full range: two catastrophic cycle-top crashes, one macro panic, two ordinary mid-cycle corrections, and the live drawdown investors are sitting in as of mid-2026. This reconciles with our below-all-time-high historical breakdown, which uses the same peak and trough figures.

Dip 1: The 2017 Cycle Top (~84% drawdown)

Bitcoin peaked near $19,700 in December 2017 and bottomed around $3,200 a full year later in December 2018. A lump-sum buyer at $19,700 was down roughly 84% at the worst point and did not see their entry price again until late 2020, about three years later.

A DCA buyer who started at the December 2017 peak and bought a fixed weekly amount straight down into the December 2018 bottom accumulated coins at an average cost dramatically below $19,700, because the overwhelming majority of those weekly buys happened at prices between $3,200 and $8,000. Illustratively, a fixed-amount weekly DCA across that twelve-month slide would have produced an average cost basis somewhere in the high single-digit thousands, not near the peak. That buyer was back above water well before the lump-sum buyer, and once Bitcoin reclaimed its old high in late 2020, the averaged position was sitting on a large unrealized gain while the lump-sum buyer had merely broken even.

The caveat is real: this required continuing to buy through a 12-month decline that felt, at the time, like Bitcoin might be going to zero. Most people did not.

Dip 2: The March 2020 COVID Crash (~55% drawdown)

This one was fast. Bitcoin fell from roughly $10,000 in February 2020 to about $3,800 in mid-March, including a single day where it lost close to half its value. Recovery was also fast: Bitcoin reclaimed its pre-crash level within a few months and went on to a new all-time high by December 2020.

For a DCA buyer, the speed cut both ways. The dip was so brief that a weekly or monthly buyer only caught a handful of buys at the lows before price recovered, so the cost-basis improvement was smaller than in a long grinding bear market. But the recovery was so quick that almost any disciplined buyer in that window came out ahead within the year. The lesson here is the opposite of 2017: short, sharp crashes give DCA fewer cheap buys, but they also resolve faster, so staying invested mattered more than the averaging itself.

Dip 3: The May 2021 Mid-Cycle Pullback (~50% drawdown)

Bitcoin ran to roughly $64,800 in April 2021, then roughly halved to around $30,000 by July 2021 on the back of the China mining ban and a leverage flush. It then recovered to a fresh high near $69,000 by November 2021, only about six months after the dip began.

A lump-sum buyer at $64,800 spent the summer underwater but was whole again by autumn. A DCA buyer through that summer bought a meaningful share of coins in the $30,000 to $40,000 range and was therefore in profit faster, and by the November recovery was ahead of the lump-sum buyer. This is the textbook mid-cycle case: a scary drop, a relatively quick recovery, and DCA quietly lowering the average entry along the way.

Dip 4: The 2021-2022 Bear Market (~77% drawdown)

This is the dip the DCA strategy page already references, and the most instructive of all. Bitcoin topped near $69,000 in November 2021 and ground down to roughly $15,500 by November 2022, through the Terra/Luna collapse, the Three Arrows blowup, and the FTX failure. A lump-sum buyer at $69,000 was down more than 75% at the bottom and underwater for over two years.

A fixed-amount DCA buyer who started at the $69,000 top and kept buying weekly all the way down captured the entire decline, including the $15,500 bottom and the long base that followed. Because so many buys landed in the $16,000 to $25,000 zone, that buyer's average cost basis ended up far below $69,000, illustratively somewhere in the $30,000s to low $40,000s depending on exactly when they started and stopped. When Bitcoin reclaimed $69,000 in early 2024, the lump-sum buyer was finally back to even after more than two years, while the DCA buyer was sitting on a substantial gain. This is the single clearest historical case for averaging through a dip rather than betting everything at the top.

Dip 5: The 2024 Mid-Cycle Dip (~29% drawdown)

After a new high near $73,800 in March 2024, Bitcoin pulled back to roughly $52,500 by the summer, around a 29% drawdown, on Mt. Gox distribution fears and German government selling. It recovered to new highs by late 2024.

A 29% dip is mild by Bitcoin standards. The DCA improvement on cost basis was correspondingly modest, and a lump-sum buyer at $73,800 only had to wait a matter of months to recover. The honest read: in shallow, fast-recovering pullbacks, DCA and lump-sum land close together. The averaging advantage scales with the depth and duration of the drawdown.

Dip 6: The Live 2025-2026 Drawdown (~20-40% so far)

Bitcoin made an all-time high around $126,000 in October 2025 and, as of mid-2026, has traded 20-40% below that depending on the day. There is no trough and no recovery point yet, so there is no completed backtest to run. What the prior five dips suggest is conditional, not predictive: if this resolves like the 2021 or 2024 mid-cycle pullbacks, the DCA advantage will be modest and the recovery measured in months. If it turns into a full cycle bear like 2018 or 2022, the DCA advantage over a top buyer will be large, but only for someone who keeps buying through the pain. Which scenario this is remains genuinely uncertain, as our on-chain bottom-reading guide lays out.

What the Six Dips Have in Common

Pulling the backtests together, a few patterns hold across all of them:

DCA lowered the average entry in every drawdown, and the deeper and longer the dip, the bigger the effect. The 84% 2017 crash and the 77% 2022 bear gave the largest cost-basis improvements; the 29% 2024 dip gave the smallest.
DCA beat the lump-sum-at-top buyer on time-to-breakeven in every case. Because the average buyer's entry was below the peak, they returned to profit before the buyer who paid the top tick. This is mechanical, not lucky.
DCA did not beat a perfectly-timed bottom buyer, and never claims to. No one consistently buys the exact bottom. DCA's pitch is that it removes the need to, not that it wins a hindsight contest against the luckiest possible trade.
Staying invested mattered as much as the averaging. In the fast 2020 and 2024 recoveries, simply not selling did most of the work. The averaging shines most in long bears, exactly when conviction is hardest to hold.

The Honest Caveats

None of this makes DCA a guarantee.

Past drawdowns all recovered to new highs. A future one is not required to. Every Bitcoin dip so far has been followed by a new all-time high, but "so far" is doing real work in that sentence. A regulatory shock, a structural failure, or a multi-year deflation could break the pattern.
DCA only works if you actually keep buying. Every backtest above assumes the buyer did not pause during the scary part. Pausing DCA during a dip and resuming after recovery inverts the whole strategy, and it is the most common mistake real investors make.
Cost basis is not the same as profit. A low average entry helps, but if you need the money during the drawdown, or you sell at the bottom out of fear, the math on paper never reaches your account.
Volatility is the price of admission. Bitcoin's drawdowns are larger and faster than almost any traditional asset. DCA smooths the entry; it does not remove the volatility you have to sit through.

The Bottom Line

Across six real drawdowns from 2017 to 2025, a disciplined fixed-amount DCA buyer who kept purchasing through the dip ended up with a lower average cost basis and a faster path back to profit than a buyer who deployed the same money in a lump sum at the peak, with the advantage growing the deeper and longer the drawdown ran. That is the mechanical reality of buying a fixed dollar amount as the price falls.

What history cannot promise is that the next dip recovers at all, or how long it takes. DCA reduces timing risk and smooths your entry. It does not remove Bitcoin's volatility or guarantee a positive outcome. The strategy works for the investors who can run it through the part that hurts, which, by the numbers, is most of the people who quit right before it would have paid off.

For the mechanics of setting up a recurring buy and which platforms do it well, see our Bitcoin DCA strategy guide.

Data as of June 17, 2026. Drawdown magnitudes and dates are approximate and drawn from public price history (peak and trough figures vary by exchange). The DCA-versus-lump-sum figures are illustrative backtests using the stated assumptions, not a precise track record.

Related reading:

Not financial advice. Past performance does not guarantee future results.