The studies favour one side. Bitcoin's 50% drops are the reason people pick the other anyway.
On the numbers, a lump sum has beaten dollar cost averaging about two times in three, and by more with bitcoin than with stocks. DCA wins when you would have bought just before a crash, and bitcoin's crashes have cut the price by half or more. If your money comes from each paycheque, there is no choice to make: you are already investing as it arrives.
Not financial advice.
If the money is already sitting in your account, putting all of it in on day one has beaten spreading it out about two times in three. The best-known evidence comes from Vanguard's work on stocks and bonds. Cointelegraph's write-up of that study says it covered 1926 to 2015 with a 60/40 portfolio, and that the lump sum finished ahead two-thirds of the time, by 2.4% on average over a calendar year. Wikipedia's page on dollar cost averaging lists academic papers going back to 1979 that reach the same verdict.
The reason has nothing clever in it. Markets rise more often than they fall, so cash that waits on the sidelines misses more gains than losses.
Bitcoin stretches the gap. Writing in Bitcoin Magazine in September 2021, Joakim Book ran a simple test: US$10 a day for five years, starting in August 2016 when one coin cost US$568.40. That adds up to US$18,260. Bought in daily slices, it returned roughly 1,300%. Spent in one go on the first day, the same money returned about 8,400%. After feeding many start dates into calculators, he could not get the daily plan to come out ahead more than about 35% to 40% of the time.
Dollar cost averaging (DCA) wins when the lump sum would have gone in just before a large fall. Book's own article has the example. Someone who bought everything in early December 2017, at prices between US$9,000 and US$16,000, was up 202% to 437% by the time he wrote. A daily plan started that same month was up 452%.
| Start | Buying a little each day | Everything on day one |
|---|---|---|
| August 2016, US$18,260 over five years | about 1,300% | about 8,400% |
| December 2017, near that cycle's top | 452% | 202% to 437% |
With bitcoin, "a large fall" has a specific meaning. Every bear market so far has cut the price by half or more. A stock index that drops 20% makes the evening news. So the one case in three where the lump sum loses is a much worse case here than it is for a balanced fund, and that is the honest argument for spreading out.
DCA does not make losing impossible, though. A CoinDesk explainer from October 2022 worked through a plan of about US$100 a month. After two years, US$2,400 of buying was down about 37%. After three years, US$3,600 was up 20%. After ten, US$12,000 was up more than 5,000%. More recently, Cointelegraph ran the numbers in March 2026 on US$250 a week. Started in January 2021, that plan had put in US$67,500 at an average cost of US$40,884 and was up 76% with bitcoin near US$71,000. Started in January 2024, it had put in US$28,500 at an average cost of US$77,312 and was down about 6%. Same habit, same amount. The start date still decided the result.
Most people asking this are in one of two situations, and the answer differs.
If your money arrives with each paycheque, buying a bit every week or month is simply investing as soon as you have it. You are not choosing DCA over a lump sum, because there is no lump. The studies above do not apply to you; Wikipedia's summary is explicit that they are about windfalls such as an inheritance or an insurance payout.
The real choice exists only for cash you already hold: a bonus, a house sale, savings that have been idle. For that money the lump sum has the better expected result, and spreading it over several months gives up some of that in exchange for a smaller regret if the market drops right after. One study cited on Wikipedia found that 6 or 12 months was the window that balanced the two. Larry Swedroe of Buckingham Wealth Partners, quoted in the Cointelegraph piece, argued for a blend of both.
My view: spreading out is insurance, and insurance has a price. It is worth paying if the alternative is that you put everything in, watch it fall 50%, and sell at the bottom. A plan you can stick with beats a better plan you abandon. Bitcoin Magazine's 2024 example of US$10 a week for five years, US$2,620 in and US$7,913 out, only worked for the person who kept buying through 2022.
Both approaches only decide how you get in. Neither one tells you when to get out, and neither reduces what you hold when a bear market starts.
Markus Thielen of 10x Research made that case in a CoinDesk column in June 2026. He wrote that DCA "offers psychological comfort, not mathematical protection", pointed to three drawdowns of about 80% in bitcoin's history, and said his own cycle-aware approach cut the worst drawdown from 80% to 44% over 15 years of history. That figure is backtested, not a promise, and he sells research. His narrower point holds anyway: someone who bought steadily through 2021 and 2022 still sat on heavy paper losses.
Bitcoin is at about US$83,802 today, or C$119,442 on CoinGecko, inside the US$82,000 to US$85,000 range that CoinDesk described on October 1. Nobody knows whether a lump sum put in this week will look like August 2016 or December 2017. That uncertainty is the entire question, and no study removes it.
Fonte uses neither method in the usual sense. Each vault follows a written halving-cycle rule with one entry date into a BTC and ETH basket, the next one around October 19, 2026. Until then the money sits in Aave USDC lending on Base, at 4.16% today, a variable rate according to DefiLlama. That is closer to a lump sum on a date picked by a rule than to weekly buying. The limits are real: the timing is fit to two or three past cycles and this entry has never been tested live. You can add money any time by Interac e-Transfer through Paytrie, starting with about $100. Paytrie's fee is the larger of CAD 5 or 0.6%, so a few bigger deposits cost less than many tiny ones. Open a vault.
Usually not on returns. A Bitcoin Magazine test found daily buying beat a single purchase only about 35% to 40% of the time. DCA comes out ahead when the single purchase would have landed just before a large fall, which is also when it matters most.
Yes. Cointelegraph's March 2026 example of US$250 a week started in January 2024 had an average cost of US$77,312 and was down about 6% with bitcoin near US$71,000. Spreading purchases lowers the odds of a terrible entry price. It does not remove losses.
One study summarized on Wikipedia found that 6 or 12 months balanced return against risk when delaying a windfall. Longer than that and more of the money sits idle while markets, on average, rise.
Not really. The lump sum studies are about money you already hold, such as an inheritance. Buying every payday is investing as soon as the money exists, which is what those studies favour.
Disclaimer: This article is general information, not financial, investment, legal or tax advice. Crypto assets are volatile and you can lose money, including your entire deposit. Figures quoted from third parties are as reported by those sources on the date shown and can change. Smart contracts can contain flaws.