Dollar-cost averagingDollar-cost averagingBuying a fixed amount of money’s worth on a fixed schedule — say $200 on the first of every month — instead of trying to pick the moment. You automatically buy more coins when the price is low and fewer when it is high.Full entry in the glossary → is buying a fixed amount on a fixed
schedule and ignoring the price. This runs that plan for bitcoin or ether over every daily close we hold
and puts it next to the alternative everyone argues about: one lump sumLump sumPutting the whole amount in at once, on one date, instead of spreading it over months. It is the comparison every discussion of scheduled buying eventually turns into.Full entry in the glossary → on the start date.
Regular buying
A fixed amount, bought on schedule. Computed on real daily closes (from
18 Aug 2011 for BTC).
Invested (157 buys)
$15,700
Coins bought
0.2496 BTC
Average buy price
$62,899
DCA — worth today
$15,945
DCA result
+$245.34 (+1.6%)
DCA — annualised (IRR)
+1.0%
Lump sum — worth today
$34,169
Lump sum result
+$18,469 (+117.6%)
Lump sum — annualised (CAGR)
+29.6%
The two
totals above are not comparable as rates, which is why the annualised pair is there. With a lump
sum the whole amount works for the whole period, so its rate is a CAGRCAGRThe single yearly rate that would take a starting amount to an ending amount over the period, compounding along the way. It answers “what yearly pace was this?” for money that sat in one place the whole time.Full entry in the glossary →.
With regular buying your last purchase has been working for a week and your first for years, so
the average dollar was invested about half as long. Earning the same profit with money that spent
half the time on the sidelines is a better rate, and the
IRRIRRThe yearly rate that makes a series of payments in and out add up to zero once each is discounted for how long it was invested. It is the right rate for money that went in a bit at a time.Full entry in the glossary → is what measures it.
On one start date the question has no answer: it flips with the date. Across every
1,095-day window in our history — the same length as yours —
the lump sum finished ahead in 77.8%
of 4,364 start dates. Windows overlap heavily, so treat that as
a description of one price history, not a probability. This share is for windows exactly as long
as yours; the FAQ below quotes fixed one-year and four-year windows, so its numbers differ — the
same question asked of different lengths.
What if I had bought
One purchase in the past — what it would be worth today.
$542.15
0.008487 BTC · bought at $117,831 · now $63,882
−45.8%
Value of the DCA position over time · USD
The chart needs JavaScript. The numbers above do not —
they are computed on the server.
What this does not account for
Exchange fees, spreads and taxes are all ignored, and buys are settled at the daily close
rather than at the moment you would have pressed the button. Fees are the big one: a retail
purchase typically costs in the region of 0.1–1.5% on an exchange's professional
order bookOrder bookThe live list of everyone’s offers to buy and sell on an exchange, matched by price. Buying through it is usually far cheaper than a one-tap “buy now” button, which hides its margin in the price.Full entry in the glossary →, and 2–4% in the simplified “buy now” flows, where an
undisclosed spreadSpreadThe gap between the price you can buy at and the price you can sell at. On simplified “buy now” flows it is often wider than the stated fee and is not shown separately, so it is easy to pay several percent without noticing.Full entry in the glossary → sits on top of the visible fee. Check the published rates for your own venue. Past behaviour of an asset is a
description of the past, not a plan for the future —
read what this site refuses to do.
Frequently asked questions
Is dollar-cost averaging better than buying all at once?
For most people the comparison does not apply, and it is worth saying that before the number.
Buying all at once requires having all of it at once. If the money arrives with your salary, in
pieces, then the lump sum on day one was never an option you passed up — it is a different person
with a different bank balance, not the disciplined version of you.
Where both are genuinely open to you, here is what this price history says. Over every one-year
window, a single lump sum on day one finished ahead of weekly buying in 70.2% of
start dates; over four-year windows, in 83.1%. That is what an asset that mostly
rose does to the comparison, and it would say the same about almost any asset that mostly rose.
What averaging in buys you is a smaller worst case and a plan you can keep during a crash —
not a higher expected return.
How far back does the history go?
To 18 Aug 2011 for bitcoin and 9 Mar 2016
for ether — daily closes in UTC from CoinGecko with Binance as a fallback. The start dates differ
because ether did not trade before then, so the form will not let you pick a date the asset has
no price for.
Are fees and taxes included?
No. These are raw daily closes. A retail purchase typically costs 0.1–1.5% on an exchange's
professional order book and 2–4% in the simplified “buy now” flows, where an undisclosed spread sits
on top of the visible fee. On a weekly plan those percentages compound across hundreds of purchases,
so treat every number here as the optimistic edge of the range.
Does it work without JavaScript?
Yes. Every number is computed on the server from the address, so the page answers a plain form
submission and each result has a shareable URL. JavaScript only adds the chart and recalculation
without a reload.
Can I use this to decide what to buy?
No. It describes what one price history did, and the sample is a handful of market cycles, not a
law. Overlapping windows make it look larger than it is. This is not investment advice — see
what this site refuses to do.
