What is dollar cost averaging?

Buying the same amount, on the same day, every week or every month — whatever the price happens to be. It is the least clever way to buy something, and for most people it is the only one that survives contact with real life.

The idea, in one paragraph

Dollar cost averaging means deciding two things once — how much, and how often — and then not deciding anything again. Say $100 every Friday. Some Fridays the price is high and your $100 buys a small amount. Some Fridays it is low and the same $100 buys more. You never have to be right about which kind of Friday it is. Over enough Fridays you end up owning a pile bought at something close to the average price, without ever having formed an opinion about the price.

That is the whole method. It has no moving parts, which is exactly why it works for people who have jobs and children and other things to think about.

A worked example

Numbers make this obvious in a way that words do not. Suppose you buy $100 of something on the first of the month, four months running, and the price does this:

You spent $400 and you own 6.75 units. That works out at $59.26 each. Now look at the four prices on their own: $100, $50, $40 and $80. Their average is $67.50. You paid meaningfully less than the average price, and you did not do anything smart to achieve it.

The reason is arithmetic, not luck. Because you spend a fixed amount rather than buying a fixed quantity, the cheap months automatically get more of your money and the expensive months get less. Volatility, which everyone treats as the enemy, is quietly working in your favour on the way in.

There is a second thing hiding in that table. The price finished at $80, which is lower than the $100 it started at. And yet your 6.75 units are worth $540 against the $400 you put in. An asset that went down over the period still left you ahead, because of when your money arrived.

Why not just wait for the dip?

Because you have to be right twice, and almost nobody is. Waiting for a lower price means naming a price that counts as low, and then being willing to buy when everything you read says the thing is finished. In practice the people waiting for the dip are still waiting during the recovery, and start buying again near the top, when it feels safe. Feeling safe and being cheap are almost never the same moment.

Dollar cost averaging takes that judgement away from you. You are not predicting anything. You are converting a decision you would make badly, repeatedly, into a decision you make once.

What it does not do

This is where most explanations go quiet, so let us be blunt about it.

• It does not stop you losing money. If the price falls and stays down, you have simply bought all the way down. Averaging in reduces how much you paid; it does not put a floor under what you own.
• It does not make a bad asset good. Spreading your buying over two years does nothing for something that goes to zero. The method controls your timing risk. It has no opinion at all about whether the thing is worth owning.
• It does not remove the hard part. The hard part was never the maths. It is making the ninth purchase after the price has halved, which is precisely when the method is doing its best work and precisely when it feels most stupid.

Is it better than buying all at once?

Often not, and it depends what you mean by better. On an asset that mostly rose, putting the whole amount in on day one has historically beaten spreading it out more often than it has lost, for the plain reason that the money spent longer in the market. If your only measure is the average outcome, the lump sum usually wins.

What averaging in buys you is a smaller worst case. Put everything in the week before a 70% fall and you are looking at a hole that takes years to climb out of, and most people do not climb — they sell somewhere near the bottom and never come back. Spread the same money over a year and the same fall is survivable, because you are still buying through it.

There is also a more honest objection to the whole question: very few people have a lump sum sitting idle. They have a pay cheque every fortnight. If money arrives in instalments, it is going to be invested in instalments, and the interesting question is not whether that is optimal but what it produced. That is the question this site's calculator answers.

Does the schedule matter?

Much less than people expect. Daily, weekly, fortnightly and monthly all end up in roughly the same place over a period of years, because they are all sampling the same price history — just at different resolutions. You can check this yourself in a few seconds: set an amount, run it weekly, then switch to monthly with four times the amount and compare.

What does matter is choosing a rhythm you will actually keep, and matching it to when you are paid. A schedule you abandon after five months is worse than a slower one you keep for five years.

Raising the amount as you go

The calculator's default raises your contribution by 2.5% a year, and there is a reason for it. A flat $100 a week is not a flat effort: prices rise, so $100 in ten years' time will ask less of you than $100 does today. Raising the contribution a little each year keeps the effort level rather than letting it quietly shrink — which is also what most people's saving looks like as their pay drifts up.

2.5% is not an official target. The US Federal Reserve's longer-run goal is 2%. 2.5% is an approximation of what inflation has actually been: US consumer prices rose about 2.5% a year on average between 1990 and 2018, and about 2.6% in 2025. If you would rather model a contribution that never changes, set the box to 0%. Nothing else about the calculation changes. There is a whole article on where that figure comes from.

Reading the result honestly

Two of the calculator's figures are worth understanding before you quote them at anyone.

Return per year is a money-weighted return, not a simple start-to-finish growth rate. Money you added last month has not had the same time to work as money you added in 2018, and a naive growth rate ignores that, which flatters the result — badly, when contributions are rising. The money-weighted figure accounts for when each purchase actually happened. It is usually lower than the headline, and it is the honest number.

The price history starts on 18 August 2011, which is why that is the earliest date you can pick. Daily prices before then come only from Mt. Gox — an exchange that traded thinly, whose records are disputed, and which later collapsed. Returns computed from that data would look better and mean less, so this site does not publish them.

The part you should not skip

Everything above describes a method for buying. It says nothing about whether the thing you are buying is a good idea, and dollar cost averaging into Bitcoin is not a safe activity dressed up in a sensible routine. Bitcoin has fallen more than 70% from a high on several separate occasions, and more than 80% on three of them. Each one lasted long enough to feel permanent. Every figure the calculator produces describes something that has already happened, and none of it is a forecast.

If you want to understand what you would be buying rather than just how to buy it, start with what Bitcoin actually is.