How a fee you ignore compounds over a year of trading

By The Fee Ledger desk · Updated 5 Nov 2026 · Education only, not advice

Short answer

Trading fees compound because each one is charged on the position size, not on your profit, and you pay it again every time you open or close. A cost that looks trivial on one trade is multiplied by how often you trade, so the total depends more on your turnover than on the rate itself. The compounding part is that every fee comes out of the balance that funds your next trade, so the base you are trading with shrinks as well.

A trading fee compounds because you pay it on position size rather than on profit, and you pay it again on every trade. One fee is negligible. The same fee charged a few hundred times in a year, each time out of the balance that funds the next trade, is not.

Why the fee is charged on size and not on your result

A venue is not a partner in your trade. It provides matching, custody and settlement, and those costs are incurred when an order fills, regardless of what happens to the position afterwards. So the charge is tied to the thing the venue actually did: it moved a certain amount of value for you. The fee is therefore a percentage of notional size, or a flat amount per order, or a widened spread, or some combination.

This has one important consequence. A fee has no relationship to whether you were right. A losing trade pays the same fee as a winning one of the same size. Your profit and loss varies; your fee is close to fixed per unit of turnover. That asymmetry is what turns a small rate into a reliable annual number.

What turnover actually means and why it dominates

Turnover is the total value you have traded, not the amount you have deposited. If you move the same balance in and out of positions repeatedly, your turnover is many times your balance.

Your annual cost is close to:

cost per trade × number of trades

Both terms are multipliers, so they have equal weight in the product. A rate that is twice as high and a frequency that is twice as high do the same damage. People spend attention on the first term, which moves in small increments and is partly set by the venue, and almost none on the second term, which they control completely and which varies by an order of magnitude between a position trader and someone who scalps.

There is a second effect layered on top. Every fee is taken out of the account, so the capital available for the next position is slightly smaller. If you size positions as a percentage of your balance, each fee quietly shrinks every trade that follows it. That is the compounding in the strict sense: the drag is applied to a base that the drag has already reduced.

A worked example with made-up numbers

Every figure below is invented for the purpose of the sum. None of it is a quoted rate.

Suppose a venue charges 0.1% per side. Say you trade a position size of $5,000, and you are flat at the end of each day, so each trade is a round trip: one entry, one exit.

Cost of the entry: 0.1% of $5,000 = $5.
Cost of the exit: 0.1% of $5,000 = $5.
Cost of one round trip: $10.

Ten dollars on a five thousand dollar position is 0.2%. On its own it is inside the noise of a single day’s movement, which is exactly why it gets ignored.

Now apply frequency. Suppose you take three round trips a day, and suppose you trade 250 days in the year.

Round trips per year: 3 × 250 = 750.
Annual fee cost: 750 × $10 = $7,500.
Annual turnover: 750 × 2 × $5,000 = $7,500,000.

If the account funding those positions is, say, $10,000, the annual fee cost is 75% of the account. The rate never changed. Only the count did.

Hold the rate and halve the frequency to 1.5 round trips a day and the annual cost falls to $3,750. Now hold the frequency and imagine the rate is 0.05% per side instead: the annual cost is also $3,750. Identical outcome from either lever, which is the point.

You can run this with your own invented or real inputs in the crypto fee calculator, and if you hold leveraged positions overnight, the recurring financing charge behaves the same way: see the funding rate calculator.

The mistake: judging a fee against the trade instead of the year

The specific error is a denominator error. When you look at a fee, it is displayed beside the position size. Against $5,000, a $5 charge is 0.1%, and 0.1% reads as a rounding error. So the brain files it as immaterial and moves on.

But the fee does not live against the position. It lives against your annual result, and there it appears 750 times. The correct comparison is not fee versus trade size. It is annual fee total versus the profit you expect to make in a year, or versus your account balance. The same $5 that looked like 0.1% becomes, in the example above, a multiple of a year’s realistic return.

Three related errors follow from the same habit:

Counting one side. A fee quoted per side is paid twice per round trip. Half your annual cost is in the exit you did not include.
Leaving out the spread. If a cost is not itemised on a statement, it does not appear in anyone’s mental arithmetic. The spread is paid at the moment of execution and recurs with exactly the same frequency as a commission. Excluding it can halve your estimate or worse. The terms are set out in the glossary.
Assuming a strategy’s cost scales with its holding period. It does not scale with how long you hold. It scales with how many times you trade. A strategy that holds for minutes and a strategy that holds for weeks can share a rate and differ by a hundredfold in annual cost.

How to reason about it the next time you see a fee

Do the multiplication before you look at the rate. Count how many round trips your approach implies in a year, multiply by two sides, and multiply by your typical position size. That gives your annual turnover. Apply the rate to that number, not to one trade. Then set the result next to your account balance.

If the answer is uncomfortable, you now know which of the two terms you can change, and frequency is almost always the cheaper one to move. Comparing published cost structures across venues is a second lever, and the comparison pages exist for that, but it is the smaller one. A better rate on a strategy that trades constantly still produces a large bill. A fair rate on a strategy that trades rarely barely registers.

The structure of the charge matters too: whether you pay as a maker or a taker changes the per-trade term, and that is covered in maker and taker fees, while fee tiers explain why the rate itself can move with turnover.

Education only, not financial advice. Trading carries a risk of loss.

Key facts

A trading fee is usually charged as a percentage of position size, so it does not scale with whether the trade won or lost.
Total annual cost is approximately cost per trade multiplied by number of trades, which makes frequency the dominant variable.
Round trips charge twice: once on entry and once on exit.
Each fee is deducted from working capital, so later trades are sized from a smaller base.
Doubling trade frequency has roughly the same cost effect as doubling the fee rate.
The spread and any overnight financing are part of the per-trade cost even when they are not itemised as fees.

Frequently asked

Is it better to lower my fee rate or trade less often?

Both act on the same product, so the bigger lever is whichever you can change more. Fee rates usually move in small steps and are partly outside your control, while frequency is entirely your decision and can often be halved. If you cut your number of trades in half, you have cut that part of your annual cost in half, with no negotiation involved.

Do fees compound in the same way as compound interest?

The arithmetic is related but runs in the opposite direction. Compound interest grows a balance by a percentage each period; repeated fees shrink it by a percentage each time you trade. Because the next trade is sized from the reduced balance, the drag builds on itself rather than staying flat.

Does the spread count as a fee for this calculation?

For the purposes of working out your real cost, yes. The spread is money you give up at the moment of execution, and it recurs on every trade exactly as a commission does. Leaving it out of the sum is the most common reason a cost estimate comes out far too low.

Why does my cost per trade feel invisible when I check it?

Because it is reported next to the position size, where it looks like a rounding error, rather than next to your annual result, where it would look like a line item. The number does not change; the denominator you compare it to does.

Risk warning: trading is high risk and most retail accounts lose money. Education only, not financial advice.