How fees and slippage eat a strategy's return

A strategy makes two per cent a month on paper. The same strategy runs flat when it goes live. The gap is usually not in the market. It is in the arithmetic.

Cost comes in two parts

Commission is a published number. It sits on the exchange’s fee page, it varies by account tier, and it can be known in advance.

Slippage is the difference between the price you expected and the price you got. You send a market order and fill somewhere worse than the top of the book. Nobody publishes that figure; it can only be measured from your own fills.

The two together are what a trade actually costs you. On a single trade the number looks trivial, and that is exactly the problem.

Multiplying a small number

Say a round trip costs 0.2% all in. On its own that is nothing.

A bot taking two trades a day makes forty trades a month. Forty times 0.2 is eight per cent a month. A strategy producing two per cent a month means nothing against that; the cost is four times the return.

Run the same strategy at four trades a month and the cost falls to 0.8%, and the picture changes completely. Trade frequency is not a preference. It is a cost line.

Slippage is not constant

Slippage does not behave like a flat tax. It grows exactly when you can least afford it.

When the book is thin, meaning the spread is wide, slippage rises. In sharp moves the book empties and a market order jumps several levels at once. When your size exceeds the depth at a level, your own order pushes the price against you.

The practical consequence: slippage measured in quiet hours is not the slippage you get on a news print. Testing a strategy only on calm data understates the cost systematically.

Putting cost into a backtest

A backtest with no costs is not a simulation, it is a wish. Three things matter when you add them.

Enter commission at your own account’s rate. The exchange’s lowest tier is usually for high-volume accounts, and it is not yours.

Model slippage by order type rather than as one flat number. A strategy entering with limit orders does not carry the same cost as one entering with market orders. Limit orders cut slippage and create a different cost: the order does not fill and the trade is missed.

Account for partial fills. A large order does not fill at one price. Your average fill is always worse than the top of the book.

The break-even figure

There is one useful calculation and it comes before you write the strategy: what does this need to earn per trade just to cover its costs?

At a 0.2% round trip and a 50% hit rate, the average winning trade has to beat the average loser by at least 0.4 points. Any strategy below that line, however good it looks, is financing its costs over the long run.

That figure doubles as a filter. Strategies whose average target is several times the cost absorb it; strategies whose target sits close to the cost flip to losing on a small change in the exchange’s fee schedule.