forex-scalping.com

Average slippage numbers look small. "0.2 pips average slippage on stops" sounds fine. But averages hide the tails, and for scalpers the tail is where accounts get hurt.

Example of a realistic slippage distribution on stop orders over 500 trades:
- 80% filled at the requested price or within 0.1 pips.
- 15% slipped 0.2–1.0 pips.
- 4% slipped 1–3 pips.
- 1% slipped 3–15 pips (almost all around news or sudden headlines).

Average: maybe 0.3 pips. But that 1% bucket contains the trades that turn a normal week into a bad one.

What this means in practice:
- Measure slippage as a distribution, not just an average. Look at the 95th and 99th percentile.
- Identify when the tail events happen. If they cluster around news, flat windows will cut them.
- Size positions so the 99th percentile slippage event doesn't exceed your daily loss limit.

A quick exercise: sort your stop fills by slippage, look at the worst 5. What time did they happen?

A slippage report sounds like a spreadsheet task, and it is. What makes it useful is the choice of what to include.

Start with requested price against filled price for every market order, in pips, signed so that positive means better than requested. Add the time of day, the pair, the lot size, and whether the trade happened within five minutes of a news event.

Then split the results. Average slippage overall is nearly useless because quiet periods dilute the news-time damage. Looking separately at calm periods, news periods and rollover shows where the cost lives.

Also look at the shape. A symmetrical spread of slippage around zero suggests honest execution. A big pile at zero and a long tail on the negative side suggests something else.

Stops need a separate column, because stop fills usually slip more than entries and the difference matters for the total cost of the strategy.

An hour of work with a few hundred trades gives a clearer view of real trading costs than any broker comparison website.

What did the first slippage report show, and was anything in it unexpected?