Plenty of brokers claim that slippage works both ways, positive and negative. Easy to say. Easy to test too, and the results are sometimes ugly.
Take a few hundred market orders and log the requested price against the executed price. Split them into positive slippage, zero slippage, and negative slippage. A fair setup shows roughly balanced size and frequency on both sides, especially in fast conditions.
A suspicious setup shows negative slippage averaging 0.8 pips and positive slippage averaging 0.1, with positive fills being rare. That pattern implies asymmetry somewhere in the chain, whether from the price feed, a fill delay, or a policy that passes on bad fills but not good ones.
There's a subtle twist: some brokers do pass positive slippage on market orders but cancel it on pending orders, or the reverse. Test both order types separately, since the behavior often differs.
The sample size matters. Thirty trades prove nothing. Three hundred starts to say something. Over a thousand and the pattern is hard to argue with.
Anyone with a spreadsheet and some patience can do this audit in a weekend. It's more useful than another indicator test.
Did the slippage on an account here ever turn out to be one-sided, and how did the broker respond when asked?
Slippage should be symmetric. Check whether it is.
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
-
FXS Prop Desk
- Posts: 1689
- Joined: Wed Oct 07, 2026 9:52 am
Re: Slippage should be symmetric. Check whether it is.
The split into positive, zero and negative is the right test. One extra step makes it stronger: also split by trade direction relative to recent movement. Asymmetric slippage often hides in orders placed in the direction of a fast move.
FXS Official Team | The Idea Lab & Strategy
-
Amanhasnoname
- Posts: 2060
- Joined: Fri Oct 30, 2026 9:45 am
Re: Slippage should be symmetric. Check whether it is.
I ran almost exactly this test on a demo and a live account side by side last year, and the difference between the two was the eye opener. On demo the fills were nearly perfect in both directions, which told me nothing. On live, over roughly four hundred trades, the losing side of slippage was about twice as frequent as the winning side, and it clustered in the first few seconds after data releases.
What I didn't expect was how much the order type mattered. My market orders were reasonably balanced once I removed the news minutes. The stop orders were where it leaned against me, almost every triggered stop filled a fraction worse and I could count the improved fills on one hand.
When I sent the spreadsheet to support, the reply was a polite paragraph about liquidity conditions and nothing else. I moved the bulk of my trading elsewhere a month later. The new account isn't perfect either, but the log now shows improvements turning up regularly, which is all I really wanted to see.
What I didn't expect was how much the order type mattered. My market orders were reasonably balanced once I removed the news minutes. The stop orders were where it leaned against me, almost every triggered stop filled a fraction worse and I could count the improved fills on one hand.
When I sent the spreadsheet to support, the reply was a polite paragraph about liquidity conditions and nothing else. I moved the bulk of my trading elsewhere a month later. The new account isn't perfect either, but the log now shows improvements turning up regularly, which is all I really wanted to see.