Slippage: always against, or just remembered that way?
Posted: Fri Oct 30, 2026 7:37 am
Every forum has the complaint: slippage always goes against the trade. Some of that is bias, since bad fills stick in memory and good ones evaporate. Some of it can be structural.
In a fast market a market order fills wherever liquidity is when it arrives. If price is running toward the trade, some execution models reject the order or demand the original quote. If price is running away, it fills at whatever is there. That asymmetry is baked in for certain setups.
An experiment settles it. For a month, log every market order: requested price, filled price, and whether price was moving with or against the trade. Count positive versus negative slippage events.
Balanced numbers suggest honest execution. Heavily skewed ones are worth bringing to the broker, with a spreadsheet instead of anger.
One more wrinkle worth tracking: time of day. Slippage during the London open is a different animal from slippage at 3am Tokyo time, and mixing them in one average hides the real picture.
Has this test ever been run? What did the split look like?
In a fast market a market order fills wherever liquidity is when it arrives. If price is running toward the trade, some execution models reject the order or demand the original quote. If price is running away, it fills at whatever is there. That asymmetry is baked in for certain setups.
An experiment settles it. For a month, log every market order: requested price, filled price, and whether price was moving with or against the trade. Count positive versus negative slippage events.
Balanced numbers suggest honest execution. Heavily skewed ones are worth bringing to the broker, with a spreadsheet instead of anger.
One more wrinkle worth tracking: time of day. Slippage during the London open is a different animal from slippage at 3am Tokyo time, and mixing them in one average hides the real picture.
Has this test ever been run? What did the split look like?