Slippage does not scale linearly with lot size

[Amanhasnoname]

A common assumption is that if 0.5 lots fills with 0.1 pip of slippage, 5 lots will too. Sometimes it holds. Often it does not, and the break point is worth finding before it finds you.

Retail execution typically draws on a stack of quotes from liquidity providers, with a given size available at each price level. Small orders sit at the top of the stack. Bigger ones walk down it, so the average price worsens as size grows. Some brokers aggregate and some route in whole lumps to a single provider, which changes the shape, but the direction is the same.

A decent way to find it: trade the same pair, same hours, in steps. 0.5, 1, 2, 5 lots, a dozen fills at each step, then compare average slippage and the worst case against size. Expect a flat part, then a bend. If the bend sits right where size goes up for the next risk tier, there is a reason to split orders or scale in.

Be careful with splitting, though. Three orders of 2 lots sent back to back are not independent, because the first one moves the book for the others, and the extra spread paid on each is a cost too.

Could anyone share a slippage-versus-size table from their own fills, even a rough one?