Execution at bigger size: where it starts to break
Posted: Tue Oct 27, 2026 6:08 am
A strategy that fills perfectly at one lot can behave differently at twenty. Liquidity at the best price is finite, and a large order walks through the book.
Signs of trouble: slippage that grows faster than position size, partial fills, increasing rejection rates, and a widening gap between demo and live performance.
Retail brokers handle size differently. Some fill large orders at the best available price with a slight markup. Others split orders across providers, showing an average fill price. Still others limit maximum order size.
Testing size increases in stages reveals the point where costs start to bite. A table of average slippage by lot size shows where the curve bends.
Splitting orders into smaller pieces sometimes helps, though it adds complexity and may reveal the intention to the market.
A smart habit is logging slippage against lot size on a scatter plot. A flat line means execution isn't affected. A line that curls upward points to the size where trouble begins, and that point becomes a practical ceiling for the strategy's capacity.
At what size did execution quality start to degrade, and what workaround, if any, helped?
Signs of trouble: slippage that grows faster than position size, partial fills, increasing rejection rates, and a widening gap between demo and live performance.
Retail brokers handle size differently. Some fill large orders at the best available price with a slight markup. Others split orders across providers, showing an average fill price. Still others limit maximum order size.
Testing size increases in stages reveals the point where costs start to bite. A table of average slippage by lot size shows where the curve bends.
Splitting orders into smaller pieces sometimes helps, though it adds complexity and may reveal the intention to the market.
A smart habit is logging slippage against lot size on a scatter plot. A flat line means execution isn't affected. A line that curls upward points to the size where trouble begins, and that point becomes a practical ceiling for the strategy's capacity.
At what size did execution quality start to degrade, and what workaround, if any, helped?