Simulating Last Look: The Broker's Hold Window in 50 Lines of Python
Last look is a short window where the other side can accept or reject your order after seeing it. Here is a tiny simulation that shows how symmetric and asymmetric versions differ. tags: python, trading, finance, simulation series: The Execution Layer cover_image: canonical_url: In the last post I showed how to detect one-sided slippage in your own fill log. This time I want to build…
This article demonstrates how to model a broker's "last look" hold window in Python, showing the difference between symmetric and asymmetric versions. The "last look" is a brief window where the counterparty decides whether to accept, reject, or re-price an order after it has been sent. The simulation shows how the fill rate and realized slippage differ between symmetric and asymmetric last look.
In symmetric last look, the same tolerance is applied to both price movements, resulting in a fill rate close to 100% and minimal client slippage. However, in asymmetric last look, only orders that favor the other side are accepted, leading to a significantly lower fill rate and negative client slippage.
By comparing the simulated results to real logs, one can determine if the broker is treating fills fairly or selectively accepting/rejecting orders based on their potential impact on the broker's bottom line. This toy model helps uncover hidden asymmetries in the execution process, providing valuable insights for traders and regulators alike.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
