Backtest routing rules before they go live
A parameterized Python framework that replays historical RFQ data against different order-routing rules and scores each one on hit rate, adverse selection and P&L.
- WhenPick routing rule variants
- ThenReplay historical RFQ data
- ThenScore hit rate, adverse selection
- ResultCompare P&L per configuration
- Python
- Pandas
The problem
Changing an order-routing or pricing rule in production is risky. Before a rule goes live, you want to know how it would have performed on real history.
What I built
- A parameterized framework that replays historical RFQ data against any number of rule variants
- Metrics for every configuration: hit rate, adverse selection and P&L attribution
- Built in Python and Pandas on public data, on my own time
Why it matters
Rule changes can be compared side by side on evidence, before any of them touch production.
Where it fits
- Execution analytics
- Algorithmic trading
- Pricing strategy
- Decision testing