03 — Backend & Data In progress
Trading Backtesting Engine
- Year
- 2026
- Tools
- Python, PostgreSQL, APIs, Docker, Testing
An engine that replays historical market data through a trading strategy, simulates the orders it would place and measures how that strategy would have performed.
Architecture
- 01 Historical data Price history from market-data APIs
- 02 Data processing Cleaned and aligned into one consistent series
- 03 Strategy engine Turns each new bar of data into buy and sell signals
- 04 Order simulation Fills those signals as simulated orders
- 05 Portfolio engine Tracks cash, open positions and total value
- 06 Performance analysis Scores the run with the metrics below
What it measures
- Return
- Total change in portfolio value over the test period
- Win rate
- Share of closed trades that made a profit
- Max drawdown
- Largest fall from a peak before a new high
- Sharpe ratio
- Return earned per unit of risk taken
- Volatility
- How much returns swing from period to period
- Trades
- How many trades the strategy executed
Case study
Status
The engine is in development. Results from real backtests, the test strategy and design notes will be added here as the project reaches its first release.