How Tick-Level Backtests Differ from Bar-Based Backtests
Summary
This overview explains how FMZ’s backtests advance through historical data and compares bar-based testing with two tick-oriented modes. In a traditional on-bar setup, each candle creates one decision point, typically using its open, high, low, close, and volume. That is fast and simple, but it cannot reveal the order of intrabar price movements or represent multiple fills within a candle very well.
FMZ’s simulation mode derives intermediate price points from a shorter underlying candle interval, giving a strategy more decision points while retaining more speed than real tick data. The article illustrates this with an hourly strategy using five-minute underlying data and reports 4,032 simulated points for one day, compared with 24 on-bar points. Real-market mode uses actual ticks with intervals as short as one second, but runs more slowly and is constrained by data volume. These modes improve price-path detail to different degrees; simulated ticks remain estimates. Even real-tick tests omit factors such as full trade history, changing order-book depth, and live network delays, so backtests cannot fully reproduce bot trading.
Key ideas
- On-bar backtests advance once per candle, making them fast but unable to reconstruct the order of intrabar prices.
- Simulation mode creates intermediate price points from a shorter underlying candle interval.
- Real-market mode uses observed ticks at intervals as short as one second, with greater data demands and slower runs.
- Neither backtest mode fully captures live trading conditions, including order-book changes and network delays.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.