How Tick Modeling Changes Strategy Backtest Results
Summary
The article compares three MetaTrader 5 tester modes using a range breakout Expert Advisor on EURUSD H1 over the first half of 2016. The strategy tracks whether ticks after a range break occur more often above or below the broken level, enters in that direction after a threshold, and exits after a set number of bars. The comparison shows that the modes produced different trade counts, profits, drawdowns, and runtimes. In this example, the fastest mode showed a gain while both tick-based modes showed losses; the author describes the strategy as unconvincing and unsuitable for live trading.
The explanation links these differences to tick quantity and ordering: simplified bars may omit entry opportunities, generated ticks follow modeled paths, and real tick history can produce different sequences. The article recommends quick modes for early screening and more detailed modes when intrabar execution or tick-dependent logic matters. Real-tick tests take longer, may fall back to generated ticks where history is missing, and still cannot guarantee future performance. The results are an illustration from one strategy and period, not evidence that any mode predicts live outcomes.
Key ideas
- Different tick modeling modes can produce different trades and performance for the same strategy.
- Simplified OHLC testing is faster and can help screen ideas early.
- Generated tick paths and real tick sequences can change entry and exit decisions.
- Real-tick testing is slower and depends on the availability of historical tick data.
- A backtest result from one strategy and period does not establish future profitability.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.