Three Hidden Assumptions That Can Distort Quantitative Backtests
Summary
This article explains how seemingly routine backtesting choices can make a strategy look more reliable than it is. It focuses on three assumptions: orders fill immediately at the modeled price, market data and trading opportunities form a continuous sequence, and prices such as a bar’s high or close were available when the strategy acted. In live markets, available liquidity, order size, halts, contract rolls, gaps, and the timing of bar completion can all change whether a simulated trade was possible.
The discussion is conceptual and offers examples rather than empirical measurements or a specific strategy test. It warns that fills at idealized prices can inflate results, gaps can conceal missed signals or impossible paths, and using completed-bar information too early can introduce look-ahead bias. The article’s practical lesson is to identify and challenge the assumptions behind a backtest, especially when timing matters. Its scope is selective: it highlights execution, continuity, and information timing, without presenting a complete validation checklist or quantifying the impact of each issue.
Key ideas
- A backtest that assumes immediate fills may overlook liquidity, order size, and realistic execution prices.
- Treating market data as continuous can hide trading halts, contract rolls, gaps, and missed opportunities.
- Using a bar’s final high or close before that information is available can create look-ahead bias.
- These assumptions may inflate simulated performance, particularly for strategies sensitive to entry timing.
- Backtest results are more interpretable when their execution and data assumptions are made explicit.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.