5 Backtesting Pitfalls That Destroy Trading Strategies

A beautiful backtest curve means nothing if the methodology is flawed. Here are the five most common pitfalls that separate profitable strategies from expensive lessons.

1. Overfitting to Historical Data

The most insidious pitfall. If you optimize enough parameters against historical data, you can make any strategy look profitable. The problem: it's curve-fitting noise, not capturing signal.

How Conquest handles it: Walk-forward analysis with strict out-of-sample windows. Your strategy must prove itself on data it's never seen.

2. Survivorship Bias

Testing only against stocks that exist today ignores all the companies that went bankrupt, were delisted, or were acquired. This systematically inflates returns.

How Conquest handles it: Our historical database includes delisted securities, adjusted for corporate actions and survivorship.

3. Ignoring Transaction Costs

A strategy that trades 50 times per day looks very different when you account for commissions, slippage, and market impact.

How Conquest handles it: Realistic commission schedules, configurable slippage models, and market impact estimation based on order size relative to volume.

4. Look-Ahead Bias

Using information that wouldn't have been available at the time of the trade. This includes using adjusted close prices, future earnings dates, or economic data before its release.

How Conquest handles it: Point-in-time data with strict chronological enforcement. If the data wasn't public when the signal fired, it's not in your backtest.

5. Ignoring Regime Changes

A strategy that works in a bull market may catastrophically fail in a bear market. Testing across a single market regime gives false confidence.

How Conquest handles it: Multi-regime analysis that automatically identifies market regimes and tests your strategy across each one separately.