Skip to content

Assumption Execution at Next Day Market Open

ai-lab-projects edited this page May 4, 2025 · 1 revision

Many algorithmic trading backtests assume that:

Orders are executed at the market open on the next trading day after the signal is generated.

This simplifies implementation and avoids lookahead bias. However, this assumption also requires careful validation.

🎯 Is This Assumption Reasonable?

It depends on several factors:

  • Order size relative to daily trading volume
    If your trade size is small compared to the average daily volume, it's likely that you can execute at or near the opening price with minimal market impact.

  • Market liquidity
    In highly liquid markets (e.g., major ETFs or large-cap stocks), opening price execution is a reasonable approximation.

  • Order type
    Market orders might fill immediately, but with slippage. Limit orders may not fill at all.

  • Volatility at market open
    Some assets experience high volatility right after the market opens, making this approximation less accurate.

🔍 Validating This Assumption in Live or Paper Trading

To test this assumption:

  • Compare your backtest results against paper trading using a broker like Alpaca.
  • Log actual fill prices from paper trades and compare them with assumed open prices in your simulation.
  • Monitor fill success rate for limit orders and the slippage for market orders.

🧪 Does Alpaca Paper Trading Reflect Realistic Execution?

Alpaca’s paper trading environment is useful but has limitations:

Feature Alpaca Paper Trading
Market order execution Always fills, at simulated prices
Limit order realism May fill unrealistically
Slippage / liquidity model Not modeled accurately
Market impact simulation ❌ Not supported

Therefore:

Alpaca paper trading is helpful for functional testing but not reliable for execution realism.

To improve simulation fidelity, consider:

  • Using historical bid/ask spreads and order book data if available.
  • Incorporating slippage models into your backtest.

Clone this wiki locally