Paper vs live: why Option Alpha bot results differ

Updated October 2026. Educational.

A common question in the Option Alpha community goes: "this bot crushed it on paper, so why not live?" (see the Best Bots thread). Here are the usual reasons.

1. Fills and slippage

Paper fills are simulated. Live orders compete for real liquidity, and multi-leg 0DTE orders can fill worse or not at all. One Reddit user noted that live profits can be lower "due to slippage and not getting filled". The Flat Flyer author even tested a pessimistic $0.10 entry slippage case.

2. Commissions, exchange and assignment fees

Small-premium strategies are fee-sensitive. Option Alpha notes that some brokers charge assignment fees per leg (for example Tradier) and others do not (TradeStation). Strategies held to expiration can rack these up.

3. You changed the bot

Option Alpha says its leaderboard stats reflect the original template settings, not edits made after cloning. Different risk sizing, start dates or exits give different results.

4. Backtest windows and overfitting

One community member reported a 1-year backtest showing strong returns while 3- and 5-year backtests went negative. Short samples flatter a strategy.

5. Timing and start date

Two traders running the same bot from different dates can see very different paths, especially with low-trade-count strategies.

Help build the dataset

TradingBotScout collects anonymous paper-vs-live reports so everyone can see the gap. Submit your results