Options backtesting starts with the contract data

Decide which kind of options test you need

A signal test asks whether an underlying-price rule identified useful entry and exit times. A contract backtest asks what happened after selecting a call, put, spread, or multi-leg structure with a specific strike and expiration. A portfolio simulation also tracks buying power, overlapping positions, assignment, and cash. These are different jobs and should not be compared as though they produce the same evidence.

• Signal test: underlying price rules only.
• Contract test: historical option quotes and contract selection.
• Portfolio test: cash, margin, assignment, and overlapping legs.

Evaluate options backtesting tools by their data model

Check whether the tool has bid and ask quotes rather than one theoretical price, preserves delisted contracts, handles corporate actions, and exposes its fill assumptions. For multi-leg trades, align quotes to the same decision time, reject stale observations under a stated rule, and check exercise or assignment handling. A midpoint mark is not proof that a spread could fill there. Theoretical pricing can support a model-based experiment, but label it separately from a test using observed historical option quotes. A free options backtesting tool can still be useful, but missing contract history or hidden fill logic limits what the result can support.

Treat risk as part of the specification

Investor.gov describes options as derivatives whose value depends on an underlying asset and notes that buyers can lose the full premium while some writers can face larger losses. Encode contract quantity, maximum position size, exit timing, and assignment treatment before looking at results. Do not infer option returns from the underlying chart alone.