Blog Research
Research
Backtesting Mistakes That Fool Investors
Five backtesting mistakes that fool investors: ignored costs, curve-fitting, lucky windows, and selling simulations as promises.
Backtests are useful the way flight simulators are useful: they teach you where a plan breaks. They become dangerous when someone sells the simulation as a weather forecast for next quarter.
Mistake 1 — Ignoring costs until the equity curve looks pretty
Brokerage, taxes, impact, and slippage are not footnotes. A strategy that “works” only in a frictionless file is a story about the file, not about live markets. Ask what cost assumptions were used before you care about the headline return.
Mistake 2 — Fitting the past until it confesses
If you try enough rules on the same window, something will look brilliant. That brilliance often dies on the next regime. Healthy research kills fragile ideas early. Unhealthy marketing keeps them alive with more parameters.
Mistake 3 — One lucky period, one lucky symbol
A single bull tape or a single index week is not a career. Ask whether the idea was checked across quieter markets, sharper moves, and more than one environment — still without treating any of that as a guarantee.
Mistake 4 — Confusing research hygiene with a product promise
SEBI-registered brokers are not authorised to offer fixed or guaranteed returns. A backtest PDF does not change that. If the salesperson needs the word “assured”, you are no longer talking about research.
Mistake 5 — Skipping look-through after go-live
Even a carefully researched strategy needs live oversight: contract notes, alerts, and platform visibility. Screenshot track records from someone else’s terminal are not your audit trail. See algo look-through and the look-through checklist.
How Airalgo talks about backtests
For curated strategies, backtesting belongs in the selection process — a reason to keep or reject — not a billboard. Live trading is meant to sit on your account through Lotusdew Securities, with records you can reconcile. More context: why backtesting matters and regulated algo trading.
Questions before you fund? Book an appointment.
Past or simulated performance is not indicative of future results. Investments in securities market are subject to market risks; read all related documents carefully before investing.
Lotusdew