The nine biases Fincanva helps you avoid

How does Fincanva reduce look-ahead bias?

Look-ahead bias is using information in a test that would not have been available at the time the decision was made — for example acting on a price or a result before it was actually known. A Fincanva backtest works against this by replaying your strategy's rules in chronological order, using only the data available up to each point in time as it moves forward through history (a walk-forward replay). It runs to the latest available market close and never into the future, so a decision at any date can only rely on what was known by that date. See How backtesting works for the replay in more detail.

How does Fincanva handle survivorship bias?

Survivorship bias is testing only on the instruments that survived to today, ignoring the companies that failed, merged, or were delisted — which flatters results because the losers have been removed. Fincanva's data removes it by construction: delisted instruments are retained rather than dropped, each instrument participates only across the span in which it actually existed, and index membership is resolved as of the simulated date rather than as it stands today. What stays yours is the choice of universe and period — a hand-typed list of names you know today reintroduces the bias whatever the underlying data contains.

How does Fincanva reduce selection bias?

Selection bias is drawing conclusions from a sample of instruments or periods that is not representative — typically a small, favorable subset picked with hindsight. Fincanva lets you test over a broad instrument universe and a long history rather than a hand-chosen slice that happens to look good. Testing the same rules across a wide universe and a long span makes a result harder to attribute to a lucky handful of names or a single kind stretch of the market.

How does Fincanva reduce overfitting?

Overfitting — also called curve-fitting — is tuning a strategy so tightly to past data that it captures noise rather than a durable signal, so it looks excellent on history and fails afterwards. In Fincanva, you define the rules; Fincanva does not auto-tune or optimize your parameters to fit the past for you, so a backtest shows only what your exact stated rules would have done — which keeps you in control of how closely your rules are shaped to history.

Can Fincanva prevent data-snooping bias?

Data-snooping bias is trying many variations of a strategy until one looks good by chance, then treating that lucky result as if it were real signal. Fincanva does not prevent this for you: the app lets you test freely, so being aware of how many variants you tried — and treating a result that only appeared after many tries with caution — is the discipline this leaves to you. If you run dozens of variants and keep only the best-looking one, that result may reflect chance rather than a repeatable edge — the more combinations you try, the more likely one looks good for no durable reason.

How does Fincanva reduce cherry-picking bias?

Cherry-picking bias is reporting only the favorable results or the periods that worked, while quietly omitting the rest. A Fincanva backtest reports the full-period result together with its drawdowns, not just the stretches that went well — harder to cherry-pick when the same run surfaces the whole history, including the worst peak-to-trough falls; see drawdown for what those falls measure.

How does Fincanva reduce confirmation bias?

Confirmation bias is seeing what you expected to see and discounting evidence that contradicts it. A Fincanva backtest applies the same rules mechanically across the whole history regardless of what you hoped would happen, surfacing the full result — including losses and the maximum drawdown, with headline figures like CAGR — rather than only the parts that confirm your idea, giving the rules a chance to disagree with you.

How does a Fincanva backtest account for trading costs and taxes?

Cost-ignoring bias is pretending trading is free — leaving out the costs and taxes that a real account would pay, which inflates results. A Fincanva backtest can include these through its simulation assumptions: under Include you can switch on Costs & interests ("Trading costs, financing & interest"), Taxes ("Tax on dividends & realized gains"), and Reinvest profits, so a result is not cost-blind. Turning these on brings a backtest closer to what a real account would have kept after costs and tax.

How does data quality affect a Fincanva backtest?

Data-quality bias is when conclusions are distorted by poor, stale, or incomplete data rather than by the strategy itself. Fincanva runs backtests on broad, daily-updated market data covering many asset classes and regions, and a backtest runs to the latest available market close. Wider, current data across more of the market gives a strategy less room to look good only because the data behind it was thin or out of date.

Backtests show what would have happened — not what will, and reducing a bias is not the same as removing it. Fincanva provides no financial advice — see Is this financial advice?.