What is Backtesting?

Quick answer: Backtesting is the process of simulating a fully specified trading strategy on historical market data to estimate how it would have behaved, so its edge, risk and cost sensitivity can be studied before any real capital is committed.

In simple words

A backtest replays the past bar by bar and lets your rules trade through it, recording every position and the running account balance. It is a rehearsal on known history, like practising a match after already knowing how the ball moved. That makes it useful evidence about how an idea behaves, but it is never a promise about a future that has not happened yet.

Purpose

Backtesting exists to convert a vague trading idea into measurable evidence and to eliminate strategies that are broken, fragile or unprofitable before they can lose real money.

Visual explanation

Professional explanation

What a backtest actually is

A backtest is a deterministic simulation. A historical price series is fed into a set of rules that decide when to enter, how much to size and when to exit, and the engine records the resulting trades, cash flows and equity curve. The output is a synthetic track record for a period that has already occurred. Because the rules are explicit and the data is fixed, a correct backtest is reproducible: the same code over the same data must always return the same trades. That reproducibility is precisely what separates a systematic strategy from a discretionary hunch.

What a backtest can establish

A carefully built backtest can reveal whether an idea had any historical edge at all, how its returns are distributed, how deep and how prolonged its drawdowns became, how frequently it traded, and how sensitive the result is to costs and parameter choices. Its most reliable use is falsification: if an idea loses money even on the friendliest slice of history with zero frictions, it is almost certainly not worth pursuing. It also forces complete specification, because a machine cannot execute an instruction as loose as buy when the chart looks strong.

What a backtest cannot establish

A backtest cannot tell you what happens next. Past behaviour is one sample from a process whose regime, liquidity and microstructure shift over time. It cannot reproduce the emotional reality of holding a losing position through a deep drawdown, and it cannot perfectly model your true fills, latency, or the market impact of your own orders. Most dangerously, a backtest can be engineered to look excellent purely by fitting the noise of the chosen history, which is why most of this pillar is concerned with the ways a backtest can quietly mislead.

Costs and frictions decide realism

A profit line that ignores costs is fiction. A realistic Indian backtest must subtract brokerage, exchange transaction charges, GST, SEBI turnover fees, stamp duty and Securities Transaction Tax (STT), plus a slippage estimate for the gap between intended and achieved price. For an active intraday strategy these frictions can exceed the entire gross edge, converting an apparently rising curve into a steady loss. Frictions are not a footnote appended at the end; for high-turnover strategies they are the single variable that decides viability.

In-sample versus out-of-sample

The most important discipline is to separate the data used to build and tune a strategy (in-sample) from data it has never seen (out-of-sample). Performance measured on data you optimised over is almost always flattering, because the rules were shaped, knowingly or not, to that particular history. Honest evaluation reserves untouched data, or uses walk-forward analysis, so the reported figure approximates what a genuinely unseen period might deliver. A single backtest number with no out-of-sample check carries very little information.

Vectorised versus event-driven engines

Two broad engine styles exist. A vectorised backtest computes signals across the whole array at once; it is fast and ideal for research and screening, but it is easy to accidentally reference future data. An event-driven backtest processes one bar or tick at a time through the same code path a live system would use, which is slower but far harder to cheat and much closer to production reality. Disciplined teams research in vectorised form and confirm the survivors in an event-driven engine before trusting the number.

Practical example

Advantages

• Falsifies bad ideas cheaply, before any capital is exposed
• Forces a vague idea into a complete, explicit rule set
• Quantifies drawdown, trade frequency and cost sensitivity
• Is fully reproducible, so results can be audited and compared
• Provides a documented baseline the live system can be checked against

Limitations

• Describes only the tested past; it cannot predict the future
• Can be fitted to noise and made to look excellent with no real edge
• Cannot model true fills, slippage, latency or market impact perfectly
• Ignores the psychological reality of holding losers through drawdowns
• Is only as trustworthy as the point-in-time correctness of its data

Why it matters in practice

• Separates a systematic, testable process from discretionary guessing
• A backtest that ignores costs and out-of-sample data is worse than none, because it manufactures false confidence

Common mistakes

• Treating a good backtest as a forecast of future returns rather than a study of past behaviour
• Reporting gross returns while quietly omitting brokerage, STT and slippage
• Optimising over all available data and leaving nothing genuinely out-of-sample
• Acting on the same bar's close that generated the signal, which is look-ahead bias
• Testing on a survivorship-biased universe of only currently listed, surviving stocks
• Trusting a smooth equity curve without inspecting its drawdown depth and duration

Professional usage

Professional quants treat a backtest as a hypothesis test, not a sales pitch. They budget realistic frictions from the first run, insist on out-of-sample or walk-forward validation, push survivors through an event-driven engine that mirrors production, and weight maximum drawdown and cost sensitivity at least as heavily as the headline return. A backtest is where an idea is disproved cheaply; a strategy is only trusted after it also survives forward testing on data no one has seen.

Frequently asked questions

Voice search & related questions

Natural-language questions people ask about What is Backtesting?.

Sources & references

Last reviewed 11 July 2026. Educational content only — not investment advice. Markets and rules change; verify current conventions with SEBI, NSE/BSE and your broker.