Definition
Backtesting
Backtesting runs a strategy over historical data to estimate how it would have behaved. Definition, what a credible test must model, and the limits of the result.
Backtesting is the practice of running a trading strategy over historical market data to estimate how it would have performed. It produces an estimate of behaviour under conditions that already occurred, not a prediction. The value of a backtest depends almost entirely on what it models: a test that omits fees, slippage, and funding is not a conservative estimate of live results — it is a different strategy.
- Also known as
- historical simulation · strategy backtest
How it is calculated
A backtest is credible only when every one of these is modelled:
· Fees, both sides, at the tier actually paid
· Slippage, scaled to order size and market depth
· Funding, for every interval a perpetual position was open
· Fill realism — no filling at the close, the mid, or the signal price
· Out-of-sample data the parameters were never fitted on
What a credible test has to include
Trading fees at the tier actually paid, slippage scaled to order size, funding for every interval a perpetual position was held, and fills that respect available liquidity rather than executing at the close or the mid. Each omission moves results in the same direction — flattering — which is why an unfrictioned backtest is not merely imprecise but systematically wrong.
In-sample results are not evidence
Parameters chosen by looking at a data set describe that data set. A strategy has to be tested on periods it was never fitted on before its numbers mean anything, and even then the sample contains only market regimes that happened to occur, which is a narrower set than the future contains.
How it gets misread
A backtest is routinely read as a forecast. It is a description of one path through one sample: same rules, different data, different result. The maximum drawdown it reports is the worst case in that sample, which is a lower bound on future risk rather than an upper one.
See also
Sources
- Pseudo-Mathematics and Financial Charlatanism: The Effects of Backtest Overfitting on Out-of-Sample Performance — American Mathematical Society (Notices of the AMS, Vol. 61 No. 5, 2014)
- Interpretive Notice 9025 — Compliance Rule 2-29: Use of Promotional Material Containing Hypothetical Performance Results — National Futures Association
- The probability of backtest overfitting — Journal of Computational Finance (Risk Journals)
Definitions are educational. Nothing here is investment advice, and no metric described on this page predicts future results.
Definitions are the easy part
Knowing what drawdown means is not the same as having a system that halts on it. QANTERION applies these limits while a strategy runs.