BackTest Studio

BackTest Studio: Risk Disclosure

DRAFT v0.1, 3 October 2026. Not legal advice; to be reviewed by a lawyer before publication.

BackTest Studio is a software tool for research and education. It is not investment advice, a trading signal service, a broker or an advisor. We do not recommend any instrument, strategy, parameter or trade. Nothing in the Software, on our website or in our content is an offer or solicitation to buy or sell anything.

1. Trading is risky

Trading leveraged products (forex, CFDs, futures, crypto and similar) can lose more than you expect and all of your capital. Most retail traders lose money. Do not trade with money you cannot afford to lose. Seek independent professional advice if you are unsure.

2. Backtests are hypothetical

  • Past results do not predict future results. A backtest is a simulation on historical data; it did not involve real money, real fills or real emotions.
  • Backtests can be wrong or misleading because of data errors or gaps, look-ahead or repainting logic in a script, over-fitting (a strategy tuned until it fits the past), too few trades, regime changes, survivorship, and unrealistic cost assumptions (spread, commission, slippage, swap, news-time widening and rejected orders are modelled only approximately, and some not at all).
  • The Software offers checks (for example a look-ahead audit, Monte Carlo, walk-forward testing, Deflated Sharpe, probability of backtest over-fitting, heat maps, cost profiles). They are statistical aids, not proof. A strategy that passes them can still lose money; one that fails them might not. Their results depend on your data and settings.
  • Trial counts matter: the more settings you try, the more likely the "best" result is luck. The experiment log helps you count; it does not remove the problem.

3. Costs and data

Built-in cost profiles (for example for a particular broker's published average spreads) are approximations and may not match your broker, account type, time of day or market conditions. Historical and live prices come from third-party sources that may contain errors, gaps or delays and may stop being available. Verify what matters to you.

4. Artificial intelligence

The optional AI analysis is generated text. It can be wrong, incomplete or misleading, and is not advice. It sends your results and code to Anthropic using your own key (see the Privacy Policy). You are responsible for what you send and for any costs.

5. Your responsibility

You decide whether and how to trade. Test with small size or on a demo account first. You are responsible for checking every result and every strategy, for compliance with the laws and broker rules that apply to you, and for your own tax and legal obligations.

6. No guarantee of performance

Any sample figures, screenshots or examples shown by the Software or in our materials are for illustration only. We make no promise or guarantee of profit or of avoiding loss.

Contact: destek@backteststudio.dev