1. Trading involves substantial risk
Trading stocks, ETFs, options, futures, currencies and crypto-assets involves substantial risk of loss and is not suitable for everyone. With leverage or margin you can lose more than you deposit. Only trade with money you can afford to lose.
SCUTA Quant is a research tool. It does not give investment advice or tell you to trade any strategy. See section 3 of the Terms of Service.
2. Hypothetical performance
Hypothetical or simulated performance results have certain limitations. Unlike an actual performance record, simulated results do not represent actual trading. Because the trades have not been executed, the results may have under- or over-compensated for the impact of certain market factors, such as lack of liquidity. Simulated trading programs in general are designed with the benefit of hindsight. No representation is made that any account will or is likely to achieve profits or losses similar to those shown. [CONFIRM REQUIRED REGULATORY WORDING]
Past performance, real or simulated, does not indicate future results.
3. Transaction costs are modelled, not measured
By default a backtest charges 1 basis point of slippage on each fill and a commission of $0.005 per share with a $1 minimum. You can change these in Run settings. Real costs can be higher and are not all modelled:
- bid–ask spreads, which widen in fast or thin markets;
- exchange, regulatory and clearing fees, and your broker’s actual commission schedule;
- borrowing costs and availability for short positions, margin interest and financing;
- futures roll costs, currency conversion and taxes.
4. Fills are assumed
By default a market order fills at the open of the next bar after the strategy decides, plus slippage. The optional "same bar’s close" model fills at a price the strategy could not have traded at, and is labelled as optimistic. In every model:
- every order fills in full: no partial fills, queue position or rejected orders;
- your orders do not move the price (no market impact), whatever their size;
- stop and limit orders fill when a bar’s range reaches their price, at a modelled price; a gap through a stop fills at the open, and real fills can be worse;
- only bar prices are known, so the order of moves inside a bar is assumed;
- trading halts, circuit breakers, outages and liquidity crises are not modelled.
5. Data can be incomplete or wrong
- Historical data comes from Massive and its sources. It can contain errors, revisions and gaps (missing days or bars). We flag gaps longer than a threshold in the data a backtest used, but cannot detect every problem.
- Prices are adjusted for splits; dividends are handled as described in each run’s assumptions. Other corporate actions may not be reflected.
- Symbols that were delisted, merged or renamed may be missing, which can make historical results look better than they were (survivorship bias).
- Intraday sessions (regular or extended hours) and time zones follow the run settings; a strategy that depends on the exact session boundary can behave differently live.
6. Overfitting and look-ahead
Testing many ideas or parameters on the same history will find strategies that worked by chance. Code can also use information that was not available at the time (look-ahead bias) in ways the engine cannot always detect. Monte Carlo, regime and walk-forward analyses and the verdict grade reduce these risks but do not remove them. The verdict is a heuristic summary, not a prediction or a rating of investment merit.
7. Prop-firm simulator
The simulator replays a backtest’s daily results against generic challenge templates or rules you enter. It is not affiliated with any prop firm and does not use any firm’s actual rules unless a profile clearly says so, with the date its rules were last checked and a link to the firm’s page.
- Rules are checked at each bar’s close. With daily bars only the closing balance of each day is known, so a real account can breach an intraday limit on a day that closes above it.
- A day is a New York calendar date; firms may use different session boundaries and time zones.
- Firms’ rules change often and include details the simulator does not model, such as news-trading restrictions, position limits, scaling plans, activation and reset fees, payout rules and how they measure equity.
- Passing a simulated challenge does not mean you would pass a real one.
8. AI-generated code
Copilot can write code that does not do what it says, contains bugs, or introduces look-ahead bias. A strategy that looks good because of a coding mistake is a common and costly error. Read and test generated code before trusting its results.
9. Your responsibility
You are responsible for your trading decisions and for checking any result before acting on it. If you are unsure whether trading is suitable for you, get advice from a licensed professional.