Education · Article 03

Five questions everyone should ask before an algorithmic trading strategy

Whether an algorithmic trading system is reputable can rarely be recognised from outside by its return. The answers to five questions, however, almost always reveal it.

Question 1: How does the provider make their money?

The most important question first. If the provider earns from trading volume, entry fees or the sale of the system itself, they earn regardless of whether you make a profit. If, on the other hand, they earn exclusively on a performance basis from new highs in your account, they sit on your side of the table. A reputable provider answers this question in the first conversation, without evading it.

Question 2: Where can I see real live data?

Screenshots and backtests are not proof, they are marketing material. Only a track record that is automatically captured from the trading account by an external service, and that you can verify yourself, including the bad months, is meaningful. If the periods of loss are missing from the history, the most important part of the truth is missing.

Question 3: How are losses limited?

Ask specifically: what happens to a position that moves against the system? A good answer names a fixed stop-loss per position. A poor answer sounds reassuring and stays vague, such as “the system balances it out again”. Systems that sit out losses or increase their stakes after losses carry structural failure within them.

Question 4: Who holds my capital?

Your money should sit in a trading account held in your own name, with a regulated broker, with deposits and withdrawals that only you yourself can trigger. As soon as a provider asks you to transfer money directly to them, the conversation is over. Not maybe over. Over.

Question 5: What happens in the worst case?

A reputable provider can answer this question precisely: with a realistic drawdown expectation, a clear statement on the possible total loss, and ideally with a metric such as the risk of ruin, together with the loss threshold it refers to. Anyone who instead explains that losses are practically ruled out has answered the question. Just not in the way they think.

WARNSIGNALE UND GUTE ZEICHENJe Frage ein Blick auf beide SeitenWARNSIGNALEGUTE ZEICHENׄGewinne sind sicher“Risiken werden offen benannt×Screenshots statt Live-Datenextern erfasster Track-Record×kein Stop-Loss erkennbarfeste Verlustgrenze je Position×Einzahlung an den AnbieterKonto läuft auf Ihren Namen×Drawdowns fehlen in der HistorieVerlustphasen offen einsehbarWARNSIGNALE UND GUTE ZEICHENJe Frage ein Blick auf beide SeitenׄGewinne sind sicher“Risiken werden offen benannt×Screenshots statt Live-Datenextern erfasster Track-Record×kein Stop-Loss erkennbarfeste Verlustgrenze je Position×Einzahlung an den AnbieterKonto läuft auf Ihren Namen×Drawdowns fehlen in der HistorieVerlustphasen offen einsehbar
Five questions, two kinds of answers. No single warning sign proves anything, but several together form a pattern that should not be explained away.

The pattern behind the five questions

All five questions test the same thing: whether the provider's interests align with yours, and whether risks are named rather than hidden. A provider who answers all five openly can still have bad months, any system can. But they have no structural interest in you not noticing them. You can find the detailed version of these assessment criteria in the educational document.

This article serves exclusively educational and informational purposes. It does not constitute investment advice or a recommendation to buy or sell. Past results are not a reliable indicator of future developments. Trading in financial instruments carries substantial risks, up to and including total loss.

Further reading: Perry J. Kaufman, “Trading Systems and Methods” (Wiley) · Ralph Vince, “The Mathematics of Money Management” (Wiley).

Free information call, bookable after the questionnaire:

To the questionnaire All articles