Three very different things end up under the same label:
The most complete systems combine all three: rules measured on data to choose where to look, agents that read the context and debate, and a human being who decides.
This is the part that people selling «miracle bots» don't tell you:
Let's take STA as a concrete example. Twice a day a model, measured on seven years of data, points to the markets with a clear direction. For each one, four agents prepare the dossier: the Market Analyst looks at the chart, the Derivatives Analyst at funding and open interest, the News Analyst at the news, the Social Analyst at the community's mood.
Then comes the most important part: the debate. The Bull Researcher builds the case for, the Bear Researcher looks for every reason not to enter, the Research Manager weighs the two cases and the risks. Only if the idea holds up does the Trader write the card: entry, stop, target, expected duration and the reasoning in plain words. Ideas that don't hold up don't become cards.
| Question | A serious system's answer |
|---|---|
| Does it show its losses? | Yes, all of them, as prominently as the wins. |
| Are the results net of costs? | Yes, and it says so. |
| Can I verify the history? | Yes, with dated data that cannot be changed. |
| Does it promise returns? | No. Never. |
| Who decides? | You: you can always ignore, stop, exit. |
If you try a system in DEMO, don't just check whether it «makes money». Look at these four numbers:
| Number | What it tells you |
|---|---|
| How many trades it proposes | If there are few, you need more time to judge it. |
| How many end at the stop | Many is normal: what counts is how much the others return. |
| The average result in R | How much each trade returns relative to the risk taken. Above zero, net of costs, is the first good sign. |
| The longest losing streak | It tells you how much you must be able to endure without quitting at the wrong moment. |
An artificial intelligence system proposes, and the responsibility for the decision stays with you. That is why a serious service always lets you ignore a proposal, change its amount and stop everything. Be wary of anyone who asks you to «let the machine do it» without showing you what it does.
The most sensible way is to watch the system work without risk: a DEMO account, with virtual capital and real prices. After a few weeks you will have seen how it reasons, how often it is wrong and how it handles losses. Only then does it make sense to decide whether to use it with real money, and how much.
Every card as soon as it closes, with the result and what STA learned from it. It's free and you don't need to sign up.
Follow the channelA DEMO account with virtual capital and real prices, the same cards as REAL users.
Try DEMO for free →This article is for information only and is not investment advice. Trading crypto-assets involves a high risk, including the total loss of the capital used. Numerical examples are for illustration.
Risk management in trading: how much to risk per trade, how to calculate the position and how to use a stop, with made up examples and a demo account.
Trading online with artificial intelligence: what AI can do, what it can't, how to try it on a demo account and how to spot those who promise too much.
Is trading online worth it? It depends on the costs: spread, fees, swap and slippage. What really weighs on a small account and how to do the maths first.