Algorithmic trading means letting a computer follow your trading rules exactly. Here is how it works, what it can do well, and where it can go wrong.
Algorithmic trading, or algo trading, means writing your trading rules down so precisely that a computer can follow them without you. The program watches the market, decides when a rule is met, sends the order, manages the trade and closes it. You decide the rules; the software does the clicking.
What an algorithm actually is
An algorithm is just a list of instructions. A very simple trading algorithm could be:
- If the 20-period average crosses above the 50-period average, buy.
- Place a stop loss 2 x ATR below the entry.
- Close the trade when price reaches 2 x the risk, or when the averages cross back.
- Never risk more than 1% of the account on one trade.
Each line is testable and unambiguous. That is the key difference from discretionary trading, where a trader might say "buy when the trend looks strong". A computer cannot act on "looks strong" - it needs a number.
What algorithms do better than people
- Discipline. The program never skips a stop loss because it "has a feeling".
- Speed. Orders go out in milliseconds, which matters in fast markets like gold.
- Stamina. An Expert Advisor can watch the market 24 hours a day, five days a week.
- Testing. Because the rules are exact, you can replay years of history and see how they would have behaved.
What algorithms cannot do
Automation does not create an edge by itself. A losing idea that is automated just loses faster and more consistently. Algorithms also do exactly what they are told, including mistakes: a wrong lot size or a missing stop loss will be repeated on every trade. And markets change - a rule set that worked in a calm year can struggle when volatility doubles.
The typical workflow
- Idea - a market behaviour you think repeats, such as breakouts after a quiet session.
- Rules - turn the idea into exact entry, exit and risk rules.
- Backtest - run the rules over historical data, including spread and costs.
- Out-of-sample check - test on data the rules were not tuned on.
- Demo or small live run - confirm the live behaviour matches the test.
- Monitor - keep watching drawdown, slippage and whether the edge is still there.
Who is it for?
Algo trading suits people who like structure and are willing to measure everything. You do not need to be a programmer to start - many traders begin with ready-made Expert Advisors on MetaTrader 5 and learn to adjust inputs - but you do need the patience to test before you trust.
The bottom line: algorithms are a tool for executing a plan consistently. The quality of the plan, and the risk you attach to it, still decides the result.
Trading forex and gold carries a high level of risk and may not be suitable for every investor. Past performance and backtests do not guarantee future results. This article is education, not financial advice.