Algorithmic trading means handing your trading rules to software and letting it place the orders. You decide what counts as a signal — a moving average cross, a breakout, an oversold reading — and the machine watches the market and acts the moment those conditions are met. No hesitation, no "just one more pip", no missed entry because you stepped away from the screen. In retail forex it almost always means an Expert Advisor running on MetaTrader.
What it is
Algorithmic trading — "algo" trading for short — is any approach where a defined set of rules is executed automatically, without a person clicking buy or sell each time. The rules can be simple or involved, but they share one trait: they are explicit. Every entry, exit, position size and filter is written down in code, so the same input always produces the same decision.
- Entry rules — the conditions that open a trade.
- Exit rules — stop loss, take profit, a trailing stop, or a time-based close.
- Position sizing — fixed lots, a percentage of balance, or something adaptive.
- Filters — session times, spread limits, news avoidance.
In big institutions the term covers everything from order execution to high-frequency market making. In retail forex it is narrower and more practical.
How it works
On MT4 and MT5 the software is an Expert Advisor: a compiled program attached to a chart. It reads incoming price data, checks it against your rules on every tick or every bar, and sends orders to the broker when a condition is met. The terminal stays open on a computer or, more often, a VPS, so the EA can run around the clock without your own machine being on. Once it is attached and enabled it needs no further input — it just keeps applying the rules.
An example
Suppose your rule is: buy EUR/USD when the 50-period moving average crosses above the 200-period, risk 1% of the account, set a stop at the recent swing low and a take profit at twice that distance. By hand you would have to watch the chart, spot the cross, work out the lot size and place the orders quickly. An EA does all of it in the same instant the cross confirms — at 3am if that is when it happens, across ten pairs at once if you tell it to.
Pros and cons
The appeal is genuine:
- No emotion — the EA does not feel fear after a loss or greed after a win.
- Speed — it reacts and sizes orders faster than any person.
- Runs 24/5 — it covers every session while you sleep or work.
- Testable — because the rules are explicit, you can backtest them before risking money.
The drawbacks are just as real:
- Curve-fitting — a strategy tuned to look perfect on past data often falls apart live.
- Technical failure — a dropped connection or a VPS reboot can stop it silently.
- No judgement — it will keep trading a broken idea through a market it was never built for.
Who it suits, and how to start
Algo trading suits anyone with a rule-based idea who would rather not sit at a screen — people who work full-time, who trade across sessions, or who know their own discipline is the weak link. It does not suit anyone hoping for a hands-off money machine; those do not exist. To start, run one EA on a demo account for a few weeks, backtest it before you commit real money, and understand the strategy it runs rather than trusting a vendor's screenshots. When you go live, use the smallest size that matters to you and monitor it from day one.
The Karnek note
An algo only helps while it is running as designed, and the one thing it cannot tell you is when it has quietly stopped. Karnek reads your live terminal and shows whether each EA is still trading, how its drawdown is tracking, and whether the results still match what you backtested. It is read-only — it watches the account and can never place a trade of its own.
Written and reviewed by the Karnek Research team. Last updated August 2026.
Educational content only - not financial advice. Past performance does not predict future results. Trading carries significant risk.