What Is Algorithmic Trading? A Plain-English Guide

By the numbers

Most trading on modern markets now involves computers following instructions. If you run a business or have capital to place, you have probably wondered what that actually means for an individual. What is algorithmic trading, how does it work in practice, and what can go wrong? This guide answers those questions in plain English, with the UK and US rules side by side.

What is algorithmic trading?

Algorithmic trading means using a computer program to open, manage and close trades according to rules written in advance. The rules can be simple. Buy when the price breaks above a set level. Close the trade if it moves against you by a fixed amount. They can also combine several inputs, such as price, time of day and how volatile the market is.

The software executes the rules. A person or a team designs the rules, tests them and decides how much of the account the program may risk. The computer then applies those rules the same way every time, at any hour the market is open.

You will see several names for the same idea: algorithmic trading, algo trading, automated trading, trading bots and, on the MetaTrader platform, expert advisors. They differ in sophistication more than in principle.

How does an algorithm actually place a trade?

Whatever the strategy, most trading algorithms follow the same loop:

• Watch. The program reads live prices from the broker.
• Check. It compares those prices against its rules and waits for its setup.
• Size. It works out how large the position should be from the account balance and the risk limit it has been given.
• Execute. It sends the order to the broker, usually with a stop loss attached so the maximum loss on that trade is defined from the start.
• Manage. It moves or closes the position according to its exit rules, then goes back to watching.

For an individual, the program usually runs on a trading platform connected to a broker account held in their own name. Many people run it on a virtual private server (VPS), a rented computer in a data centre, so it keeps working when their own laptop is switched off.

Who uses algorithmic trading?

Institutions came first. Banks, hedge funds and market makers use algorithms to split large orders into small pieces, to quote prices continuously and to react in fractions of a second. The markets involved are vast. The Bank for International Settlements found that global foreign exchange trading averaged $9.6 trillion a day in April 2025, up 28% on 2022, with the UK the largest trading centre at 38% of turnover.

Regulators expect firms that trade this way to keep tight controls. Under the European MiFID II rules, investment firms running algorithms need kill switches that can cancel all outstanding orders, limits on order size and position size, and price collars that block orders priced far from the market.

Individuals came later. Retail platforms now let a private trader run rules-based software on their own account, most often in forex, gold and stock indices. In the UK that usually means contracts for difference (CFDs) or spread bets. In the US, retail forex is traded through brokers registered with the National Futures Association (NFA).

What can algorithmic trading do well?

• Consistency. The program applies the same rules every time. It does not hesitate after a loss or chase the market after a win.
• Hours. The forex market trades around the clock on weekdays. Software can monitor it while you run your business or sleep.
• Risk discipline. A well-built program attaches a stop loss the moment a trade opens and sizes each position to a set limit.
• Testability. Clear rules can be tested on historical data before any money is used. A test of the past shows how the rules behaved before. It cannot show how they will behave next.

What are the real risks?

Automation changes how trades are placed. The risks of the market underneath stay the same.

Leverage

Retail forex and CFDs are traded on margin, so a small deposit controls a much larger position. The UK Financial Conduct Authority caps retail leverage at 30:1 on major currency pairs, 20:1 on gold, minor pairs and major indices, and lower on other assets. In the US, NFA rules cap retail forex at 50:1 on major pairs and 20:1 on the rest. At 30:1, £1,000 of margin controls a £30,000 position, so a 1% move against you costs £300. Leverage magnifies losses exactly as it magnifies gains.

Most retail accounts lose money

UK and EU brokers must publish the percentage of their retail CFD accounts that lose money. A 2026 compilation of those published warnings across 49 brokers put the average at 71%, ranging from 51% to 81%. Software does not change that arithmetic on its own.

Changing markets

Rules that suited one period can struggle in another. A strategy built for calm, trending markets may lose repeatedly when conditions turn choppy. Every strategy has drawdowns, stretches where the account falls from its high point, and you need to know in advance how deep you are prepared to let one run.

Technical failure

Code does exactly what it is told, including the wrong thing. In August 2012 a defective automated order router at Knight Capital sent more than 4 million orders in about 45 minutes, leaving the firm with losses of more than $460 million. Knight later paid $12 million to settle SEC charges. As the SEC's Daniel Hawke put it, “Knight Capital's failure to ask these questions had catastrophic consequences.” For an individual the failures are smaller but real: a dropped connection, a VPS outage, a platform update or a broker changing its conditions.

What protections do UK and US traders have?

The rules differ on each side of the Atlantic. This table summarises the main retail protections.

The UK rules have applied to CFDs since 1 August 2019. The US leverage caps took effect in October 2010.

A practical checklist before using any trading software

• Is the broker authorised in your country? Check the FCA register in the UK or NFA BASIC in the US.
• Is the trading account held in your own name, and can you withdraw your money whenever you choose?
• Are the rules explained in plain language? You should understand when it trades and why.
• Does every trade carry a stop loss from the moment it opens?
• What is the most you could lose on a single trade, and in a bad month?
• Can you switch the software off and close positions yourself at any time?
• What happens if the server, platform or internet connection fails?
• Does anyone promise a fixed return? Treat that as a red flag.
• Have you spoken to a qualified adviser about how this fits your wider finances and tax position?

Who does algorithmic trading tend to suit?

It tends to suit people who want a rules-based approach, are comfortable with leveraged products and their risks, and can set aside capital they can afford to lose without affecting their business or household. It also suits people who prefer to oversee a process than to watch charts all day. It tends not to suit anyone who needs the money for a fixed purpose, expects steady monthly income, or would struggle to leave a losing period alone. If you are weighing it against other uses of capital, our comparison of automated trading and buy-to-let property sets out the trade-offs, and our weekly Market Watch reports what moved markets and what is scheduled next.

Frequently asked questions

Is algorithmic trading legal in the UK and US?

Yes. Individuals can run trading software on accounts with regulated brokers. The limits on leverage and the protections described above apply whether you trade by hand or through software.

Do I need to know how to code?

No. Many people use ready-made software. You do need to understand what its rules do, how much it risks per trade and how to switch it off.

Can algorithmic trading lose money?

Yes. Automated strategies go through losing trades and drawdowns, and leverage magnifies losses. Only use capital you can afford to lose.

Is algorithmic trading the same as high-frequency trading?

No. High-frequency trading is one specialised kind, run by firms competing on speed. Typical retail algorithms hold trades for minutes, hours or days and follow much simpler rules.

Sources

• BIS: Global FX trading hits $9.6 trillion per day in April 2025
• FCA: FCA confirms permanent restrictions on the sale of CFDs and CFD-like options to retail consumers
• BrokerChooser: Leverage and margin limits in the UK
• Journal Plus: NFA forex rules in the United States
• BrokerRank: Retail loss rates across CFD and forex brokers (2026)
• SEC: SEC charges Knight Capital with violations of market access rule
• Presencis: MiFID II algorithmic trading controls (RTS 6)

This article is general information, not financial advice. Trading leveraged products such as forex and CFDs carries a high risk of losing money. Past performance is not a reliable indicator of future results. Speak to a qualified adviser about your own situation.