Automated Trading: How It Works, the Real Risks and How to Evaluate a Service
Updated 2026-11-16 · 10 min read
Automated trading means using computer programs to place buy and sell orders according to predefined rules, without a person clicking each trade. It can bring discipline and speed, but it also brings risks of its own: strategies that only work on past data, deep drawdowns and outright scams. This guide explains how trading bots, copy trading and PAMM accounts work, and gives you a practical checklist for judging any automated trading service.
What is automated trading?
Automated trading is the use of software to execute trades according to a set of rules written in advance. The rules can be simple, such as buying when a short-term moving average crosses above a long-term one, or complex, combining many indicators, risk limits and position-sizing formulas.
Once the rules are coded and connected to a brokerage account, the program monitors the market and sends orders on its own. A person still decides what the rules are and how much risk to allow. The software then applies those rules consistently, at any hour the market is open.
You will also see the terms algorithmic trading, algo trading, systematic trading and trading bots. They overlap heavily. Algorithmic trading usually refers to the rule-based logic itself, and automated trading to the fact that execution happens without manual intervention. In everyday use, most people treat them as the same thing.
Automated trading vs manual trading
In manual trading, a person analyses the market, decides and places each order. In automated trading, the analysis and execution follow fixed instructions. The table below summarises the main differences.
| Aspect | Manual trading | Automated trading |
|---|---|---|
| Decision-making | Discretionary, made trade by trade | Rule-based, defined in advance |
| Emotional influence | High: fear and greed can override plans | Low during execution, but humans can still switch the system off at the wrong time |
| Speed | Limited by human reaction time | Orders sent in fractions of a second |
| Hours covered | Only when the trader is watching | Can run whenever the market is open |
| Main weakness | Inconsistency and fatigue | Rules may stop working when market conditions change |
| Testing | Hard to test objectively | Can be backtested on historical data (with caveats) |
How automated trading systems work
Most automated trading systems share the same building blocks, whatever the market or platform.
- Strategy rules. The conditions for entering and exiting a trade: which signals, which instruments, which timeframes.
- Risk management. Stop-loss levels, maximum position size, maximum number of open trades, and sometimes a daily or overall loss limit that stops the system.
- Data feed. Live prices that the program reads continuously.
- Execution. A connection to a broker, usually through an API or a trading platform, so the program can send, modify and close orders.
- Monitoring. Logs, alerts and human oversight to catch errors, disconnections or unusual behaviour.
Backtesting and forward testing
Before a strategy goes live, it is normally backtested: the rules are applied to historical price data to see how they would have performed. A sensible developer then runs a forward test (also called paper trading or demo trading) on live data without real money, and finally trades small real amounts.
Backtests are useful but limited. They assume prices, spreads and fills that may not match reality, and they describe the past, not the future. A good backtest is a reason to look further, not proof that a strategy works.
Types of automated trading: bots, copy trading and PAMM
Retail investors usually meet automated trading in one of three forms. They differ mainly in who controls the strategy and where your money sits.
Trading bots and expert advisors
A trading bot is a program you run yourself, either bought, rented or built. On popular forex platforms these are often called expert advisors (EAs). In crypto, bots typically connect to an exchange through API keys.
You keep control of the account and settings, but you are also responsible for configuring the bot correctly, keeping it running and understanding what it does. Many bots sold online come with impressive-looking results that cannot be verified.
Copy trading and signal services
With copy trading, your account automatically replicates the trades of another trader or strategy provider, usually in proportion to your balance. Signal services are similar, except they send trade ideas that may be executed automatically or manually.
Your money normally stays in your own account with the broker, and you can stop copying at any time. The risk is that you are relying on someone else's judgement, and past rankings on copy-trading platforms can change quickly.
PAMM, MAM and managed accounts
PAMM stands for Percentage Allocation Management Module. Investors allocate money to a manager, whose trades are distributed across all investors in proportion to their share. MAM (Multi-Account Manager) is a related set-up that offers more flexibility in allocation.
These structures are offered by some brokers and can be legitimate, but the arrangement matters. You should understand who holds the funds, who can withdraw them, how fees are calculated and whether you can see every trade.
Comparison of the main structures
| Structure | Who controls trading | Where funds usually sit | Typical costs | Key question to ask |
|---|---|---|---|---|
| Own trading bot | You, through the software | Your own broker or exchange account | Licence or subscription, plus trading costs | Can the results be independently verified? |
| Copy trading | The strategy provider | Your own broker account | Spread mark-ups, performance or subscription fees | How long and how consistent is the provider's record? |
| PAMM / MAM | The account manager | Broker account linked to the manager's structure | Management and performance fees | Can the manager only trade, or also move money? |
| Managed automated service | The service's strategy | Ideally your own account in your name | Often a profit share | What exact permissions does the service have? |
Some services combine elements of these models. For example, Cortex Automation runs an automated strategy on accounts that stay in the client's own name at their broker, with fees based on a share of profits. Whatever the provider, the principle to look for is the same: trading permissions without withdrawal permissions.
Advantages of automated trading
Automation has genuine benefits when it is built and used carefully.
- Discipline. The system follows its rules every time, which removes impulsive decisions during execution.
- Speed and precision. Orders can be placed immediately when conditions are met, reducing delays.
- Consistent risk control. Stop-losses and position sizes are applied automatically rather than forgotten.
- Coverage. A program can monitor markets during hours when you are asleep or working.
- Testability. Rules that are written down can be tested, measured and reviewed, unlike gut feeling.
- Diversification. Several strategies or instruments can be run at once, which would be hard to manage manually.
None of these advantages makes a strategy profitable on its own. Automation executes a strategy faithfully, including a bad one.
The real risks of automated trading
This is the section that matters most. Automated trading can lose money, sometimes quickly, and the risks are not always obvious from marketing material.
Overfitting: when a strategy only works on the past
Overfitting (also called curve fitting) happens when a strategy is tuned so closely to historical data that it captures random noise rather than a real, repeatable pattern. The backtest looks excellent, but live results disappoint.
Warning signs include a very smooth backtest equity curve, many finely tuned parameters, and results that collapse when the settings are changed slightly or the test period is moved. A robust strategy should behave reasonably across different periods and market conditions, not only on the data it was designed on.
Drawdowns and the maths of recovery
A drawdown is the fall in account value from a peak to a subsequent low. Every strategy has them, and the maximum drawdown is one of the most important figures to understand before you start.
Losses and gains are not symmetrical. Consider a hypothetical account of £10,000:
- A 10% loss takes it to £9,000. It then needs about an 11% gain to return to £10,000.
- A 25% loss takes it to £7,500. It needs about a 33% gain to recover.
- A 50% loss takes it to £5,000. It needs a 100% gain to get back to where it started.
This is why leverage and aggressive position sizing are dangerous. A strategy that occasionally loses half its value may need a very long time, or very high risk, to recover.
Leverage and market risk
Many automated strategies trade leveraged products such as forex, CFDs or futures. Leverage magnifies both gains and losses. Sudden events, such as surprise central bank decisions or market gaps over a weekend, can move prices past stop-loss levels, leading to larger losses than planned. With some products and account types, losses can exceed the initial deposit.
Technical and operational risk
Software can fail. Internet connections drop, servers restart, platforms update and APIs change. A bug in position sizing or an order that is sent twice can cause serious losses in minutes. Good systems include monitoring, error handling and limits that stop trading when something unexpected happens.
Changing market conditions
A trend-following strategy may struggle in a sideways market; a strategy that profits from calm conditions may suffer when volatility spikes. Markets change, and a strategy that worked for several years can stop working without warning.
Scams and misleading offers
Automated trading attracts fraud because it promises income without effort. Common red flags include:
- Guaranteed or fixed returns, or claims that the bot cannot lose.
- Pressure to deposit quickly, or bonuses for recruiting friends.
- Requests to send money to the provider's own account or wallet rather than your broker.
- Screenshots instead of verifiable account statements.
- Unclear company details, no physical address, or anonymous managers.
- Difficulty or delays when you try to withdraw.
Regulators in many countries publish warning lists of unauthorised firms. Checking these lists before engaging with a provider is a sensible habit.
Checklist: how to evaluate an automated trading service
Before committing money to any automated trading service, bot or manager, work through these questions.
- Is the track record verified? Look for live results that are independently verified through a third-party tracking service connected directly to the trading account, not just a spreadsheet or screenshots.
- Is it a real account or a demo? Demo results ignore slippage and real execution. Live, real-money history is far more meaningful.
- How long is the record? A few good months says little. Look for a history long enough to include difficult market periods.
- What was the maximum drawdown? Ask for the largest peak-to-trough loss and ask yourself honestly whether you could tolerate it.
- Where are the funds held? Ideally, money stays in an account in your own name at a broker, and the provider has trading access only, with no ability to withdraw.
- Is the broker reputable and regulated? Check the broker's regulatory status directly on the regulator's register, not just on the broker's website.
- How are fees structured? Understand subscription, management and performance fees. With profit-share models, check whether a high-water mark applies, so fees are only charged on new profits.
- Is the strategy explained? You do not need the source code, but you should understand the general approach, the markets traded, the use of leverage and the risk limits.
- Can you stop at any time? Check how to disconnect the service and close positions, and whether there are lock-in periods or exit penalties.
- Are the claims realistic? Be wary of any promise of steady high returns with little risk. Legitimate providers talk about risk as much as return.
Who is automated trading suitable for?
Automated trading may suit people who understand that losses are possible, can leave the money invested through difficult periods, and only use capital they can afford to lose. It is generally not suitable for emergency savings or money needed in the short term.
It is also wise to treat automated trading as one part of a broader financial plan, rather than the whole of it. Starting with a small allocation and reviewing results over a meaningful period is a cautious approach.
Key takeaways
- Automated trading uses predefined rules and software to place trades without manual clicks.
- Bots, copy trading and PAMM accounts differ mainly in who controls trading and where your money sits.
- Automation brings discipline and speed, but it does not make a weak strategy profitable.
- Overfitting, drawdowns, leverage, technical failures and scams are the main risks.
- Insist on a verified live track record and, ideally, funds held in your own account with trading-only permissions.
This article is general information for educational purposes and does not constitute financial advice. Trading involves risk, including the possible loss of capital, and past performance does not guarantee future results. Consider your own circumstances and, if needed, seek independent professional advice.
Frequently asked questions
Is automated trading profitable?
It can be, but there is no guarantee. Profitability depends on the strategy, risk management, costs and market conditions. Many automated strategies lose money, and past results do not guarantee future performance.
Is automated trading legal?
In most countries, using automated trading software on your own account is legal. However, firms that manage money or offer investment services to others are usually required to be authorised by the relevant financial regulator, so check a provider's status directly with the regulator.
What is the difference between copy trading and PAMM?
In copy trading, trades from a provider are replicated in your own separate account. In a PAMM structure, a manager trades a pooled arrangement and results are allocated to investors in proportion to their share. Fees, control and withdrawal arrangements differ between the two.
How can I tell if a trading bot is a scam?
Common warning signs include guaranteed returns, pressure to deposit quickly, requests to send money to the provider rather than a broker, unverifiable screenshots and problems withdrawing funds. Check regulator warning lists and ask for an independently verified live track record.
What is a drawdown in automated trading?
A drawdown is the decline in account value from a peak to a later low. For example, an account falling from £10,000 to £8,000 has a 20% drawdown. Larger drawdowns need proportionally larger gains to recover.
How much money do I need to start automated trading?
Minimums vary by broker and service. Whatever the amount, only use money you can afford to lose, and consider starting small while you assess how the strategy behaves with real funds.
Want to see how an automated system works, with the money in your own account? See results and simulator →