This is probably the most common problem among traders, especially newbies. You have developed a strategy and tested it on historical data, and it worked amazingly. But suddenly you start trading on a real account, and everything that could fail does fail: trades are not executed perfectly, profits are reduced, and sometimes you even end up in the red. Why? Let’s find out.
1. Backtesting provides ideal conditions
With backtesting on historical data, trading takes place in a vacuum, where:
- Orders are executed immediately at the desired price.
- No spreads, commissions and/or slippage (or requotes if it is an instant execution broker).
- No human factors (fear, panic, greed).
In the real market, things are not so smooth. The price can change while you are executing your order, especially on news and during high volatility. And if you trade huge amounts, the market may simply refuse to accept you at the desired price.
2. Overtraining
One of the most common mistakes is to create a flawless strategy on historical data, but not implement it in the future. It is like preparing for an exam, but on the exam you are asked completely different questions. A good history does not guarantee that you will succeed in real trading.
How to avoid:
- Test on different timeframes and symbols.
- Divide the data into practice and test data (out-of-sample testing).
- Use walk-forward analysis.
3. Changing market conditions
The market is not geometry, and it is not static. The financial market is a dynamic environment. If a strategy works in a calm market, it may act completely differently in a highly volatile or crisis market. Trends change, liquidity changes, the algorithms of large players change – all this affects your trading.
4. Emotions and psychology of a trader
In backtesting, you simply look at the chart and say to yourself: “Yes, here is the entry, here is the exit.” In real trading, when the price goes against you, you are afraid of losing, hesitate or break the rules. At some point, even a winning strategy will become a losing one because of your emotions.
5. Technical problems
During backtesting, there are no:
- Connection breaks and order execution delays.
- Terminal or platform failures.
- Sudden breaks or sharp movements due to news.
But in real trading, all this happens. Especially if the broker does not provide high-quality execution.
How to make a backtest more realistic?
- Think about spreads and commissions – even a small commission can have a big impact. Include slippage – based on the volatility of the asset, include slippage of 0.1%-0.5%.
- Test the strategy in the real market (forward test) – first on a demo account, then on a real account with a minimum volume.
- Monitor statistics – monitor the maximum drawdown, profitability stability and risk/reward ratio.
- Do not over-optimize – the strategy should work not only on history, but also in real conditions.
Final words
Backtesting is informative, but does not simply accept its conclusions on faith. The real test of a strategy is real trading in real market conditions. Use backrest as a means of preliminary verification, but always try the strategy first on a small base of real capital before increasing the trading size. This is the only way to know if it works.