Table of Contents
The MetaTrader 4 (MT4) platform remains one of the most widely used platforms for Forex traders worldwide, known for its Expert Advisors (EAs), indicators, and flexibility. The strategy tester is a key feature of MT4, which allows traders to backtest Expert Advisors (EAs) on historical data. This testing process helps to evaluate how an EA would have performed in the past. However, the accuracy of the MT4 strategy tester is often a subject of debate, as the platform has several limitations that affect real-world trading results. In this article, we will explore the functionality, limitations, and practical applications of the MT4 strategy tester.
Please check my video about this topic:
1. Overview of MT4 Strategy Tester
The MT4 strategy tester allows users to simulate trades based on historical data. It is accessed by pressing Ctrl + R, where users can select their preferred settings, including:
- Expert Advisor (EA): The automated strategy that will be tested.
- Testing Period: The range of historical data over which the EA will be tested.
- Asset/Currency Pair: The specific currency pair or asset for testing.
- Model of Testing: Open price only, every tick or control point.
The strategy tester offers insights into how an EA might perform under certain market conditions, but these simulations can differ significantly from real-time trading.
2. Common Testing Models
MT4 offers three main testing models:
a) Open Price Only
This is the fastest testing method, where trades are only opened and closed at the open price of a new bar (candle). It’s beneficial for strategies where trades are opened or closed based on price action at the open of a new candle. While fast, this model lacks precision because it doesn’t account for intra-candle price fluctuations.
b) Every Tick
This is the most accurate model in MT4, where trades are opened, closed, and managed on every tick of the price data. However, it is computationally heavy and takes longer than the other models. Despite its accuracy, it can still miss certain real-world elements, especially regarding spread variations and execution speed.
c) Control Points
This model is somewhere between the previous two and provides moderate accuracy. It uses a bar’s open and close prices to simulate price action between the bars, though it lacks precision compared to every tick.
3. Spread Handling in MT4 Strategy Tester
One of the MT4 strategy tester’s main criticisms is its handling of spreads. In the real world, the spread (the difference between the bid and ask prices) fluctuates constantly, depending on market conditions, news releases, liquidity, and the specific broker. However, the MT4 strategy tester typically uses a fixed spread, which doesn’t accurately represent these real-time variations.
Even if a trader uses a “zero spread” account, brokerage firms can still introduce a small variable spread during highly active trading periods, such as the overlap of the US and European sessions or significant news events. These variable spreads are not well-captured in MT4’s backtesting environment. This discrepancy can lead to backtesting results that look impressive on paper but fail in live market conditions due to these overlooked spread changes.
For example, a strategy that appears to work well in backtesting, which generates trades on a tick-by-tick basis, may encounter significant issues in live trading because of unaccounted-for spread widening.
4. Testing for Manual Trading Strategies
Those testing strategies for manual trading use the Price Only” model. This is often preferable. The method is faster and generalizes whether a strategy is profitable without focusing on extreme precision. The focus is to check if the strategy generates consistent results over time. Traders typically use this method to fine-tune the number of trades, rules, or parameters without worrying too much about minor fluctuations.
However, the “Every Tick” model is more suitable for fine-tuning and testing highly accurate strategies, even though it can take much longer to simulate.
5. Expert Advisors in Live Trading: The Spread Dilemma
One key issue with using EAs in live trading is that they don’t have the flexibility to adapt to market conditions the way a manual trader can. For instance, when a target is missed by just a few pips or the market moves against the position, an EA might not adjust its parameters dynamically unless programmed with complex decision-making rules. This lack of adaptability can lead to suboptimal results in live trading compared to backtesting.
Also, because spreads can widen during volatile times (especially during news events), EAs that generate many trades quickly can suffer from higher costs due to spread widening. The backtesting environment cannot account for these small but crucial real-world conditions, and as a result, many traders find their EA strategies underperform in live accounts.
6. Combining Multiple Expert Advisors
One solution to these issues is using multiple EAs simultaneously. By employing different EAs that cover various market conditions or price action strategies (e.g., buying on support and selling on resistance, or vice versa), traders can minimize drawdowns and improve overall profitability.
For instance, one EA might buy on support while another sells on resistance. Combining these with other strategies can help fill market gaps left by one EA, and overall, the drawdowns will be smaller, and profits cumulative. Combining several EAs works best when the trader backtests each one individually before using them in tandem on live accounts.
7. MT4 vs. MT5 Strategy Tester: Which is More Accurate?
While MetaTrader 4 is a powerful and flexible platform, it falls short regarding accuracy in backtesting compared to MetaTrader 5 (MT5). MT5 provides more features, including a better representation of market depth, more granular spread handling, and the ability to test strategies using accurate tick data. In MT4, tick data is often approximated, leading to inaccuracies in simulated trades.
Many traders prefer MT4 due to its familiarity, ease of use, and extensive community of developers and support. For manual strategies or traders who want quicker, more straightforward backtesting, MT4 is an excellent choice.
Conclusion
The MT4 strategy tester is an indispensable tool for traders, but it comes with notable limitations, especially concerning spread variability and real-world execution. While it’s effective for general testing and strategy refinement, traders must be cautious when transitioning from backtested results to live trading. Testing strategies with multiple EAs, understanding the impact of spread changes, and considering the advantages of MT5 can help bridge the gap between backtesting and real-world performance.
For those who rely on automated trading via EAs, combining several strategies to offset each other’s weaknesses is a sound approach. While MT4 remains a reliable and familiar platform, understanding its limitations is essential for achieving long-term success in live trading.
He is an expert in financial niche, long-term trading, and weekly technical levels.
The primary field of Igor's research is the application of machine learning in algorithmic trading.
Education: Computer Engineering and Ph.D. in machine learning.
Igor regularly publishes trading-related videos on the Fxigor Youtube channel.
To contact Igor write on:
igor@forex.in.rs
