Using the Strategy Tester
- By FXT
- JAN 11, 2023
Checking the Test Parameters
The test parameters need to be checked when running a robot test to ensure the correct information and data is being used. It is important to check the correct Expert Advisor, correct Symbol and correct Timeframe are being used. Checking the date tested and also the Model (what data is being used to test on) is important to ensure the output is reliable. If using the Model "Open Prices Only" or "Control Points", this can sometimes lead to better than expected (and better than actual) results. Remember, we want a realistic picture of the outcome so that we can feel confident in putting the robot to work for us.
Running The Test
A Quick Snapshot
Once you are happy with the inputs and settings, you can run the test. If you just want to see if the parameters are functioning and trades can be placed, a quick snapshot is a good way to start the testing process. It simply helps to confirm the bot is working and is able to function. Ideally, you want to see data in the Graph and the Report.
A More Visual Approach
You have a couple of options when running the test. One way is to use the Visual Mode, which will display the historic data along with orders, stops and take profits so you can visually assess the way the robot works and if you need to make changes. If the visual mode is running too slow, you can speed up the process by dragging the bar next to Visual Mode to the right during the test. The bar looks a bit like a volume slider and controls the speed of the tester. You can slow it down to see the action play out a little slower also.
The strategy tester has a range of features which you can utilize to help understand your trading strategy better, and to test or optimize the inputs of the robot. Since you have already screened for a trading strategy and seen the overall outcome, you may wish to run an optimization.
What Can I Optimize?
Optimization means the strategy will test a range of parameters that you set and provide the end result on each step. For example, if you want to check what the best Take Profit figure to use is, you can run an optimization on this and see what the results show. Let's say you want to test the Take Profit from between 300 points to 2000 points, checking every 100 points in between. The Strategy Tester will pull data and show the results of each iteration for you to review. Additionally, you can test multiple parameters at a time to find optimum combinations of Stop Loss and Take Profit figures and more.
Considerations When Optimizing
Overoptimizing a strategy could mean that you are testing and refining the sample to heavily, and when the market changes, your strategy may no longer fit the conditions of how the market is flowing. Perfect tests do not always lead to perfect trading in the real world, so it is important to consider if the strategy will continue to perform outside of the testing environment. As an example, perfecting the stops and targets for a tight and low volatility period may not harvest the same results when the market becomes more volatile. Consider alternate market conditions and try test with those considerations in mind.
What is the Goal?
This will vary depending on the trader, but the goal in general is to earn a profit from the strategy. Factors that come into play when deciding the profit and worthiness of the trading strategy include risk tolerance, drawdown, timeframes of being in loss vs timeframe being in profit, consistency of the equity curve, win loss ratio and percentages, number of trades and expectancy per trade. All this may sound a bit complicated, so let's simplify.
The goal for a trader might be to earn as much as possible. Their drawdown and risk tolerance may be very high which might seem very risky to another trader. A more risk-averse trader might be looking for a slow and steady return with a very low drawdown and risk tolerance, even if the returns aren't as high. Unless a trader is looking to hedge an alternate strategy in a portfolio, most traders will want to see profitability as a must have. The other data such as win loss percentage and ratios is dependent on what people want out of their trading portfolio.