Why is it your strategy works only on the strategy tester?
“My strategy works fine on backtests but it doesn’t work when trading live.”
Unfortunately we get lots of such messages.
The basis of this problem is the difference between the MT4 strategy-tester and live trading.
On the strategy-tester everything is perfect:
- There is no slippage.
- The spread always remains the same.
- The opening prices of pending orders are never moved away.
- When there is a gap, positions are closed where the stop-loss was placed. The market is not able to jump over the stop-loss.
In our tests we measured slippage values during live trading.
In a 5 digit EURUSD, slippage moved between 0 – 8 points, the average was around 3 points, while the spread was 15.
The examined strategies used stops of minimum 200 points. This means that the slippage was less than 2% of the whole range.
We often hear the following comments from our clients:
„Anyone unable to make money at large spreads will neither be able to do so at low spreads.”
„Slippage is just a few points, it has no significant effect on trading.”
As a result, we decided to examine the effect of slippage on forex robots.
Our method was the following: We saved the slippage values measured on live accounts and we “spoiled” some robots on purpose to move their stop-loss a little bit away, in line with the slippage.
E.g.: If the stop was 200 points, and slippage was 2 points, than the EA put the stop at 202 points from order’s opening price. If slippage was -4 points, then the stop was put 196 points from opening price. All the stops of new positions were modified with a different stops in line with the measured values. If the slippage was 0, (i.e. there was no slippage), then the stop remained at the original price where it was palced.
In order to have an accurate simulation we run 10000 tests on each robot.
In these tests, we always used the slippages in a different order. We used the historical data of January 2015 for the tests and we did not change any other setting of the forex robot.
Our purpose was to determine the difference between the results.
Our first EA strategy was the following:
It entering the market using the RSI signal. If the market went in the wrong direction it started opening new positions double the LOT size of the previous one, as in the Martingale strategies. Its result was +4.86$ without slippage.
With slippage the result was between -2373.5$ and +1337$. We divided this range into 100 equal segments, and we examined the number of results that can be found in each segment:
As a result, we can conclude that this strategy is very sensitive to slippage, because our profit moves in a 3000$ range.
In order to safely run a Forex Robot, we need to select a broker with low slippages and spreads.
With our latest product – the only one on the market – we are able to measure the real spread and slippage of your broker and its effect on your strategy.
Since we scan a lot of brokers, we can tell you where you can keep the risks due to such errors at a minimum. You will thus be able to keep your operational risk under control, ensuring your success based on your expertise and strategy.
Whereby you are curious as to howYour strategy is influenced by slippageget in contact with us, and so that we can prepare your very own personalized test.
If you are also interested in knowing which broker has the lowest spreads and slippages sign up for our new services as a tester!
In our next article we will show you the test results of a conservative (ie: NOT a duplicating Martingale) forex robot, with regards to spread and slippage.
We look forward to reading your opinions!