I was curious how many people have tried the [B]’So Easy It’s Ridiculous’ System[/B] that is in the School of Pipsology?
To be clear, I set the charts up just as it said within the examples. I also followed the rules exactly as they were specified. If the rule states that you can only go long if the RSI is over 50, I wouldn’t go long if even if the RSI was 49.9. This cut out some good potential trades, but it also, saved me from potential losing trades.
I back tested this system to July 1st and the results were outstanding. Out of a total of 9 trades, only one was unsuccessful. The result was 1,937 pips gained from July 1st to December 16 in 8 successful trades.
This made me wonder if back testing is actually 100% accurate. What is the catch here? If it was this easy, everyone would be doing it.
It would seem to me that the MA lines would not cross right away in real life. For example, the 5 EMA crosses under the 10 EMA on December 4th, so in my back testing a went SHORT and then closed the trade on December 7th when the 5 EMA crossed the 10 EMA again.
Would that EMA crossing on the 4th show up on the 4th if this were live data? It would seem to me that the crossing wouldn’t become apparent until after the trend already began to occur. So, wouldn’t I not notice the cross over until December 5th or December 6th after the 5 EMA had time to adjust or would it be apparent on the close of the December 4th?
This would obviously mean, the profits that I received during the back testing would not be the same as if I had done it all with live data.
I may not be asking the correct question, but if you understand what I am trying to get at, please shed some light on the subject for me.
I believe we’re talking about a trend-following system using daily data. Most trend-following systems have a win/loss ratio below 50%.
You are not using enough data in your backtest.
The catch is that it’s actually not that easy in practice. Drawdowns in TF systems be severe and can last months, even years. Many cannot handle them and give up.
Please explain. What is the point in back testing 6 months if it’s not enough data? The school of pipsology says to backtest systems before testing them in demo accounts. This is exactly what I did. Should I be back testing for a couple of years?
I want to know exactly why the backtesting is no good. I don’t mind vague answers, but when it’s the only answer, it makes it difficult to learn from it.
Yes. You need a large enough representative sample. How do you know if you have a large enough sample? There is no hard answer, but one way is to test on 70% of your data. Then test on the full 100%. If there is a significant difference, you might not have enough data.
Using daily data, I used at least 12-15 years for my backtests.
Keep in mind that backtesting is merely a tool to help you develop a trading system, not a guarantee of future results.
I don’t mean to be vague. If you like detailed explanations, feel free to visit my thread (shameless plug)
How long does it typically take to back test 10+ years?! I was thinking it was a pain going back to July :eek: Actually it didn’t take more than 30 minutes for me to go back half a year. I never did check to see how far MT4 would go back.
charrell
What time frame did you back test, daily, 4 or 1 hour charts? If you used 4 or 1 hour charts six months back testing sounds good to me. If daily charts a year or two would be more than enough to give you an idea if the system works or not.
PS it wouldn’t be in the babypips school if it wasn’t a valid trading method, try it on demo account. Forward testing not only shows you if it’s profitable or not but also gives you lots of valuable hands on experience trading it.
I only tested the system on the daily charts. I will back test it back a couple of years tomorrow and see how it does. I do plan on testing it on a demo account for a couple of months to see how well it does for me.
When backtesting a system, you need to have enough data that you can see how well your system would have performed during several market cycles (trending up, trending down, and consolidating). For the daily charts, this usually requires at least 10 years to feel confident moving forward in a live account. At least this is the experience of myself and other traders that I communicate with.
Forex Realm and Online Trading Concepts both have reliable information on many technical indicators. Here is what they say about the Stochastic Oscillator and the RSI indicator:
