Hi guys,
I am trying to backtest different trading strategies and have a question. Was 2006 a particularly rough year in forex (ranging with not alot of trends)? I have tried some backtesting that seems to work from 2008 til now but 2006 doesn’t seem to be very profitable for any of my strategies.
Thanks in advance.
Hello,
Well first: may I be the first to congratulate you on the fact that you’re actually backtesting your strategies using a decent amount of sample data (a few years worth of data) unlike most (myself included when I started) that test their strategies using a few days worth of data and THINK they’ve finally found ‘it’!!!
To answer your question: if your system is a trend following system it would most certainly have worked very well from about the time that the sub-prime ‘debacle’ ‘reared it’s ugly head’ right up until the ensuing credit-crunch for the simple reason that there were HUGE uptrends or downtrends (depending on what you were trading i.e. I trade equities so I’m referring here to the Dow for example but you’ll find that pairs like EUR/JPY and GBP/JPY moved in the same ‘fashion’). The point is that if you ‘caught’ a trend and stayed with it there were fortunes to be made (IF you followed your trading systems that is). I cannot comment on what happened prior to these periods on FOREX though but what I can tell you is that you should continue with your tests as far back as you can because what you’ll most probably find is that certain pairs will have had bad years and good years at different times (if you have it available look at something like EUR/SEK before sub-prime for example i.e. that pair trended for months if not years before that if memory serves me correctly).
I personally don’t believe that one can be profitable OVERALL (over a long period of time) trading FOREX only trading only ONE pair because when one pair is range bound there are almost always others that are trending and the nett result should be profitable so don’t lose ‘faith’.
Let me also say this though: it took me a LONG time also to realise that there is no SINGLE trading system that will work in ALL markets ALL of the time (oh how nice and simple that would be and believe me when I say that I spent years trying to prove that there was INDEED such a trading system) so just because ONE system does not make ‘millions’ each and every single year it does not mean that a combination of systems, each achieving different results at different times, will not have overall good results.
Regards,
Dale.
There is value in backtesting, but at best it should only be taken as a guide, trading at the coal face is totally different to backtesting, and you can’t take into account all the news announcements that may or may not have spiked you out and the such.
Well you’re right to a point.
Put it this way and as you probably already know: when I talk about ‘backtesting’ I’m really actually talking about ‘paper trading’ (literally) not running some type of ‘automated backtest’. That way: you SHOULD be able to take all things into account e.g. price spikes due to news data releases (that of course depends on how reliable and honest your broker is of course and that unfortunately is an unknown unless you’ve been trading with them for a good while). The only thing to watch out for when ‘paper trading’ is that you don’t ‘fool yourself’ or ‘rationalise’ bad trades because you can ‘see into the future’ as it were. In other words: because you have the benefit of hindsight it’s sometimes easy to ‘fool yourself’ that you would NOT have taken this or that (bad) trade for this or that reason when in reality, WITHOUT the benefit of hindsight, you INDEED would have taken that (bad) trade. That’s the only pitfall. That, and, of course, variable spreads (although by ‘flipping’ between bid and ask charts at the appropriate times, you should be able to eliminate this unknown to a large degree).
Of course and as we know: backtesting is no ‘guarantee’ of future results but let’s face the fact that if you backtested a trading system using, say, ten years worth of data, and it NEVER showed a profit then at least you know that the said system is useless and you’d be a fool to think that THIS year it’s going to be profitable.
Regards,
Dale.
Back Testing can help allot with understanding your trading system, however a problem that I found was looking ahead from where a possible trading indiaction appeared. I.e into the future, which cannot be done when trading live!!
Ive backtested my theory on the GBP/USD from Jan 2009 to Jan 2011 current. Its taken me 9 months to do so, however I now know my system like the back of my hand. The key when back testing, and the difficult part which people fail to mention is this:
Think LIVE when using historical data, and try to blank out what the right hand side of the chart shows you. Its very easy to see that you would have lost your trade, and say to your self “well i would’nt have traded at that point anyway”.
I find scrolling the chart to hard right hand side of the opening point of my trade helps! this way you can then scroll foward in time and see what WOULD happen, not what HAS happened.
Thats my key objective when back testing. Its boring, but well worth the graft.
James
May I also add that I’m a day trader, hence why the backtesting took nine months to complete. Going over each and every trade for each and every day equates to allot of time and work.
I did not use a automated back tester as ive tweeked my system when I see problems forming. Of course, I then had to go back over the already analysed data, and finally the remianing data which was left.
A very very long process. But I have learnt allot, and even wrote up a trading guide and question sheet to constantly remind myself of the rules that I must stick to in different trading situations.
James
Thanks for this. I guess I was looking for that holy grail that does not exist-lol. A few days backtesting is what many people do before trading live? No wonder their accounts get vaporized. I try to stay with the “majors” because valid or not I am concerned with liquidity with the non majors…One more question- How would one know when a currency generally stops trending before trading- other than trial and error- taking trades over the course of time and losing them
Amen…I found myself “subconsciously” “adjusting” my paper trades by looking into the future…Not good…FYI- I also prefer to paper trade versus the auto backtesting as others have already mentioned
Well this depends on allot of factors, the first and what I would put as the most important is the time scale you are trading. A 1 hour chart may have had many trends stopping and starting with in it when drilling down to a 5 min chart, and even more when drilling further down to a 1min chart.
There are many indicators out there which will HELP you determin when a trend is possibly running out of steam, but this is not going to be 100% accurate. If we all knew when a trade was about to finish we would all be in the money!
There is a huge amount of technical analysis out and about which aids the trade trending approach, far to much to even talk about!!!
At the end of the day, trading in its simplest terms is about trading a trend, wether it be a small trend for a few pips or the big on that we all whish for which could be 1000 pips or more!
In 2006 China made a decision in regards to the RMBUSD relationship. That particular decision influenced the global currency markets fundamentally.
Oh my – - let me ask, do you like to be right?
.
.
.
.
.
.
.
I do , and that is a big problem in this biz. . I am working on it and I believe I am getting past that stupid stupid stupid stupid . . . . stupid notion.
As in other area’s do you want to be right, or do you want to be
‘profitable’
‘happy’
‘good’
etc. .
I am choosing the latter, but for a person who likes to be right, it is a bit difficult, I am taking the red pill so to speak. when i should be taking the blue.
PeachTird,
I’m not too sure where you have taken this thread, but it seems to have moved from the original question which was, what was a good starting point to benchmark a new Forex strategy? You’re now talking about emotional and social factors such as ‘do we want to be happy, good, and profitable’.
Of course we want to be good, and being good means we are being profitable. There is no other benefit from trading apart from making a NET amount of money. You don’t build up relationships with colleagues, you don’t make friends, and it’s not like any other job.
People often say in life that you shouldn’t be working for the money, and people may disagree with me here, but anyone who says they trade for fun, and not to make money is simply either crazy, or not telling the truth.
The three areas you listed are Profitable, Good, and Happy.
There is no distinction between them when relating to trading. If you losing more money than you’re making then you are none of the above. If you are making more money then you are losing then you are all of the above. However, you can’t be just one of the above three factors you listed.
Maybe in a short-term view you could have won a few trades based on luck, therefore you are Profitable, but may not be happy and as you know it was luck! But in the long term view there all the same.
Does that make sense???
And if someone offers you a blue and red pill, ask what they are before hand! :rolleyes:
James
apologies for sidetracking the thread.