Im a fairly new forex trader, and i’m begining to develop my own trading strategy. my idea is the use two sma (5 and 13) lines and the rsi on a hourly chart. when the 5sma crosses over the 13 sma, and the rsi is rising above 50 ill buy. when the 5 sma crosses below the 13 sma, and the rsi is falling below 50, ill sell. my target is 20 pips ber trade, and ill risk 30 pips on the trade. I plan on trading the eur/jpy almost exclusively, and will never trade on any news releases. i have never done any backtesting, and i plan to start soon. But can anyone poke any obvious flaws in this strategy?
Yea the obvious flaw is you are risking more than your planning to make. Lets say you take 10 trades and 5 of them are winners, 5 are losers…
your 5 winners net you 100 pips of profit.
your 5 losers net you -150 pips of loss.
This means your strategy lost you -50 pips of profit with a 50% success rate (which is pretty realistic). Change your money managment and you only have to be right 50% of the time. To make money your going to have to be right 3 out of every 4 trades. Thats a pretty high hit percentage, especially if your just starting out.
i hear ya daedalus, it just seems to me, in my limited experience, that when i see those two indicators cross, the next bar always seems to shoot right up.
the market does sometimes drop, but in a market like eur/jpy with a pretty big range, 20 pips is a decent target to reach
but as Dae said even if your right 3 out of five times you will only break even, you should either aim higher or make your stop loss less.
My thinking when it comes to this strategy is that when the 5sma crosses the 13 sma, it happens because the price is trending, so im trying to capture a small piece of that momentum. When this situation occurs and i look at the charts, quite often the price goes up, up and away. it happens more than 50% of the time. if the market does go against me, im only down 30 pips. is my logic off? can anyone recommend a free service where i can backtest this system?
The problem is that, eventually, you’ll hit a string a losing trades. That’s why the other comments are saying that you will end up losing more than you win in the long run.
You can backtest your strategy with MT4.
A lot will also depend on which time frame you’re using. Yen pairs tend to move a lot, so 20-30 pips doesn’t give you much room without getting stopped out, unless you’re going to trade the smaller time frames.
Of course, smaller time frames will subject you to more volatility, so those MA crosses can turn back on you quickly. As previously stated, with your Stop Loss higher than your Take Profit, you will probably lose more than you win in the long run.
Terry
thanks for the feedback fellow trader, much appreciated…
so is this a situation where i have to stick to my stops, but let my profits run?
lets say i trade two lots. if i put a 30 pip stop (risking $60), and i put a 60 pip limit on one lot, and let the other lot run (meanwhile moving the stop up 60 pips in order to break even), is that more feasable?
how do you guys determine risk/reward? is it the same on every trade, does it depend on your capital?
so is this a situation where i have to stick to my stops, but let my profits run?
lets say i trade two lots. if i put a 30 pip stop (risking $60), and i put a 60 pip limit on one lot, and let the other lot run (meanwhile moving the stop up 60 pips in order to break even), is that more feasable?
how do you guys determine risk/reward? is it the same on every trade, does it depend on your capital?
The problem with this question is that it is very subjective, it depends
very much on personal feelings, tastes, or opinions.
But the basic idea is that you will not win every trade, however good
your technical or fundamental analysis is, therefore you need a cushion.
But if you can win 50 out of 100 trades, with a R:R of greater than 1:2 then
surely you will be profitable? Just do the math.
As to trailing stops, it also depends upon your platform whether this
function is available. If not you will either need to spend the time to monitor
those trades, or just use a fixed S/L : T/P.
All this needs to be hammered out in your own mind while demo trading.
That is why it is so subjective.
Well lets play it out.
Lets assume again your just trading 1 lot to keep things simple. Again, assume a 50% win rate.
your 5 winners net you 300 pips of profit.
your 5 losers net you -150 pips of loss.
for a net profit of 150 pips now using a 1 risk: 2 reward ratio. See why this is important?
What is also important is removing risk from your trade as you mention, moving up stops, etc. Its very realistic to assume that every trade won’t hit your +60pip goal. So we need to reduce that risk systematically as the trade performs. So do some backtesting with your system and see how many stop out for -30, and how many go the distance for +60. I think what you’ll find is that a trade may go lets say +48, and then reverse in your face. Are you willing to take a -30 pip loss on this trade? Probably not. But at the same time i have found its best to keep your trade managment systematic and unchanged trade to trade so it removes your emotion from the situation. If you don’t belive me, you haven’t traded real money yet and just trust me, i’m trying to save you the pain!
So then see what would happen to the results to if you say moved your stop up to breakeven after each trade hit +30 pips. Then your risking nothing to make +60 pips… See how that management affects a string of the say the last 10 trades your system fired off.
Its time consuming, but it is a necessary process that will either make you or break you when its time to go live!
Someone very wise once said to me:
“Nothing in trading is really valid or invalid, only testing makes it so”
Everything everyone here is saying makes good sense but don’t take anyone’s word for it. Do your own testing and the truth will quickly become apparent as to what works.
Without having tested your strategy myself, it does seem to me that it is trend following in nature. So when the market decides to range your strategy will get chopped up pretty badly (and markets do spend alot of time in ranges).
But again, do your own testing over many many trades and see what is true for yourself and make up your own mind.
ScroogeMcDuck - nice name!
Be careful about changing your stops or TPs purely to alter the risk:reward ratios. It’s hard to calculate it properly because it rarely takes into account the probability of a move actually happening.
Better as others say to backtest and see what works. Most strategies are woefully poor when backtested, so it’s really, really important to do this first. Be careful too about doing it by eye without measuring your results - it’s very easy to spot the occasions when it works and miss out the ones where it fails miserably.
To give you a headstart though, I tested a bit for you as I know it can be brain-scrambling when you start. Based on just this year’s data, you might want to tweak it a bit - see chart 
Try this. Drop it to 30 mins TF, scrap your TP levels, and instead close out everyday by midnight. Change your stop to 150 (yes, 150), and this year, you’d have got the 2nd graph.
Best of luck!
Hey triphop, how did you get that graph of the results?? It seems like a program because I can’t imagine you’d manually test it out to be what looks like 100’s of times (but if you did, kudos!).
If it’s a program, may I ask what program? I’m curious but know that I probably never will find a automatic backtester that can handle the subjectivity of candlestick trading. It sure would be awesome to get instant results on the success rate of a certain pattern for a certain pair!
I got some freetime, but not that much free time!
It’s Ninjatrader. Bizarrely, it’s not popular at all, but for those who can’t program anything more complex than a VCR (me), it allows you to knock together strategies really easily with a wizard-type thing. This really took about 15 minutes.
As for testing candlesticks, no reason why not. They’ve got set rules, and if it’s subjective (for example when candle x penetrates the other candle by a significant amount), you could just test different amounts (10pt, 20pt, 30pt etc).
Beats the hell out of just believing everything Steve Nison says 
One of the reasons im developing this strategy is because i feel like this plan forces me to become a more disciplined trader. If I only make a trade when the two sma’as cross, im limiting myself to a certain amount of trades, taking emotion out of the equation.
Also, I know that a market can shoot up 30 points and reverse, thats why my initial profit target was only 20 pips, something i think is quite reasonable. And i feel that regardless if the market is trending or ranging, when the two sma’s cross, doesnt that mean that theres a short term rally in price (either up or down)? So in a market like eur/jpy with a daily range of 100+ pips, 20 pips is a decent target.
P.S. thanks for the feedback, this site is an incredible resource, and i hope to one day be a member of the fx-men!
You make a good point, but have you done any research to see how often it does that? If it does it quite often, then you’d be stopped out of quite a few trades with a 20 pip SL. However, if your SL was 40 pips (just as an example), then you wouldn’t be stopped out so often and would have more profitable trades.
Scrooge, did you look at those graphs? Whether or not you think 20tp is quite reasonable or not, you would bee wiped out with those settings this year. Backtesting isn’t the be all and end all of everything, but you should at least start with something that has worked at some point in the past.
Don’t take my word for it. Test it and save yourself a headache.
Hey TripHop, thanks for the complement, and thanks for your research!
O.K. so it looks like my 1st attempt at a trading strategy is not going to work 100% of the time, so maybe I can tweak it a little so that it can work a percentage of the time. Im going to look into adding another indicator or something to the mix.
BTW TripHop, my strategy also included the RSI as part of the equation, did you incorporate this into your study?
Yep, the two tests are both using your settings suggested at the start of the thread, both of them using 50 RSI (couldn’t improve on that), but the 2nd uses different time frames, TP and SLs.
I found when I started out I always made my stops too tight, and it screws with your winnings.
just give the cowabunga a try at first and then currupt it in your own image. Ninjatrader is a cool program man, i might give that a go!!
I’ve thought about trading using the cowabunga system and a few other trading systems I’ve come across, but I eventually decided that I would learn a lot more if I attempted to create my own system. However, I also know that it’s impossible to rely on just one system. For example, what might work in a trending environment wont work in a ranging one.
How many different systems do you seasoned traders have in your “toolkit”? Do you rely on one or two main systems? And do you wait for the system to set up, or do you fit the system to whatever is happening in the market?
For example, my system relys on two sma’s crossing, so I am mostly waiting for the setup to occur. Are there any systems that can “predict” whats going to happen with a decent probabilty, so you can get into a trade whenever you want without the wait?