Simple Strategy for GBPUSD

[Boletus]

Hi folks,

This is a very simple strategy for trading GBPUSD that I want to share with you; I have backtested it for 3 months (sadly this is all the data I have available in the charting system I use), over which time it generates, at the moment, a very good profit. (8470 euros in three months if you stake 5 euros per point; admittedly there is a hideous 3400 drawdown at one point, but this is due to the over-simple script version of the strategy making a series of ridiculous trades that I don't think a human being would make; and even then it comes out with a big profit).

My hope is that by combining this too-simple strategy with such things as support, resistance, price action and common sense to filter out bad trades, an extremely good income might be generated. Or not ... time will tell, since I am now playing this strategy for real (as of three weeks ago) and have so far won with it.

The strategy uses a four-hour-bar chart; it doesn't work in other time frames or on other currencies. You need to be able to check the chart at least every few hours. The idea is simply to short GBPUSD whenever the stochastic indicator registers over 80 and buy it whenever the stochastic is less than 20. Exit as soon as the stochastic moves to the opposite extreme. The stop loss is always 40 points; this gives the best results in my backtesting.

Now, if you look at GBPUSD on a 4-hour chart, it seems to be generally very volatile. (Unless this is just the case for the past 3 months!). It's full of turning points, many of which I've circled on the attached chart which you can hopefully see.

The stochastic helps enormously to figure out where these turning points are; as I say, over this period, if you simply bought and sold according to the stochastic alone you'd come out with a big profit. But I imagine that, if you leave some of the possible trades aside by insisting that

a) The price must be showing signs of reversal
b) The price must currently be at a support or resistance level and
c) Avoid trading against any strong trend where such a trend is obvious and there is no good reason (such as fundamental considerations) for thinking the trend will fizzle out pronto

... .then perhaps this strategy could be highly lucrative.

What I like about it is that there are a limited number of points when a legitimate trade can be made; it is then simply a case of trying to filter out the bad ones to minimise losses. The script actually loses 66% of the time, and yet wins overall; the winning trades tend to be pretty large.

A 40-point stop-loss at 1 euro per point costs you 40 euros; but then a single winning trade might easily make 200 or more, as you can see from the chart. So really you're using up these 40-euro "bullets" gunning for those big hefty reversals.

I'm not pretending there's anything magic or original about this; I'm only trying to apply what I've learned from forex books. After a few months of looking at various charts and trading in various time frames, I've decided to concentrate all my efforts on this one.

Right now I have money on GBPUSD going down; that's because it fits all of the above criteria at the time of writing (at the rightmost position on the chart I've attached to this post) and I'm hearing a lot of bearish sentiments about the UK economy. If, of course, GBPUSD now instead surges upwards by 500 points, it'll just about wipe the floor with me since I'll keep expending 40-point stop losses only to have them tripped ... but of course that's why I ought really to be trading 1% of my tiny account, not 5% as I am currently.

Any feedback welcomed ....

Boletus

PS

This is a ProRealTime script for anyone who can use such a thing:

stochasticFast = Stochastic[21,3](close)

conditionGoShort = stochasticFast > 80
conditionGoLong = stochasticFast < 20
conditionExitLong = conditionGoShort
conditionExitShort = conditionGoLong

if conditionGoLong then
BUY 1 SHARES AT MARKET THISBARONCLOSE
elsif conditionExitLong then
SELL 1 SHARES AT MARKET THISBARONCLOSE
endif

if conditionGoShort then
SELLSHORT 1 SHARES AT MARKET THISBARONCLOSE
elsif conditionExitShort then
EXITSHORT 1 SHARES AT MARKET THISBARONCLOSE
endif

 

[Boletus]

I missed a crucial bit of information off this. The stochastic has to be set to 21,3 on the four-hour chart. i.e. 21 four-hour periods, not 21 days. Possibly I've inadvertently "retrofitted" this value to a 3-month period, when this value isn't any good for any other 3-month period! Time will tell .... At the moment I'm still in profit from it.

 

[hirnant]

Hi Boletus
Thanks for the info - I've just set it up and looks good.
Do you set yourself a TP of 200 pips or do you wait for the market to turn?

Sorry I mean the Stochastic....not the market!!

 

[Boletus]

Hi Hirnant,

I wait till the stochastic hits the other side. Eg. if I sell short when the stochastic is over 80, I close the position when it's below 20. I actually have not followed this strategy since posting this - I got cold feet; but meantime I've LOST money on other strategies and this one WOULD have made a great profit!!! So now my resolution is to trade this alone, with 1% of my account. Slow but steady does it. I've also turned on Bollinger bands since I see that very often the price actually turns when it hits a band, although the basic strategy remains the same.

The trade I posted above worked out perfectly, and there's another one at this very minute (although of course any individual trade may or may not work -- the idea is to make money overall, which so far it does).

 

[Boletus]

I'm thinking, the main problem with doing this is that it's not going to work when GBPUSD gets into a strong trend. Then hopefully it's just going to be a case of trading only in the direction of the trend. In other words, trading pullbacks and leaving contra-trend trades on the table.

 

[hirnant]

The Bollinger certainly adds extra confirmation and helps with the timing.
I assume you are waiting for the candle to close above or below the band?

 

[hirnant]

Have you looked at Eur/JPY with the same settings?

 

[Boletus]

Yes, or at least I figure that when the price is at the band, there's an increased chance it's a major pivot. My whole strategy here is based on trying to spot the correct pivots while weeding out the bad ones, so anything that helps must be good. Well, I got stopped out of the above trade, so will have to try again. I've shorted GBPUSD again now since a poor monthly manufacturing report came out from the UK (I'm using the free calendar at Forex Factory) and it looks to me like there's a strong resistance line at about 16270.

I had a look at EURJPY but my programme doesn't make a profit on it. In fact it doesn't even make a profit on AUDUSD which looks quite similar to GBPUSD. So I'm concentrating my efforts on GBPUSD, trying to learn the ups and downs of this one pair.

 

[hirnant]

Hmmmm. I go t stopped out on cable as well, although currently 50 pips up on the Eur/Yen.

I too am shorting the cable based on news and your settings. Have you tried going down to a 1 hour candle to get an early indication and then moving up to 2 hour, etc?

As it comes to the end of a cycle based on the stochastic how about setting a trailing stop to avoid coming out of the move too early, i.e. let your profits run.

 

[Boletus]

Cool, I hope this one comes off! As I say, I'm a rank amateur at forex, but this strategy looks promising to me.

Yes, I'll even look at 5 minutes to try to find the turning point. Anything that helps. Your suggestion sounds very good to me; I'll definitely give it a go if the stochastic drops to 20 and doesn't show immediate signs of turning.

I'm also trying to keep track of scheduled economic announcements to try to build up a feel for which ones signal pivots; although it seems like often the pivots come about for reasons having little clearly to do with the news.

 

[hirnant]

Rookie.....me too. 18 months and still no consistency.

Spent way too long staring at 5 minute charts, which is why this strategy appeals to me.

As ever, it's all about entries and exits - blinding obvious that one really, but notoriously difficult to do.

I will update you with my progress.

 

[Boletus]

I know what you mean Hirnant, these 5 minute charts sent you crazy after a while.

Well, this last series of trades has been gruelling and not very rewarding for me -- my profit has been in the region of 10 pips. I got stopped out 6 times (6 x 40 pips = 240 pips) and eventually made 250 pips on the drop, when it came. Once the stochastic hit 20, I followed the falling GBP with a stop that I moved down manually. Now I'm going to be looking for the next rally upwards.

 

[Boletus]

A quick update now I have more experience .... You cannot simply execute a strategy involving one indicator and expect to make a profit indefinitely. I got over-excited during a phase when GPB/USD was in a "frothy top" and this strategy happened to work. But my original post was extremely vague anyway, so I'm not totally sure what I was trying to say.

However, from my extensive reading on these forums and in books, and my experience to date, it seems likely to me that making only a handful of trades a month, looking at a daily chart, is liable to be more profitable for most people than peering at 5 minute charts, or even hourly charts.

For the inexperienced, indicators are highly seductive .... but now I lean towards the view that they are really only useful once you've learned to trade successfully without them. Then they are the icing on the cake. However, I don't rule out the possibility that my current opinion may also be quite wrong ... More updates in the course of time ....

 

[oiltanker]

taking stochs both ways will mean losses [unless its ranging]. find the trend say off the weekly then take those in the trend direction ideally off a round number. the loss will be the last one as the trend turns. the other problem is the turns might happen at 2am.

have you looked at hieken ashi to help? hieken 1hr might help with 4 hr trading?

 

[Boletus]

Thanks Oiltanker, good advice. I am going to leave the indicators largely alone for a bit, but I see that looking at long trends and timeframes seems to be a good way to go. I would now not dream of trading against a major trend.

 

[Pipmaster1]

Boletus,

Interesting method. Something that may help qualify your stochs being extreme is RSI 3 with extreme being 90 and 10. I also use RSI 14 with extremes of 70 and 30. Having stochs extreme along with RSI 3 may help filter out some bad trades. Having all 3 extreme would be an even tighter filter! My RSI's are on a 30 min chart, not a 4 hour however. Anyways just a thought.