This is an idea that appears viable in visual backtesting. I haven’t actually traded it, and it would be a bit difficult for manual trading, as the orders would need to be executed rather quickly. We will use two ideas that will require a broker that allows hedging, but you can trade one or the other separately, if you wish.
The first idea is to trade breakouts of the envelope trading bands. The other is to trade the range between the bands. I would strongly recommend the EUR\USD pair on a 1 hour TF chart. NOTE: these are NOT bollinger bands. Some charts have this indicator as a moving average envelope. We want 20 period setting on the close, 0.1% deviation from the average.
Here’s a pic to give you a visual: chartss.bmp (791 KB)
So, here are some rules, starting with the breakout idea.
Buy when price breaks the upper band, with SL at the lower band, and move SL to breakeven at +5 pips, or so. Close the buy when the price goes below the upper band, or you can experiment with fixed TPs.
Sell when price breaks the lower band, with SL at the upper band, and move SL to BE at +5 pips, or so. Close sell when it goes above the lower band. (I wouldn’t close it with just a touch, as price often bounces off the bands, after breaking out.)
And the ranging idea…again, this requires ability to hedge, or use different account, because you might have an opposite order open, temporarily, with the first idea.
Sell when price touches upper band, with 5 pip +spread SL, and opposite, lower band for TP.
Buy when price touches lower band, with 5 pip +spread SL, and upper band for TP.
the average distance between the bands is about 25 pips, so this is a constant 1:5 RR ratio, very good.
When price is +5 pips, move SL to BE.
So, if we were to trade this on the above chart starting at the beginning of this week, we would have a breakout sell, which nets 126 pips, followed by a ranging buy for 25 pips, followed with a breakout buy that hits BE, repeated a few times, with a ranging sell, also hitting BE, with a breakout buy finally getting away, and at +55 pips at the end of the chart.
This looks very promising. I quickly back tested manually the last 2 months, and WOW!
Looking forward to some live forward testing to see if it keeps up to expectation. The simplest, most effective system I’ve come across so far. And I’ve been looking. Out performed the system I’ve been working on for the past 3 months, by about 4 times as much pips. Love that it works when trending or ranging.
I’m actually checking it on the M30 chart as well, and still looks very good.
I am still very cautious, unless the candle clearly moves beyond the upper band for a buy, or lower for a sell, I wait for the 2nd candle as confirmation of direction. Might cost me 10 pips. But I’ve noticed sometimes it’s a fake signal. I also use a max 20 pip stop loss, otherwise I don’t stick to my money management rules if I place the SL at the opposite band. But on back testing it was only hit a couple of times. Not enough to worry me compared to the times it makes a profit. I think I had 64% wins. But the wins ranged from 30 pips to 150 pips.
I can’t sit at my pc all the time, but if I work on moving the SL to BE, the win ratio will increase exponentially I think.
As I said, it was a quick back test and I didn’t check for moving SL.
Will try it on the forward testing.
I will, at some point in the next few days, code an EA for this. It will deal with moving the stops and eventually ( though not at first cut) allow for money management to allow the stop at the opposite band.
Thanks for your interest!
It may be the best to figure out some kind of buffer, because when the price just sits ranging right on a band, that can cause an awful lot of small losses that add up. Fewer trades that succeed are better than many that don’t, even with small losses.
Also, I would strongly recommend you either make two EAs, one with the breakout idea, the other with the ranging idea, or make the EA have the option to turn either style on or off-this way people can decide what they want to use, if the other idea doesn’t seem to pan out. Thanks again!
Guys, nice little strategy.
I back tested it a little bit yesterday on my Forex Tester 2. Though I only worked with the trending way of trading, it looked very promissing.
I also used the regular SL rule of placing it on the oposite band and moving SL to BE when profit is +5 pips.
Unfortnately, I forgot to save the dang test.
Ill do it later today. I want to take at least 30 trades on 2012, following the rules as explained in the 1st post. Ill let you guys know of the results!
This was a trade example from last Friday. A nice 45 pip profit.
It was on a M30 chart, got in a bit late, but was entered the moment I spotted it. To stick to money management, placed SL at 20pips.
Exit at the arrow indicated. Was heading of for the weekend so closed the trade. But looking back it was such a bad place to exit, I think.
the offset is 0.1 of average (see the posted charts, they have the settings just above the chart). Not smoothed or exponential, just simple moving averages.
Which is the best time suitable for this strategy ? I mean, Asian, London ???
If you want to use time as a tool for this, the london is best for breakouts, asian is best for ranging inside the bands.
Generally typing, the bigger sessions, the london and US will be best for breakout types, and the asian (everything else) will be ranging, so best for inside the bands trading.
So you’re looking for either a bounce or a breakout from the MA envelopes… But the issue here is… Which one is the best one at a given time…
It seems to contradict itself.
Now, if you could find a way to determine when the market is ranging and when the market is trending, you could trade the bounce or breakout respectively.
True, that it is difficult…but generally speaking a skilled, quick fingered person (or EA) can trade both at the same time, and just have hedging entries. It may prove to be a bit easier, at least for the learning portion, to concentrate on breakouts first, until you get the hang of it, and then the ranging idea, each by themselves.
