Weekly Gbpjpy Breakout System | Page 3 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Breakout Trading - Simple Weekly System 183 replies

Another great Breakout GBP/JPY System (Dachel's Simple system) 62 replies

FXKRED system about 200 pips Daily on GBPJPY 189 replies

System on gbpjpy 82 replies

mphpopular "system gbpjpy - breizhfx" 23 replies

Attachments: Weekly Gbpjpy Breakout System
Exit Attachments

Weekly Gbpjpy Breakout System

  • #41
  • Feb 25, 2008 5:16pm Feb 25, 2008 5:16pm
Quoting harold4x
sometimes nf data is screwed up i use my own platform from rjofx what has impressed me is the profit in 2005 and 2006 last year shows the robustness when markets get wide ranges i think this little strategy out performs that other gbpjpy 8 hr deal by a lot
I think you may be right. It looks good.
  • #42
  • Feb 25, 2008 6:25pm Feb 25, 2008 6:25pm
Hello.

Thanks, it is a interesting system.

With FXCM data, using high and low from Friday 8:00 EST to Friday 15:00 EST, simulating from January 17, 2005 to February 13, 2008, it makes 2360 pips.

Before 2005, from 2001 to 2005, risking max 10% per trade, it doesn't work.

I also close positions when reaching stop_loss or limit or Friday at 15:00 EST.

Regards.

Paco.
Attached File(s)
File Type: txt simulation.txt   42 KB | 657 downloads
  • #43
  • Feb 25, 2008 9:57pm Feb 25, 2008 9:57pm
paco i think you are using wrong times my broker closes 5 pm on friday est so if you look on 1 hr chart its 10am till 5 but just take last 2 4 hour candles and check to see if data is about the same as other brokers. your findins are much different than mine 2200 pips for 2005 1300 pips for 2006 and 4100 pips for 2007 i did this manually so there are no errors
  • #44
  • Feb 26, 2008 1:35pm Feb 26, 2008 1:35pm
OK.

Then, If I take from 9 EST until close, I also obtain 4058 pips in 2007.
If i take from 10 EST until close, I obtain 3695 pips in 2007.

Then, I think last two bars are from 9:00EST to 13:00EST (4 hours) and from 13:00EST to 17:00EST (4 hours).

I'm going to test with other pairs such as GBP/USD.

Thanks.
Attached File(s)
File Type: txt 9-close-(2007).txt   14 KB | 510 downloads
File Type: txt 10-close-(2007).txt   14 KB | 628 downloads
  • #45
  • Feb 27, 2008 8:28am Feb 27, 2008 8:28am
Hello there.
Nice system.
  • #46
  • Feb 27, 2008 10:12am Feb 27, 2008 10:12am
Quoting harold4x
paco i think you are using wrong times my broker closes 5 pm on friday est so if you look on 1 hr chart its 10am till 5 but just take last 2 4 hour candles and check to see if data is about the same as other brokers. your findins are much different than mine 2200 pips for 2005 1300 pips for 2006 and 4100 pips for 2007 i did this manually so there are no errors
Harold,
How does Daylight Savings Time effect MT4 charts when going back years?
  • #47
  • Feb 27, 2008 10:50am Feb 27, 2008 10:50am
Quoting drayco
Harold,
How does Daylight Savings Time effect MT4 charts when going back years?
It depends on the broker providing the data feed.
With some, the time remains constant. With others it changes with DST.
  • #48
  • Feb 27, 2008 5:00pm Feb 27, 2008 5:00pm
bill, have you done a manual backtest of weeks where you had to enter a second trade because the first one got stopped out?

if so, what is the winning percentage of the second trades?

thanks, greg




Quoting billbss
I do it as harold outlined and using the method he used in his backtesting.

On Sunday evening, I place both a buy stop order and a sell stop order.
If one of them gets hit, I ride it out until I either hit the PT or SL.
I don't replace it.
If it hits the SL, that automatically puts me in the opposite trade. - If I was long, it puts me short and vice versa.

I ride that trade out until SL or TP.

If I'm in a trade, I close it out Friday evening before my broker closes.

I only take 2 trades a week. The second trade only occurs, if I get stopped out on my first trade.

If I hit the PT on the first trade, I cancel the other order.
  • #49
  • Feb 27, 2008 5:04pm Feb 27, 2008 5:04pm
harold, you broke even 5 weeks on stops and reverses for profit.

can you tell us how many times you had double losses from when the second trade also stopped out?

thanks, greg


Quoting harold4x
i have done a manual backtest for year 2005 ihave altered his plan just a bit i use the same 2 candles on north finance charts if they didnt total 100 pips i would go to 3rd candle to increase range and it that didnt do it i would equally add the difference to top and bottom. my target and stop is the size of the range not double max trades for the week is 2 they only happen if 1st stop is hit here are the results 5 weeks we broke even where stop was hit but reversal won. the rest are 36 wins for 3600 and 14 losses for 1400 total pips for 2005 2200 this is significant since the market was a lot differnt and not as volitile many systems crashed during this year 2006 was also tough but only for half of year so when i get time i will check it out this could be a nice set and forget for me
  • #50
  • Feb 28, 2008 5:40pm Feb 28, 2008 5:40pm
Had a look at other pairs such as GU, EU. The pattern doesn't seem exist there.
  • #51
  • Edited Feb 29, 2008 4:24pm Feb 28, 2008 8:09pm | Edited Feb 29, 2008 4:24pm
Due to a network error, you all have been saved from one of my novels. God was sending me a message for sure. So let me make it simpler...

I think this little channel has some power through the rest of the week. A cursory check indicates if we make a target one way or the other, there is a reasonably high probability of some payout if price reverses to our entry point. If it gets back to that point the last couple of weeks it either immediately or eventually continued on to the opposite original entry. This week was an easy winner, last week would have been a challange. I think there is more low hanging fruit here if we were to take the time to work on it. More time is NOT what I have. I am tossing this out for someone to look at if they have an interest. Might be total bunk after looking at more data... I don't know. Just tossing out something I saw.

Love what I see with this though. Okay, okay... I'll shut up now....
  • #52
  • Mar 2, 2008 6:33pm Mar 2, 2008 6:33pm
Wow, I must have had too many beers the first time I read this method. I really got this wrong, and I am glad I looked over the rules again.

I like the idea of adding an extra 4 hours to this if the price for the original breakout box doesn't equal 100 pips. And it seems to work reasonably well in the case where that occurs. I also agree with the concept of adding pips to each side of the box if after adding the additional hours the price still doesn't amount to 100 pips.

I still notice a large percentage of the cases I looked at had the situation where if the price goes to half the target (as determined by the high/low of the box) that you can make pips from the fact price will tend, eventually to retrace to trigger the opposite side of the box. I have to look at this more closely to figure the possible money management rules for this.

The bigger concern I have is, what to do about these gap opens. Today was pretty big. In the discussion to date it looks like the concensus view is, we should place a limit order and wait for price to come back and trigger. I think this is reasonable when the gap it smallish... 50 pips more or less. But for these really significant ones, maybe it would be better to wait for price to break back into the range by some percentage and only then reestablish the stop order.

So, for example, today the Market open showed a gap of 102 pips from the breakout box low to the open of the first candle. In such extreme cases it might be wiser to wait for the price to retrace 1/4 or 1/3 of the total box height. Again, in this case, breakout box height is 117 pips, so 1/4 would be 29 pips. So we might decide to wait for price to return (102+29=) 131 pips from the open before putting on the sell stop. I've seen where the retrace on these big days has a way of just continuing on.

Also, I am trying out another twist. Because from the open, price continued on to exceed the 117 pips, and I believe there is a phenomenon of price traveling the height of the breakout box once it reaches this point and retraces across the trigger price, I am placing a half lot position to try to benefit from this.

I'm still deciding how best to handle the gap. I'd love to here what Nomask thinks since you've analyzed these weeks more closely in your spreadsheet work.

I really like this system. I'm a bit surprised there hasn't been more discussion on this thread. I'm happy you folks bumped it so that I found it. This week could be a challange though, so we'll see if it works despite the gap, and which way it will run.

Nomask, thank you again for sharing this.
  • #53
  • Mar 4, 2008 12:50am Mar 4, 2008 12:50am
Nomask or anyone here,

Please show the buy stop and sell stop figure for this week based on last friday bars? I just want to confirm my understanding on this technique. A chart would greatly help me. I would appreciate it Thanks.
  • #54
  • Mar 4, 2008 1:06am Mar 4, 2008 1:06am
Fxhaz, I only have a moment so can't post a chart. But the high was 207.39 and the low was 206.22. These are my broker's numbers on IBFX. Yours might be different.

Figure from 1400 GMT to the close GMT (2059). We had a huge gap open on Sunday that has not yet closed. In the discussion in the thread we stated you would wait until price closed the gap to initiate a position on that side, which this week would be a Sell Limit.

I think there could be some improvement over this "no brainer" way of handling a gap. I'll be looking into it when I get a chance.
  • #55
  • Mar 4, 2008 7:38am Mar 4, 2008 7:38am
hey qman..... got your mail..... this is very similar to a 3 line break system, i've coded that long ago......

did a quick read and if i understand the rules correctly, gaps are probably going to whack us around here..... nomask's simple and well defined rules should make turning his thoughts into an ea fairly simple..... with a couple extra options we could compare different entry times, ie sunday's open might not be the best...... h
to trade and code, keep both simple... no call to impress....h
  • #56
  • Mar 4, 2008 8:31am Mar 4, 2008 8:31am
Hayseed, thanks for looking in... And, AaaaHaaa!, ... Good thinking! Way to think outside the box! Literally.

Three line break? Hmmm. Sounds familiar. I must have seen that around and never jumped in. Did you develop an indicator to go with that? I've been looking back here and there and I think there are several opportunities/methods for trading in and around this channel.

I really misread the rules the first time through. I somehow thought the plan was take profit a 1X the distance of the channel, and if that fails, then your goal becomes 2X on the next entry. That kills profit obviously, but the reliability of the second cross was of note. --- I guess if I was smarter I might have realized that points to the possibility of a better entry time.

I also noted a high propensity for price to move all the way through a modification of the channel once the first 1X (remember, that was my mistake) TP was hit. If price goes to the 1X point and reverses to your original entry price trigger, it tends to eventually go to the other end of the channel. Sometimes it meanders there, sometimes direct, sometimes it will give you a scare, but it gets there far more often than it doesn't. This look at it needs some work though.

If we can get a "simple indicator that builds the channel each week it would help to more quickly see if my anecdotal observations are just me being fooled by market randomness and a very greedy eye.

I'll go snoop about for your 3 line break though. That sounds interesting too.

Thanks again.
  • #57
  • Mar 4, 2008 9:51am Mar 4, 2008 9:51am
I wonder how this model would look on Oanda's 3H charts? I just finnished reading all of the posts in this thread and unfortunately I use Oanda, no 4H chart. Just wondering if anyone else has tried using this model on Oanda's 3H charts. Im currently sitting at a work computer and unable to check this, I cant wait to get home and sift through some 3H charts to see what I find. I dont have much time durring the day to study charts anymore due to work and commute, 122 miles a day. This trading model is exactly the type of thing I've been looking for. I'll post my findings after I get home if it's worth posting. Although 3H and 4H is a very big difference.

Heny
  • #58
  • Mar 4, 2008 10:23am Mar 4, 2008 10:23am
Henyloc, I got some of the same time issues you do, and so I can relate. The 4 hour bars for my broker don't match the open and close times for Nomask's broker. So, I investigated and believe if you work the 1 hour bars/candles starting from Friday 1400 GMT you will get the same result.

One contributor came up with a good twist. If the price spread, high to low for that original review period does not equal at least 100 pips, extend your analysis back an additional 4 hours (so back to 1200 GMT). If the spread is still less than 100 pips, do the math to see how short of 100 you are, take half of that, and add that amount to each side. So your minmum channel width is 100 pips, with a TP goal of 200 pips minimum. If the first try fails, your new TP becomes 300 pips minimum. And of course, if you don't reach the goal by the end of the week (say 1 hour before closing) exit your position, whatever it happens to be.

Now, the only trick for you is to find out your brokers offset from GMT, if any. Do you need help with that?
  • #59
  • Mar 4, 2008 11:02am Mar 4, 2008 11:02am
Quoting Qman
Henyloc, I got some of the same time issues you do, and so I can relate. The 4 hour bars for my broker don't match the open and close times for Nomask's broker. So, I investigated and believe if you work the 1 hour bars/candles starting from Friday 1400 GMT you will get the same result.

One contributor came up with a good twist. If the price spread, high to low for that original review period does not equal at least 100 pips, extend your analysis back an additional 4 hours (so back to 1200 GMT). If the spread is still less than 100 pips, do the math to see how short of 100 you are, take half of that, and add that amount to each side. So your minmum channel width is 100 pips, with a TP goal of 200 pips minimum. If the first try fails, your new TP becomes 300 pips minimum. And of course, if you don't reach the goal by the end of the week (say 1 hour before closing) exit your position, whatever it happens to be.

Now, the only trick for you is to find out your brokers offset from GMT, if any. Do you need help with that?
Oanda is kinda strange with the GMT thing. Im not sure If I completely have it figured out. On all my charts the 1hour candle correlates exactly to my local time. Unlike MT4 , the current candle will always be offset based on the location of the server in use for that broker. If the close of the week is 5PM EST , that would be 2PM my time PST. So how I am understanding this is, I would take the close of my 2PM hourly and go back 8hours. If within that timeframe I do not have a 100pip or more range I would go back an additional 4 hours but no more. After that if needed I would just add the difference divided equally to each side. If my stoploss is hit thus triggering my trade in the opposite direction, would I reset the trade initially stopped out? Or did I miss somthing and a re-read of this thread is in order. I really appreciate the help!.


Heny
  • #60
  • Mar 4, 2008 1:04pm Mar 4, 2008 1:04pm
Heny, you've got it, old chap!

Say a long trade is triggered. You'll stay in that until you make your profit target (channel range X 2), or the end of the the week, or your stoploss is hit. Stop loss is the opposite side of the channel, which means as you take that loss you are opening your short position. The new position now has a target price of channel range X 3. You hold that new position until you are stopped again, or the end of the week.

That is how the rules currently read. I can't attest to viability of opening a third position. I don't think that happens very often at all. Until we have a chance to look back at three or four years we can't get an idea about it.

I am hoping that Hayseed will bless us with an EA, and possibly an indicator that can help us evaluate this. His point, if you missed it, is that we might find a better time to put our orders in place beyond the Sunday open. We can't figure that kind of thing out until we get some brain power crunching on this. I just don't have any time to work the details as of yet. I'm hoping I'll get some space next week. Until then I'm working this in a very small account just because it looks so lovely and consistent. In other words, "blind faith".

Good luck. I'd love to hear anything you might conclude from your studies.
0 traders viewing now