4H Box Breakout | Page 105 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Box Breakout indicator 197 replies

Breakout EA based on New York Box 3 replies

4H Weekly Box Breakout [TD] 4 replies

EURUSD Box Breakout 6 replies

Breakout Box 5 replies

Attachments: 4H Box Breakout
Exit Attachments

4H Box Breakout

  • #2,081
  • Aug 19, 2009 11:30am Aug 19, 2009 11:30am
Earlier in the thread someone had mentioned using this Method on a Daily basis... (i.e refiguring your box every evening) I tried it last night with the EURJPY and USDJPY last night...both went short within minutes and both hit my tp which was 2atr+spread sl were set at atr + spread. When I say ATR I am referring to thebox total not the Average True Range... I'll keep playing with it and let you know how it goes...
  • #2,082
  • Aug 19, 2009 11:51am Aug 19, 2009 11:51am
For those of you who are still using the original rules as I am, the week ended with BE on the short and a loss on the long.

My results are:

GBP/JPY BE on short
-3.44% on long

EUR/JPY BE on short
-2.64% on long
Attached Image (click to enlarge)
Click to Enlarge Name: be + loss.gif Size: 65 KB
  • #2,083
  • Aug 19, 2009 12:05pm Aug 19, 2009 12:05pm
Quoting hafizal79
I think it will go long tommorow......
I think it could be a long move short, or a short move long , or better yet, I've had my two trades for the week and am content to wait for a new week.

I've made some very minor changes to the system but I'm sticking to a maximum of two trades a week. When the market is ranging like this week I personally prefer not to keep guessing what the market might do. I'd rather relax for the rest of the week and enjoy the profits I've gained so far this month. But to each his own.
Greetings from Guanajuato, Mexico
  • #2,084
  • Aug 19, 2009 12:28pm Aug 19, 2009 12:28pm
Quoting Ace284
For those of you who are still using the original rules as I am, the week ended with BE on the short and a loss on the long.

My results are:

GBP/JPY BE on short
-3.44% on long

EUR/JPY BE on short
-2.64% on long


I am going now back to last year and testing trading according to original rules with moving S/L to B/E when price moves 1 box in our direction with final goal of TP3. so far it looks very good. Obviously losses happen in ranging market. I am testing according to no partly profit taking.
I can only say, it requires a lot of patience and discipline on traders' part.
Let's see, I might start trading 100% according to your rules
I set only take 50% of 50 pips but I am ready to discard it.
as you saay, it is probabilities games. Yes, ay be I trade 1 account according to your rules and another one with minor change . I also trade no more than twice a week.
  • #2,085
  • Aug 19, 2009 12:37pm Aug 19, 2009 12:37pm
By the way. we have 2 trades according to the system each week. Is it only 1 short and 1 long or it can be 2 shorts or 2 longs? For example, price triggerd short trade, reached TP1, we moved S/L to B/E. B/E got hit. then price moved back down, should we take short position again?
  • #2,086
  • Aug 19, 2009 12:58pm Aug 19, 2009 12:58pm
Quoting shiko
By the way. we have 2 trades according to the system each week. Is it only 1 short and 1 long or it can be 2 shorts or 2 longs? For example, price triggerd short trade, reached TP1, we moved S/L to B/E. B/E got hit. then price moved back down, should we take short position again?
Ace has consistently posted a Long and a Short box for the week, never 2 longs or two shorts. I don't think he's ever suggested re-entering a losing trade. Others have suggested variations and they may or may not be successful with the changes, but it's not part of the original system.
Greetings from Guanajuato, Mexico
  • #2,087
  • Aug 19, 2009 1:05pm Aug 19, 2009 1:05pm
Quoting yen44x
Ace has consistently posted a Long and a Short box for the week, never 2 longs or two shorts. I don't think he's ever suggested re-entering a losing trade. Others have suggested variations and they may or may not be successful with the changes, but it's not part of the original system.


Thanks, I have not followed this topic from the start, so I did not know about it. It is good to know now that it is only short and long per week.
I am comparing now what is better, taking partial profits or no.
  • #2,088
  • Aug 19, 2009 1:59pm Aug 19, 2009 1:59pm
ok, some interesting facts. If we let this system work exactly as Ace says with moving stop loss to B/E once the length of the box reached and with total goal of TP-3, we would have about 34%, risking 1% per trade since January 1, 2009.
The worst case scenario per week is 2% loss if both trades went wrong.
If we are looking for TP3 and moving stop loss as Ace say, we can have either TP1 or TP3 win. Never 2%. It is if we let the price hit take profit without closing it manually.
  • #2,089
  • Aug 19, 2009 2:09pm Aug 19, 2009 2:09pm
Quoting Ronald Raygun
MoveStopTo is the pip distance from the order entry price (you should be able to put in a negative value for that).

I'll look into it.
Thanks Raygun

I have modified your EA to add the separate OrderModify for ECN brokers as I needed it to do some backtesting on reliable Live ECN historicial data I have. I also added a Step function to the trailing stop so that every pip does not send a OrderModify on the SL. It makes backtesting go quicker.

I won't post this mod as I think you are doing this and I don't want to interfere with your sequence of releases. I wil leave that up to you.

In running some visual backtesting, I noticed some potential problems:

1. You do not have the option to close on Friday. I think this is important as I noticed that often, a trade stays open over the weekend and interferes with potentially profitable new trades on Sunday/Monday. Also, the lack of a close on Friday, does not allow me to compare my ideas with Aces concept of letting things run with no interference and the order closes at the SL or automatically closes on Friday. I will probably dig this code out for myself and mod your EA but I think you should add this to your published EA

2. I have been reviewing my idea of the Trailing Stop with a new Pending Order concept and I am convinced that this method will add serious pips to the basic trading of Aces method. I am simply not prepared to walk away from a 500 or 600 pip profit already earned when a serious reversal is threatening to wipe it out, just because the original method says this is fine because back testing shows the overall method to be still profitable. That just does not make sense to me.

If this modification works, with no risk, then it can be used but only with an EA. It is too hands on to work consistently well manually. My coding skills are not good enough to handle point 2 so I hope you can work out a coding plan for this concept.

Dave R.
  • #2,090
  • Aug 19, 2009 2:28pm Aug 19, 2009 2:28pm
Hi all,
This is my first week of this system and I am using Danno96's revised rules as they seem to give a higher pip count with his back testing. So far down 49 pips allowing for spreads. Just resentered EURJPY short.
I am having to draw in the box lines manually as I cannot get the indicator on the first post to work. Please see the screen shot and I would be grateful if anyone has any suggestions on how to make it work correctly.
Also I am assuming you guys are using spot prices from your broker and rolling every 24 hours rather than near quarter prices. My broker is IG index.
Cheers, DJ
Attached Image (click to enlarge)
Click to Enlarge Name: ScreenHunter_01 Aug. 19 20.23.jpg Size: 169 KB
  • #2,091
  • Aug 19, 2009 3:24pm Aug 19, 2009 3:24pm
Quoting droland
Thanks Raygun

I have modified your EA to add the separate OrderModify for ECN brokers as I needed it to do some backtesting on reliable Live ECN historicial data I have....
If you have a modification, I encourage you to release it. Just put in some prefix or suffix so any new users know what to get.

I'm trying to find a way to use file versioning in my programming so that every time I save the file, a new version is made somewhere (and added to the name) something like "4H Breakout EA build 314159" or something like that. If you could direct me to some programs which do that, that would be great. Also, check your e-mail (whichever one you used for my EA support website)
  • #2,092
  • Aug 19, 2009 4:38pm Aug 19, 2009 4:38pm
Quoting Ronald Raygun
If you have a modification, I encourage you to release it. Just put in some prefix or suffix so any new users know what to get.

I'm trying to find a way to use file versioning in my programming so that every time I save the file, a new version is made somewhere (and added to the name) something like "4H Breakout EA build 314159" or something like that. If you could direct me to some programs which do that, that would be great. Also, check your e-mail (whichever one you used for my EA support website)
Sorry, I don't know of any automated version extenders. I love your example of the version number though pi=3.14159265. You have a Physics or Engineering background I presume.

Dave R.
  • #2,093
  • Aug 19, 2009 4:48pm Aug 19, 2009 4:48pm
OK guys please ignore my previous post in terms of the setup. I now have a FXDD account and all is well including the period seperators. On ODL MT4 the period seperators were one 4hr candle behind. Now I cannot get the boxes the same as Ace has them at post 2082. The box sizing does not include the 20 pip buffer. Any suggestions? Mr Raygun?
Many thanks
Attached Image (click to enlarge)
Click to Enlarge Name: ScreenHunter_02 Aug. 19 22.35.jpg Size: 137 KB
  • #2,094
  • Edited 5:12pm Aug 19, 2009 4:59pm | Edited 5:12pm
Quoting droland
Sorry, I don't know of any automated version extenders. I love your example of the version number though pi=3.14159265. You have a Physics or Engineering background I presume.

Dave R.
I actually have neither. I just watch MythBusters too much.

As for everyone else, there is a new minor feature added to this EA and for all future EAs I put in. It's the ability to trade with ECN-style brokers who require a trade and the stoploss/takeprofit to be added separately.

I have also added a feature to close all trades by a certain time of the week (like friday?).


Days = "Sunday is 0, Monday 1, Friday 5, etc";
UseCloseOnWeekEnd = False; If set to true, the EA will close all trades and not open new ones after the time specified below.
ClosePositionsDay = 5; This is the day of the week that the EA will close all trades (0, Sunday, 1, Monday, 2 tuesday, 3 wednesday, 4 thursday, 5 friday
ClosePositionTime = "23:00"; The time on the day specified above. Make sure to use "HH:MM" format.
Attached File(s)
File Type: mq4 4H Weekly Box Breakout.mq4   18 KB | 1,288 downloads
  • #2,095
  • Aug 19, 2009 5:30pm Aug 19, 2009 5:30pm
[quote=Ronald Raygun;2969660]I actually have neither. I just watch MythBusters too much.

I have now run RayGun’s EA on good live historical data. I know there are some people on this forum who are adamant that we don’t touch anything. The “if it ain’t broke don’t fix it” attitude. Fortunately, I have been an anti-rule person all my life and have done well because of it.

If anything is really good, it deserves to be poked and prodded to ensure it can stand up to inspection. If weakness’s are found, they should be corrected.

If the world took the “ain’t broke don’t fix it” attitude we would still be driving model T’s and flying to Europe in DC 3’s. If anything really is perfect, it will remain that way after the prodding and poking. If not it should change.

My backtest data only goes back to May 1 of this year so this is not statistically significant but the results are extremely interesting.

In Raygun’s EA, there are some differences to the original method that need to be explained:

 

  1. Raygun allows additional entries during the week if the market goes back through the “Box” during the week.
  2. The EA does not close automatically on Friday. I am not sure if I like this and I hope in a future EA this option will be added.
  3. I have chosen to back test with no TP and simply use a Trailing Stop to close my position.
  4. The option to re-enter a trade after a Trailing stop exit HAS NOT been used as it is not coded yet.

I have attached the results of a backtest using 2% risk which shows a 77% return in 3.5 months.
My next post will show the results of an aggressive 5% risk.
Dave R.

Attached Image (click to enlarge)
Click to Enlarge Name: StrategyTester 4HR-2.gif Size: 8 KB
  • #2,096
  • Aug 19, 2009 5:44pm Aug 19, 2009 5:44pm
Further to my last post, I am attaching here the results of the 5% risk backtest.

Just a comment on risk/reward. Mathematics. If we are risking 100 pips and our goal is 400 each time, our risk/reward is 1:4. In other words, if we Lost four times in a row and then won 1 time, we would break even. 4 losses at 100 pips versus one win at 400 pips.

If we can move the stop loss to BE quickly such that on average every four trades has two trades that move to BE, then our risk is effectively cut in half or 0.5:4 or 1:8.

We can now lose or BE 8 times in a row, then win once with no money lost.

OK on to the back test. Using 5% risk but moving to BE in 90 pips, we have a 370% return from May 1 to Aug 12.

I have included a zip file of all the results and settings for reference.

All criticisms welcome. If anyone has got 90% quality data going back several years, backtesting would be very welcome.

Dave R.
Attached Image (click to enlarge)
Click to Enlarge Name: StrategyTester 4HR-5.gif Size: 8 KB
Attached File(s)
File Type: zip BackTest Data 4HR.zip   31 KB | 649 downloads
  • #2,097
  • Aug 19, 2009 5:46pm Aug 19, 2009 5:46pm
Quoting Ronald Raygun
I actually have neither. I just watch MythBusters too much.

As for everyone else, there is a new minor feature added to this EA and for all future EAs I put in. It's the ability to trade with ECN-style brokers who require a trade and the stoploss/takeprofit to be added separately.

I have also added a feature to close all trades by a certain time of the week (like friday?).


Days = "Sunday is 0, Monday 1, Friday 5, etc";
UseCloseOnWeekEnd = False; If set to true, the EA will close all trades and not open new ones after the time specified...
That's excellent Raygun. I will do some more backtesting this evening with the Friday close thing.

Dave R.
  • #2,098
  • Aug 19, 2009 7:11pm Aug 19, 2009 7:11pm
Dave,

I noticed your test is on 1 hour timeframe (EA attached to 1 hr). Should it be attached to 4 hr timeframe as ACE's system?

Thanks.


Quoting droland
Further to my last post, I am attaching here the results of the 5% risk backtest.

Just a comment on risk/reward....
  • #2,099
  • Aug 19, 2009 7:24pm Aug 19, 2009 7:24pm
Quoting EATrader
Dave,

I noticed your test is on 1 hour timeframe (EA attached to 1 hr). Should it be attached to 4 hr timeframe as ACE's system?

Thanks.
Rayguns EA allows you to use the 1 HR timeframe but boxing over the same 4 hour period. The reason for this is that different brokers start their trades at different times. Therefore using the 4HR charts, the first bar, may be starting at 21:00 on one broker or 22:00 on another broker. Yourboxes therefore will be different.

By using the 1 hour chart, I look back over four - one hour bars instead of 1 - 4 hour bar and can therefore match Aces setup. I have always doubled checked my box with the ones published by Ace and they are within a few pips of being identical.

I have attached my setup on a 1 hr chart as reference.

Dave R.
Attached Image (click to enlarge)
Click to Enlarge Name: 1 hr chart.gif Size: 37 KB
  • #2,100
  • Aug 19, 2009 7:44pm Aug 19, 2009 7:44pm
Quoting droland
Further to my last post, I am attaching here the results of the 5% risk backtest.

Just a comment on risk/reward....

VERY, VERY impressive! Good work RR and Dave!
8 traders viewing now