What is your Brokers Time Zone? At first glance it does not seem to be using the correct day open hour or the time zone is not UTC+2.
However, I will redo my back-test using the same tick data and time period and repost it here for a better comparison.
Just give me a couple of days as today I will be quite busy and will not be able to do it today.
[B]EDIT: I have just taken a quick look at EURUSD Tick data and can confirm that either your Time Zone is not UTC+2 or your tick data is not properly shifted. If it is a case of different Time Zone, then you will have to open at a different time and calculate the High/Low differently.[/B]
I had another three losses last night. Although I realise I missed a couple of days earlier in the month (being away), and i donāt know what results those two had, my ongoing figures for the month are now negative (-5 pips).
Does this tie up with other peopleās results?
Also, Vijay, how are you getting on this month? Not heard from you for a while.
Double winner today on the GBP/USD, just got stopped out with 38 pip profit; +5 on position one (fixed 5 pip TP/SL) and +33 on position two (fixed 5 pip SL & 10 pip TS).
My lot size is not all that important. Suffice to say that I am a very conservative trader and rarely risk more then 1% of my account and never more then 2% on any given trade. Todayās trade was 1% and what Iāve been doing lately is a 3 to 1 ratio; 0.75 / 0.25.
After reviewing Vijayās myfxbook page I concluded that position 1 hits more often then 2 and 3. So, if I win on pos 1 and lose on pos 2, then Iām still in the money. One could argue to increase pos 2, but it seems to me that you donāt often get runners like this morning before a retrace and your TS gets taken out.
Initially I was using a 5-6 pip TS, but after Mr. C posted his test results with a 10 pip TSL I changed my strategy; time will tell how well it works. Once he develops his EA I will most likely abandon my method and go for his. Iāve asked Vijay about the one heās developing, but he has pulled it from myfxbook and seems reluctant to reply about it both here and via email.
Forgot to add something else on money management and for some reason the forum wonāt let me back in now to edit.
If I happen to be online when the trade activates, then as soon as itās up by about 7-8 pips I move my SL on pos 2 to break even (+/- 1 pip depending on spread). For me preserving capital is important. Perhaps moving the SL to BE is something for Vijay and Mr. C to consider incorporating into their EAās, but at what point? Opinions anyone? And I know they already have a trail gap, but as Iām not a programmer I donāt know how that works; perhaps what Iām suggesting is one and the same.
Iām not sure about what Time Zone I used for this backtest, I think itās UTC +2 OR +3 but I will confirm it and let you know. What time zone you used ? +2 ?
Iām java developer, so Iām still learning MQL but Iām getting some fun, itās interesting.
Did you try to test it live and compare the results with your backtests ?
The [B]Time Zone in use is Very Important[/B] for [I]Vijayās[/I] strategy (and my own adaptation) to work properly!
If Pending Orders are placed at 00h00, then your broker must be UTC+2. If it is not, then you have to offset the open as well as calculate the correct highs & lows based on the NY Close. In other words, you cannot simply use the D1 High/Low values.
One way to compensate for different time zones in an EA is to calculate The Highest High and Lowest Low of a Period of 24 Bars of H1 data which is offset or shifted by a certain number of hours that coincide with the NY Close.
Your EA is obviously not doing that, so if your Broker (and the tick data) does not coincide with UTC+2, then your code will not work correctly as it stands.
Sorry, I should have known better and said āif not already thereā. But maybe for Vijayās original EA it may not be a bad idea for his EA to move pos 2 and 3 to BE once pos 1 is hit (plus spread).
His original strategy has no trailing (or break-even) at all. That is why my adaptation does, but it seems that he has not yet fixed his EA to work properly with the trailing stop.
I donāt know if he is very busy or not these days, but lately, he has not participated much, nor offered any feedback on the matter.
Hi Carnino, I know Vijay is testing the strategies live but, did you try it ? in demo account.
I did backtest again and now Iām sure itās UTC +2, anyway the results was not profitableā¦ as I told you I fixed the bugs and now itās trailing with step of 5, sl 5 and I set TP but far away, 100 pips. Also the expiration of orders to 10 days.
I will post the last backtest soon.
Edited to post the backtest, look BreakoutX2.zip (35.4 KB)
No, the bugs are not all fixed. If you really want to analyse the problem, then read your own report. If you look at the āBreakoutX1ā files you will see that the daily highs and lows do not match that of a UTC+2 Daily Chart, nor to the buy and sell prices match tick data for UTC+2.
For example, on the 2013-10-02, according [I]Dukascopy Tick Data[/I], the Day High was 1.36062 and the Day Low was 1.35042, yet you place your the pending orders on the open of the next day with a buy at 1.36179 and a sell at 1.35379. Even with a pip or two of offset, those prices do not coincide with the data. I also verified the high/low using [I]HistData[/I] and data from my own broker, [I]RoboForex[/I] (all had similar values to [I]DukasCopy[/I] data).
You yourself have said that you are a Java programmer, only now starting to code in MQL4. So donāt you think that you should first consider that your EA might be wrong and still have bugs?
Besides [I]Vijayās[/I] testing, two other separate users (āMata Nuiā and myself), using different EAās have confirmed that it is indeed profitable. That is not to say that any of us are infallible, and obviously there is plenty of room for improvement, but first, take the time to make sure that it is not your EA at fault before stating outright that it is not profitable.
[B]EDIT: This was obviously posted by me before you placed the X2 version up. I will have a look it and comment later![/B]
Hi Iām sorryā¦ I didnāt want to be negative, and Iām absolutely not saying that strategy doesnāt work.
You right, my EA can still have bugsā¦ so Iām asking about it.
Can you please check the file BreakoutX2 in the previous post ?
After looking at your X2 version, I am now even more convinced that either your EA or your tick data is faulty.
For example, your X1 and X2 are both for the same testing period, yet Highs and Lows do not coincide between the two. To make it worse, X1 has pending orders starting on 2013-01-02 and X2 only starts having pending orders on 2013-01-08.
Sorry, āfelipebrā, but you are going to have to take a serious look at your setup and fix the issues before continuing with the procedure.
