I wasn't going to work anymore on the single grid EA, but in light of my own experiences with hitting max orders broker limits, and others similar experiences, I have decided to enhance the risk-mitigation strategies by adding additional ways (e.g. max open orders) to trigger both the high-water and low-water marks.
Again, the low water mark, once activated will behave the same way (reduce profit target, always still end in profit or breakeven). I have just added additional ways to trigger that state.
For those who don't use the risk mitigation...
Ignored
Viper, could you change the profit calculation for opening new positions or add it as an option in v2.5 so that it calculates the tp target on the basis of ((levels*increment)+increment)). This way using the extend take profit will do what it was originally intended to do.
Also is it possible to solve the problem of spike spreads closing sequences in loss by only permitting the ea to close the trades if the bid price reaches our tp target only. Someone mentioned earlier about using stealth mode. I thought perhaps if this were only used for the tp/sl target on the buy side it may solve this particular problem?
Lots of repsect Viper. You've turned a caterpillar into a butterfly.
Firstly thanks very much for the huge amount of work you have put in.
I don't post much but perhaps one way of showing my thanks is to show my forward testing results. Mostly with the default settings.
Testing started on the 19 Oct and in less than one month I am up 25% with no real dramas. When going live I will be more cautious but I believe with a sensible approach your ea has real potential.
I wasn't going to work anymore on the single grid EA, but in light of my own experiences with hitting max orders broker limits, and others similar experiences, I have decided to enhance the risk-mitigation strategies by adding additional ways (e.g. max open orders) to trigger both the high-water and low-water marks.
Again, the low water mark, once activated will behave the same way (reduce profit target, always still end in profit or breakeven). I have just added additional ways to trigger that state.
I was demo testing this last night and had left it open for multiple pairs. I noticed that when one pair (E/U) hit the TP, all others were automatically closed. Am I supposed to make any change to the magic number in each pair?
Secondly, I want the initial orders at the difference of 10 pips as opposed to 20 pips default settings and then leave the increment as is.
For example in EURO/USD say the current price is at 1.4900. I want buy order placed at 1.4905, 1.4915, 1.4925 and sell orders at 1.4895, 1.4885 and 1.4875.
You must set a different magic number for each pair you trade to avoid closing out on every trade once TP is reached.
Also, for your second issue. Set the increment to 10 (or 100 for a 5 digit broker), then change the 'incrementadjustforfirstlevel' to -5 (or -50 for 5 digit).
That should solve your problems.
Luck is where preparation meets opportunity.......
You must set a different magic number for each pair you trade to avoid closing out on every trade once TP is reached.
Also, for your second issue. Set the increment to 10 (or 100 for a 5 digit broker), then change the 'incrementadjustforfirstlevel' to -5 (or -50 for 5 digit).
That should solve your problems.
Ignored
Thanks mate. I should test the new settings tonight from home. Can't access MT4 from work.
BTW, have you tried this EA in live? if yes, would you please share your experience?
Once I get the EA up and running, I intend to post my demo results on a regular basis to help test.
To be 100% honest, i haven't tried it live yet. Have forward tested but my results weren't terrific and im refusing to go live till im confident i can make a steady income without destroying the account.
When i traded options a couple years back, i was burnt bad by the beginning of the financial crisis and lost ALOT of money (well for me it was). Have learnt alot since then and much more cautious now (its both a good and a bad thing).
I still have confidence with this EA and the people working on it. As stated before, i have spent 100's of hours with this trying all sorts of different risk mitigation strategies but have been unsuccesful (as a whole). Im trying to make this a 'set and forget' EA.
My belief:
If you can find a way to dig yourself out of a massive DD without incuring a huge loss and giving away all your profits then there is no limits to what you can achieve with this EA.
Is it possible?? Probably not but i can't seem to give it up or stop trying. Once i find something i will let people know. If not, then i will keep trying to answer questions and get ideas off others which could hopefully leed to something substantial.
Luck is where preparation meets opportunity.......
To be 100% honest, i haven't tried it live yet. Have forward tested but my results weren't terrific and im refusing to go live till im confident i can make a steady income without destroying the account.
When i traded options a couple years back, i was burnt bad by the beginning of the financial crisis and lost ALOT of money (well for me it was). Have learnt alot since then and much more cautious now (its both a good and a bad thing).
I still have confidence with this EA and the people working on it. As stated before, i have spent 100's of...
Ignored
Thanks for your detailed reply mate. Well, my story (in terms of loss) is very similar. I've lost my one month's gross salary in Forex in last 3 months and when you combine that with my loss in shares/cfd that's way too much and I seriously can't afford it.
I was with a broker who didn't allow hedging in the same currency pair, which was the reason for my loss. I've sinced changed broker and now on MT4 and hence this interest in EA, etc
At this point I intend to forward test this EA (Can you please show me how to backtest the same??) in my demo account only when I am infront of it so that I can see what its actually doing and without being greedy if it can give me some $25-$50 a day, I am happy.
thanks for your help as usual. I must say you guys have been great and have put in a lot of efforts in this EA.
My sincere thanks to all people who have devoted considerable amount of time and effort in this EA.
Is it possible in VipergridEA_v4 EA that when the EA is running and your net positions is in profit by x dollars you can manually close those positions and shut the EA?
I am asking this because when I am infront of the PC and this EA is running and say If I see a profit of $20 I say that's my limit and choose to close it down?
Is that possible? If yes, closing one position manually will cause all the pending orders to cancel?
Is it possible in VipergridEA_v4 EA that when the EA is running and your net positions is in profit by x dollars you can manually close those positions and shut the EA?
I am asking this because when I am infront of the PC and this EA is running and say If I see a profit of $20 I say that's my limit and choose to close it down?
Is that possible? If yes, closing one position manually will cause all the pending orders to cancel?
if yes, will it affect its future operations?
Thanks a lot.
cheers
Ignored
Puks,
If you run the EA visually on a backtest, you will see that profit usually isn't made until the last few pips of the final increment. If price goes straight towards TP then profit will happen earlier, of course.
What you are asking is similar to the "high-watermark". It closing some trades that are slightly in profit in order to save margin. You have to specify how deep the drawdown has to be before this is implemented.
I honestly don't think backtesting this EA is all that useful as you can get vastly different results depending on the day and time you start testing at. However, one consistent way to get meaningful results is to run one grid at a specific time each day. I have done a few backtests using M1 data from August 1st, 2009.
The grids are all looking to capture 70 pip moves on the GU starting a little before the london open. You will see the results compare a 1-level grid up to a 9-level grid, with every grid equalling 140 pips. All tests had the TP extended by 10 pips (final level was increment + 10 pips). I used .1 initial lots on a 10k account, which is risky. One should really stick to .01 lots per 10k equity.
I have labelled the backtests in the zip file as follows: eg. "GU3L15i5am" means GU pair, 3 levels, 15pip increment, grid starts at 5AM GMT
Enjoy! Please feel free to add comments or suggestions!
Puks,
If you run the EA visually on a backtest, you will see that profit usually isn't made until the last few pips of the final increment. If price goes straight towards TP then profit will happen earlier, of course.
What you are asking is similar to the "high-watermark". It closing some trades that are slightly in profit in order to save margin. You have to specify how deep the drawdown has to be before this is implemented.
Hope this helps. Keep the ideas coming though!
Ignored
Thanks Jay. Can you please advise the potential probelm associated with below settings?
Lot size = 0.01, Equity in account = $600
Levels = 3
Increment = 5 pips (4 digit broker)
Pair - euro/usd
Setup would be like LIne - 1.4900. B1 = 1.4905, B2 = 1.4910, B3=1.4915, TP = 1.4920 and same levels starting from 1.4995 on short side.
Now there is a "decent probability" that the market (at London open and Ny open) will travel atleast 25-30 pips in each direction.
so If I start this EA "just" at these times and after making profit, I can turn it off, do you think that's a good strategy?
sorry I haven't been able to test this as I am still getting my head around with settings but I intend to do some rigeous testing tonight (sydney).
Thanks Plasma. I suppose I can do both. I can fix the temporary spread spike closing the set down with a loss with a stealth mode type algorithm. But if the spread is extra wide at the time when the TP is hit, profits will suffer (maybe even cause a loss) and there is nothing I can do about that of course. I'm not a genie :-)
Of course your request is going to cause a delay while I implement and test...
I was demo testing this last night and had left it open for multiple pairs....
Ignored
Puks, You only need to change the magic number if they are running on the same pair. If they are on different pairs you do not need different magic numbers.
Also note, if the increment is set to 20, the difference between the first buy level and the first sell level is going to be 40, not 20. The increment begins at the pivot and adds at each level.
Such small increments are NOT going to work well if your spread is close to that size or you have volatile spreads with your broker. Please be careful.
Firstly thanks very much for the huge amount of work you have put in.
I don't post much but perhaps one way of showing my thanks is to show my forward testing results. Mostly with the default settings.
Testing started on the 19 Oct and in less than one month I am up 25% with no real dramas. When going live I will be more cautious but I believe with a sensible approach your ea has real potential.
Ignored
That's great Walt. Just please understand the risks prior to going live. I would hate to have you report in a few months that you lost half your account. We are working hard to lessen the probability but I'm afraid it will always be possible.
You must set a different magic number for each pair you trade to avoid closing out on every trade once TP is reached.
Also, for your second issue. Set the increment to 10 (or 100 for a 5 digit broker), then change the 'incrementadjustforfirstlevel' to -5 (or -50 for 5 digit).
That should solve your problems.
Ignored
You have been a big help, with your contributions and your assistance with other's questions.
I do have one correction though, I believe I fixed the 5-digit issue so you would simply enter -5 whether you use a 4-digit broker or a 5-digit broker. I have not actually tested this EA on a 5-digit broker, but I have my live account on a 5-digit broker. When I started with them, nothing I had written and tested worked because of this issue. So I had to learn to fix it. I fixed this in the same manner I had fixed those - I think it's correct.
Puks, You only need to change the magic number if they are running on the same pair. If they are on different pairs you do not need different magic numbers.
Also note, if the increment is set to 20, the difference between the first buy level and the first sell level is going to be 40, not 20. The increment begins at the pivot and adds at each level.
Such small increments are NOT going to work well if your spread is close to that size or you have volatile spreads with your broker. Please be careful.
Viper
Ignored
Thanks Viper. I will be careful.
With my broker the spread is 2 in euro/usd so here is what I intend to test. Its the difference between B1 and S1 that I want to lessen.
I will try heap of different combinations and will put my results here.
I will start with b1 at 1.4905, b2 at 1.4915 and b3 at 1.4925 with TP at 1.4935 and in short side, s1 at 1.4895 and so on.
Is it possible in VipergridEA_v4 EA that when the EA is running and your net positions is in profit by x dollars you can manually close those positions and shut the EA?
I am asking this because when I am infront of the PC and this EA is running and say If I see a profit of $20 I say that's my limit and choose to close it down?
Is that possible? If yes, closing one position manually will cause all the pending orders to cancel?
if yes, will it affect its future operations?
Thanks a lot.
cheers
Ignored
Yes closing one order will close the set. No it doesn't effect future operations. It will behave as though the set closed naturally. You can also run a script I use sometimes to shut everything down. I turn off the EA (by hitting the Expert Advisors button so it doesn't start a new set) and run the attached script.
You have been a big help, with your contributions and your assistance with other's questions.
I do have one correction though, I believe I fixed the 5-digit issue so you would simply enter -5 whether you use a 4-digit broker or a 5-digit broker. I have not actually tested this EA on a 5-digit broker, but I have my live account on a 5-digit broker. When I started with them, nothing I had written and tested worked because of this issue. So I had to learn to fix it. I fixed this in the same manner I had fixed those - I think it's correct.
I'll...
Ignored
Thanks mate, very much appreciated.
By the way, you are 100% correct. Wether it is a 4/5 digit broker, ignore the extra '0'. Its not needed.
Your work on this EA has been fantastic. Thanks for all your effort.
Luck is where preparation meets opportunity.......
Puks, there are entry filters that include time of day. You can set these to session open times and set MaxSetEntryCount=1. Mind you, all times are broker time, not GMT or local time. So if London open starts at 4AM broker time, just set the MinHourToStart=4, MaxHourToStart>=4 and MaxSetEntryCount=1. This will wait until that time to start the set and will only allow one set to execute. Then you would need to edit it for NY open.
