Great EA in backtest! - page 117
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
ì have test (ok for a short time) the 2.0 and the euro 2.2 version. in the 2.2 version are lesser trades than in the 2.0 version and it seems like the results are better. where is the problem why people say this one is a bad ea? do u trade live ur a (wich) version (?) aara?
I have outcome data..
this represents the combined values of 1344 trades
WIN LONG
Min 0.00048 0.00063429 0.00132 5 3 4 9 5 5 0.00044 0.00044444 0.00047778 0.00009143 0.00006857 0.0001
Ave 0.002546876 0.002534261 0.00812069 12.33262712 7.773305085 8.216101695 16.72033898 8.872881356 9.406779661 0.001307665 0.00120777 0.001089179 0.000474909 0.000277216 0.000262536
Max 0.01052857 0.00876571 0.02431429 19 13 13 24 13 13 0.00294286 0.0041 0.00296 0.00105714 0.00095714 0.00068857
LOSE LONG
Min 0.00031429 0.00077143 0.00113143 5 3 5 9 5 5 0.00051 0.000325 0.0004875 0.00009143 0.00005429 0.00011428
Ave 0.002850238 0.002860655 0.008342778 12.59375 7.930555556 8.291666667 16.65625 8.965277778 9.378472222 0.001322559 0.00132135 0.001234676 0.000489157 0.000307411 0.000297619
Max 0.00801429 0.00999143 0.02431429 19 13 12 25 13 13 0.00267857 0.003325 0.0031 0.00107714 0.00076 0.00076857
Thus I observe:
To WIN LONGS SellPossibilityMid*SellPossibilityQuality: BuyPossibilityMid*BuyPossibilityQuality: UndefinedPossibilityMid*UndefinedPossibilityQuality: UndefinedSucPossibilityQuality: SellSucPossibilityQuality: BuySucPossibilityQuality: UndefinedPossibilityQuality: SellPossibilityQuality: BuyPossibilityQuality: UndefinedSucPossibilityMid: SellSucPossibilityMid: BuySucPossibilityMid: UndefinedPossibilityMid: SellPossibilityMid: BuyPossibilityMid:
Min not less than .00048 exceptions noted not less than .00132 no contrast no contrast wider range no contrast no contrast no contrast lower not below .000443 lower no contrast not below .0000686 exceptions noted
Ave winners smaller average than losers smaller on average lower on average lower on average lower on average lower on average higher on average lower on average higher on average lower on average lower on average lower on average lower on average lower on average lower on average
Max exceptions noted not above .00876 no contrast no contrast no contrast wider range Not above 24 no contrast no contrast exceptions noted exceptions noted not above .00296 not above .001057 exceptions noted not above .000688
[/PHP]
[PHP]WIN SHORT
Min 0.00028571 0.00039429 0.00154 5 3 3 9 5 4 0.00058 0.0005 0.00044 0.00014 0.00005714 0.00006571
Ave 0.002659267 0.002756386 0.007791549 12.13118812 8.487623762 7.690594059 16.36633663 9.584158416 9.04950495 0.00129987 0.001087373 0.001322631 0.000463423 0.000270092 0.000296301
Max 0.00820286 0.00874286 0.01938857 19 12 12 24 13 13 0.00281818 0.0026 0.0049 0.00107714 0.00074571 0.00072857
LOSE SHORT
Min 0.00075429 0.00037143 0.00179143 6 5 3 8 6 5 0.00058 0.00055714 0.00058 0.00015143 0.00011143 0.00007429
Ave 0.002793779 0.002824976 0.008300714 12.36016949 8.533898305 7.538135593 16.6059322 9.427966102 8.966101695 0.001341504 0.001164867 0.001353249 0.000486768 0.000288983 0.000303535
Max 0.00911429 0.00843429 0.01748571 19 13 12 24 14 13 0.00240714 0.00263636 0.00315 0.00097143 0.00082857 0.00074857
Thus I observe:
To WIN SHORTS SellPossibilityMid*SellPossibilityQuality: BuyPossibilityMid*BuyPossibilityQuality: UndefinedPossibilityMid*UndefinedPossibilityQuality: UndefinedSucPossibilityQuality: SellSucPossibilityQuality: BuySucPossibilityQuality: UndefinedPossibilityQuality: SellPossibilityQuality: BuyPossibilityQuality: UndefinedSucPossibilityMid: SellSucPossibilityMid: BuySucPossibilityMid: UndefinedPossibilityMid: SellPossibilityMid: BuyPossibilityMid:
Min lower the better not below .000394 wider range lower the better lower the better no contrast not below 9 exceptions noted exceptions noted no contrast lower the better wider range wider range lower the better lower the better
Ave lower on average lower on average lower on average lower on average lower on average higher on average higher on average higher on average higher on average lower on average lower on average lower on average lower on average lower on average lower on average
Max not above .0082 exceptions noted wider range no contrast not above 12 no contrast no contrast not above 13 no contrast exceptions noted not above .0026 widr range wider range not above .00075 Not above .000729
it should make more sense to look at it in excel than pasted here. oh crap that was with the one order per bar setting true...that is going to mess it up...start over sherlock, Oh well at least I've got the analysis pages built in excel now...
it's in this... https://www.mql5.com/en/forum/174700/page71
what value i need to put in Max bars in history and max bars in chart ?
Keep at it Aaragorn!
Been following your progress, and I find myself coming back to this EA. Hopefully soon I'll be caught up and have some more input.
Thanks,
based on my findings I made this filter.
however the results are incredibly harsh. Using this it only makes about 36 trades a year and the proportion of drawdown isn't alot better. So the CT logic filtering may yet be useful but it's going to take alot more work to make it so.
I did find one error in my previous 'one order per bar' work. The error is that I never reset the switch to false after the order was blocked, so the switch remained in block mode..this probably accounts for some of the devestation that resulted in previous tests using it. It will be interesting to see if (how much) it changes now that I've repaird the glitch.
{
if(UseCTfilter)
{
if(Decision == DECISION_BUY)
{
if(SellPossibilityMid*SellPossibilityQuality < 0.0048)
{
CTBlockBuy = true;
}
if(BuyPossibilityMid*BuyPossibilityQuality > 0.00876)
{
CTBlockBuy = true;
}
if(UndefinedPossibilityMid*UndefinedPossibilityQuality < 0.00132)
{
CTBlockBuy = true;
}
if(UndefinedPossibilityQuality > 24)
{
CTBlockBuy = true;
}
if(SellSucPossibilityMid < 0.000443)
{
CTBlockBuy = true;
}
if(BuySucPossibilityMid > 0.00296)
{
CTBlockBuy = true;
}
if(SellPossibilityMid < 0.0000686)
{
CTBlockBuy = true;
}
if(BuyPossibilityMid > 0.000688)
{
CTBlockBuy = true;
}
}
else
{
if(Decision == DECISION_SELL)
{
if(SellPossibilityMid*SellPossibilityQuality > 0.0082)
{
CTBlockSell = true;
}
if(UndefinedPossibilityQuality < 9)
{
CTBlockSell = true;
}
if(SellPossibilityQuality > 13)
{
CTBlockSell = true;
}
if(SellPossibilityMid > 0.00075)
{
CTBlockSell = true;
}
if(BuyPossibilityMid > 0.000729)
{
CTBlockSell = true;
}
return(0);
}//if DS
}//DB else
}//if use ct filter
}//ct filterit's in this... https://www.mql5.com/en/forum/174700/page71
hi
it is good not
pls send setting file
tnx bye
1.93 Euro - now with CT filter option
I give you the CT filter.
This is using the results of the data dump I did of 2318 trades.
WIN LONG
Min 0.00048 0.00021715 0.00132 5 3 2 9 5 4 0.00044 0.00044444 0.00047778 0.00009143 0.00006857 0.00005429
Ave 0.002420114 0.002442273 0.007918565 12.20187793 7.541079812 8.151408451 16.90023474 8.73943662 9.360328638 0.001270039 0.001214834 0.001060205 0.000458102 0.00026831 0.00025383
Max 0.01052857 0.00876571 0.02431429 19 13 13 26 13 13 0.00294286 0.0041 0.00296 0.00113714 0.00095714 0.00068857
LOSE LONG
Min 0.00031429 0.00057143 0.00113143 5 3 5 9 5 5 0.00051 0.000325 0.0004875 0.00009143 0.00005429 0.00009714
Ave 0.002732355 0.002717004 0.007946232 12.51891892 7.813513514 8.310810811 16.68378378 8.927027027 9.389189189 0.001265285 0.001296737 0.001158426 0.00046529 0.000296386 0.000281166
Max 0.00801429 0.00999143 0.02431429 19 13 13 25 13 13 0.00267857 0.003325 0.0031 0.00107714 0.00076 0.00076857
[/PHP]
from these findings I concluded..
To WIN LONGS not less than .00048 not above .00876 not less than .00132 Not above 24 not below .000443 not above .00296 not below .0000686 not above .000688
[/PHP]
and...
[PHP]WIN SHORT SellPossibilityMid*SellPossibilityQuality: BuyPossibilityMid*BuyPossibilityQuality: UndefinedPossibilityMid*UndefinedPossibilityQuality: UndefinedSucPossibilityQuality: SellSucPossibilityQuality: BuySucPossibilityQuality: UndefinedPossibilityQuality: SellPossibilityQuality: BuyPossibilityQuality: UndefinedSucPossibilityMid: SellSucPossibilityMid: BuySucPossibilityMid: UndefinedPossibilityMid: SellPossibilityMid: BuyPossibilityMid:
Min 0.00028571 0.00031429 0.00135143 5 3 2 9 4 4 0.00058 0.0005 0.00044 0.00012286 0.00005714 0.00006286
Ave 0.002465013 0.002474476 0.007647709 12.03196931 8.193094629 7.460358056 16.87212276 9.361892583 8.765984655 0.001246447 0.001062041 0.001254263 0.000442934 0.000255228 0.000272806
Max 0.00820286 0.00874286 0.02388 20 13 12 26 13 13 0.00281818 0.00285714 0.0049 0.00113714 0.00074571 0.00072857
LOSE SHORT
Min 0.00056571 0.00037143 0.00135143 6 4 2 8 6 5 0.00058 0.0005375 0.00054286 0.00012286 0.00009429 0.00007429
Ave 0.002724241 0.002690219 0.007808735 12.02229299 8.445859873 7.52866242 16.56369427 9.461783439 8.974522293 0.001289283 0.001138241 0.00130382 0.000458153 0.000280255 0.000288626
Max 0.00911429 0.00843429 0.01748571 19 13 12 24 14 13 0.00253 0.00263636 0.003275 0.00097143 0.00082857 0.00074857
from these findings I concluded...
[PHP]concluded SellPossibilityMid*SellPossibilityQuality: BuyPossibilityMid*BuyPossibilityQuality: UndefinedPossibilityMid*UndefinedPossibilityQuality: UndefinedSucPossibilityQuality: SellSucPossibilityQuality: BuySucPossibilityQuality: UndefinedPossibilityQuality: SellPossibilityQuality: BuyPossibilityQuality: UndefinedSucPossibilityMid: SellSucPossibilityMid: BuySucPossibilityMid: UndefinedPossibilityMid: SellPossibilityMid: BuyPossibilityMid:
To WIN SHORTS not above .0082 not below 9 not above 13 not above .00075 Not above .000729
the result of which is less profit but also less drawdown. I believe this would be safer for small accounts like mine. Perhaps it might even make actual profit in live forward test? That is the question. All that great looking backtest stuff doesn't mean a thing if it doesn't perform live forward. With any luck this will be an improvement? At very least it gives a few more parameters to change to see how it effects it.
the settings I used are in the EA. As always this is exclusively for the EURUSD on the 1 hour chart. (Unless you tailor its settings for something else.)
I'm starting to get the feeling that I've learned about all I can from this EA (perhaps not all I can learn from Dave about how to use it better, I'm not getting as good as his results live) that I've done about all I can do with it, and it's time to move on. I have another EA I'm starting to focus on now. As I move on to that project I'm taking with me the dynamic traling stop the CT uses, I kinda like that feature. I'm still wanting to develop a system that uses a higher probability trade and increased leverage.
Thanks Aaragorn, gonna forward test this one for a week, will try and post something.
Nanoc,
Just download the file and use the default settings.
There is no settings file atm.
Also, feel free to read the entire thread. I know I did.
* EDIT *
I also notice with the filter there is no long trades, or at least non profitable. gotta check that again, I can't believe it
Thanks Aaragorn, gonna forward test this one for a week, will try and post something.
Nanoc,
Just download the file and use the default settings.
There is no settings file atm.
Also, feel free to read the entire thread. I know I did.
* EDIT *
I also notice with the filter there is no long trades, or at least non profitable. gotta check that again, I can't believe itgreat scott you're right! that can't be good....I can't believe that is accurate. I'm gonna have to look what it's doing.