Blessing 2 v5.2 with Stochastic Indicator - page 30
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
If someone wants to play around with the 1min settings I have taken it this far in a 2 week backtest of EUR/CHF when I last backtested it it did this:
Net: $624
Gross: $825
Max DD" 4.27%
Not too bad so far...
Alot of changes were made.
I have not forward tested this one yet, still working on the 5 min but if someone wants to test or tweak it here ya go ~
Please share your data.
SAME EA just different name from my testing.
Fifth Element did a fine piece of work IMO.
ChartIf you change only the "Float Percent" to 15 instead of 5, then you can run these 1 min settings on the 5 min chart EUR/CHF.
Report and graph below.
Chart
nice improvements. some time ago i had managed to come up with an m5 set file and i was very happy, forward tested it for 3 months, it had gone successfully through all backtests already but i crashed aftere 2-3 live months. so i think its a dangerous thing to use the ea on m1 and m5 but as i see you amde some changes. thats very good but can you post the exact EA for m5 and the settings as a .set file, that would be really great, bacause i am interested to test the settings too
thank you
nice improvements. some time ago i had managed to come up with an m5 set file and i was very happy, forward tested it for 3 months, it had gone successfully through all backtests already but i crashed aftere 2-3 live months. so i think its a dangerous thing to use the ea on m1 and m5 but as i see you amde some changes. thats very good but can you post the exact EA for m5 and the settings as a .set file, that would be really great, bacause i am interested to test the settings too
You are VERY correct feline about 1 min systems and how "drifty" they can be. Even the 5 min, but I have come very close in 5 min charts before. I also observed that higher timeframes just took longer for, what I call "drift" to set in. I had a huge problem with VTTrader's platform and how candlesticks "repainted" themselves after you signed back in. You would be bought in but the BUY signal was no where to be found, it was just gone. I'm very hopeful that MT4's platform will be much more stable.
With the Blessing EA I think there might be a possibility in having a productive LONG TERM EA. You see if you have a EA that has a pattern that trades it well, then you wait for the GOLDEN spots to suface, or what other people call "crashes." These spots are what you have to overcome in your settings. You have to have a "sharp" EA and yet you also have to have the right balance or general setting. If it's "too sharp" it will trade very well at first but crash. Then you test the heck out of those troubled area's to find a more general setting for the EA. This is where most traders stop in frustration, but the market is actually showing you where your weakness is and what needs to be fixed. That's why I call these area's GOLDEN because that is what they are worth once you figure them out
And the 5 min settings I'm still working on, although if you take that 1 min setting I posted and simply change the FLOAT PERCENT from 5 to 15 and you will have a pretty good setting for the EUR/CHF 5min chart to start with.
All of these are at their beginning stages in development. It will take some forward testing and some "Golden" areas so we can shift in a good general setting for the long term. We'll give up some pips but it will be safer and well worth it.
I do want to have 3 EA's up and running but 2010's close. I think blessing will be one of them, I'd be very surprised if it doesn't pan out. We'll get to know it through back-testing and get the safest, sharpest setting we can. Then forward test it and "dull it out some" as we hit those golden trouble spots, then we should only have to deside whether we want leather or cloth interiors in our new cars...
Happy testing!
Chart ~
chartwaker it is obvious that youre a smart man like all of us here and you know what youre doing and what is that you exactly want. youre absolutely right about everything you say. and theres only one thing that i am worried about -> backtests are not 100% accurate. i also relied on them very much few months ago but they failed me. ill give you the simplest and most recent example. The stoch version of blessing (stoch-only) seems to be very productive and i made literally hundreds of backtests for short and long periods of time , different settings also and sometimes i ran optimizations. it showed a maximum of 800 dd for entire year 2009 (0.01 startingLOTS) and if you ran a backtest for the previous 3 years including 2006 it would say like 2.6-2.8k maximum drawdown. and it was also very profitable, good profit factor and everything. i started running it , to that moment i hadnt encountered a dramatic dd with blessing and thought of it almost as of the so talked-about HOLY GRAIL in the forex eas. and guess what happened. i started trading with it and it performed very very good for about 1.5 months, after that it ran into a DD that went up maybe to 3.0-3.5k at leasst. so it beated the backtests results from the last 36 months. so this is the only thing that i really want you to know if you dont already know that. we should forward test the settings first, and i mean for a long period of time. it really sucks that you have to waste additional time and you cant skip over it.. thats why at least for know i still use the original april 09 setfiles but only on stoch version. i dont think i have any additional mods and i also try to put the biggest ammount of $ i can afford right now, so i can be if not 100, thenb at least 95% sure it will survive. thank you very much for your work and we will continue sharing thoughts and ideas, i am sure of that
but we have to be very very careful.
p.s. so youre makin those backtests on mt4 or what ? because i use only mt4 and some experience people say its not good for backtesting. i dont know..i just dont have the knowledge to be sure about those things
have a nice day 
A backtest of Mod4 on the m5 chart using the M5 settings survived both Oct 08, and a bad spot this summer between June and July. I've got it forward testing now, but the backtests show promise if it can dodge that bullet. Not to mention that the test was run with the minimum 5k allowed for the EC pair. More agreesive settings with a higher deposit managed to plow through it as well, but ultimately if this thing can survive starting off 10/15/08 with these settings, I'd say it's well on it's way.
Awesome work mate.
Hi Board ~
Feline, thanks for the kinds words. I think the backtest should only be used for finding a starting point, the "fine tuning" we do as it goes forward and the troublesome (Golden) spots appear and then we'll adjust the settings to lessen the those bumps bumps...
Yes no EA is perfect but we sure should be able to get a good general running EA IMO.
Here's my process of what I'm thinking of doing. [what say you?]
1) Backtest for a good starting setting and getting to "know" the EA
2) Forward test and wait for those golden moments when the market shows you where your weaknesses are and adjust and make the ride as smooth as I can.
3) When you think you are getting close to where you want to be, open 3-5 demo accounts with DIFFERENT brokers so you KNOW that the broker you are useing doesn't suck and make your work SEEM for nothing, and watch things for at least a month.
4) Fund the best one and rock & roll ~
We'll get there
Ganapati, thanks for the rundown on the different versions, I have yet to play around with them and noted your comments, -thank you. I like the fact that I happen to be in the conservative one, but I bet the aggressive one would be best for the 1 min charts. I still find it hard to believe a 1 min chart could be reliable, but yet there MUST be a way huh?... But the 5 min I do like.
Let's see what the forward tests show us.
Let's keep testing and talking, the more we do the faster we can get there. And if I can help in anyway please just ask. And please keep in mind we are only at the starting point and NEED some blunder points to show us what needs attention ~
So I don't know if I would use "real bullets" just yet...
I'm also working on a way to "keep" the settings top notch. What does everyone think of this?
Take 6 months of data, as your benckmark backtest, then in 2 weeks we will have 2 NEW weeks of data available as the oldest 2 weeks drop off the 6 month backtest. Run some scans and note what settings show improvements, note if the profit graph "drifts" as the new data replaces the old. If it does, make the adjustments and smooth out the ride again. I would think once we know what to watch this will be easy to do. Most dirfts start about 4-6 weeks after running, hence the idea to start watching the scans every two weeks and make notes on what may need tweaking on the fly.
Happy trading peeps!
Chart ~
A backtest of Mod4 on the m5 chart using the M5 settings survived both Oct 08, and a bad spot this summer between June and July. I've got it forward testing now, but the backtests show promise if it can dodge that bullet. Not to mention that the test was run with the minimum 5k allowed for the EC pair. More agreesive settings with a higher deposit managed to plow through it as well, but ultimately if this thing can survive starting off 10/15/08 with these settings, I'd say it's well on it's way. Awesome work mate.
wow that made my day Lev!
I didn't expect that so early in the development of this, thanks for sharing that. I'm starting to get excited but trying not to. ha ha ~
Chart
i think we should optimize the parameters on a shorter period and then we should maybe run a longer test. i think september this year is perfect for testing because of the strong and consistent upward movement that we witnessed during that month. i have to test bb versions more thoroughly and will come back with some results
good luck my friends
what is the difference between Mod3, mod4 and 5.2 ?
I am running MOD3 without stoch is this identical to 5.2 ?
I have been trying to back test EURGBP on 4h and 1H on FXDD ..both do well and the crash in August. Is this a glitch ?