Most Famous Blessing 3.9.6 EA and Setfiles | Page 88 | Forex Factory

Options

Go to
Permalinks
Print Thread

Related Instruments

EUR/USD, GBP/USD, Gold/USD

Similar Threads

Can anyone convert the Blessing 3 v3.9.6.09 into MT5? 10 replies

Blessing Multipair EA 453 replies

Blessing EA by ananthhh 13 replies

'Blessing' EA Modification Wanted 3 replies

Forward testing results of my ea modified from Blessing 3.9.6 18 replies

Attachments: Most Famous Blessing 3.9.6 EA and Setfiles
Exit Attachments

Most Famous Blessing 3.9.6 EA and Setfiles

  • #1,741
  • May 2, 2020 6:59am May 2, 2020 6:59am
Quoting holyGrill
I am currently running Blessing on a demo account on EURUSD (15min) and AUDNZD (5min) with the set files attached to post: https://forex-station.com/viewtopic.php?t=8472216 (which btw contains great explanation of Blessing's settings and functioning). (main settings used BollingerEntry, MaxTrades=15, starting lot size = 0.01) I'm finding that, while the win % is good, it places very few trades: It made 20c on $100, (using FBS cent account) in roughly 1 week (maybe 10 trades or so placed, for both pairs). I've noticed from charts that...
That 'great explanation of Blessing's settings' is just a ripoff of the manual which the poster doesn't own up to until the end.

The version of Blessing posted there is an older version that is the benchmark so many others have worked from. I've posted more recent versions in this thread that do a number of very important things the original doesn't do. You can read more of this thread to learn about them.

Blessing does much more and is more configurable than the manual suggests. It can be adjusted to trade quite well in an incredibly diverse number of ways. I'm currently using it to trade 20+ pairs in a scalping configuration that has been quite successful so far. It appears to be a more efficient use of funds than many other approaches.

I've worked with a number of good multi-pair EAs and concluded that a cluster of Blessing instances could probably do much better. Certainly more work, since you have to optimize each pair, but the more recent versions have sped that up quite considerably. They have also greatly reduced CPU loading, which is important when running so many instances.

Good luck!
  • #1,742
  • May 2, 2020 3:28pm May 2, 2020 3:28pm
Quoting richard96816
I'm currently using it to trade 20+ pairs in a scalping configuration that has been quite successful so far
Hi Rich! I hope you are coronafree and are not locked inside home? At least not over the trade-free weekend...
Could you please post an example of one set-file for the scalping configuration? Probably small grid distances? What about the TP and SL?
As I know you, probably without the hedge function?

Thanks and stay healthy!
  • #1,743
  • May 2, 2020 6:57pm May 2, 2020 6:57pm
Quoting ursinho4711
{quote} Hi Rich! I hope you are coronafree and are not locked inside home? At least not over the trade-free weekend... Could you please post an example of one set-file for the scalping configuration? Probably small grid distances? What about the TP and SL? As I know you, probably without the hedge function? Thanks and stay healthy!
Thank you. Hope everyone is staying safe.

http://orthomolecular.org/resources/omns/v16n04.shtml

I'm exploring yet another extreme of Blessing, a grid size of one. Yes, Blessing is so good it even does that very well. I may increase grid size on some winners as the market calms down.

The problem with sharing scalpers is that small time frames are least likely to work across brokers.

I'll post some soon. Busy filling up my accounts with Blessing scalpers ...
  • #1,744
  • May 3, 2020 3:37am May 3, 2020 3:37am
Quoting richard96816
{quote} That 'great explanation of Blessing's settings' is just a ripoff of the manual which the poster doesn't own up to until the end. The version of Blessing posted there is an older version that is the benchmark so many others have worked from. I've posted more recent versions in this thread that do a number of very important things the original doesn't do. You can read more of this thread to learn about them. Blessing does much more and is more configurable than the manual suggests. It can be adjusted to trade quite well in an incredibly diverse...
Thanks for this Richard, I didn't realize there was a newer Blessing version already! (can't wait to test it. And going to scour this thread from start to end . I see the manual is here: https://www.jtatoday.com/blessing.html (assuming that's the latest manual?)

When you say scalping configuration I take it that's not martingale-style (which can consume large % of capital temporarily during large drawdowns)?

Regarding optimization, a MT5 version of Blessing would make it easier to optimize (better multi-variable, genetic algorithm in MT5) but don't think there's a MT5 version available yet (would take quite a bit of work to convert).
  • #1,745
  • May 3, 2020 12:01pm May 3, 2020 12:01pm
Quoting richard96816
Busy filling up my accounts with Blessing scalpers ...
That's why I started -some time ago- to try this multipair Blessing version. I paused now as I have more important things to do, Corona messed up my "real work life". But on the other side that is a trigger for me to search even deeper for the holy grail...

Blessing and many grid traders use points (or pips) for scaling. Since I started Forex trading I never liked to think in pips. For me it is not a scientific way to calculate. I know that many business administrators are not the best in mathematics, counting pips is easier than calculating percentage.
That's why I love fibo grids. They are independent of static pair based pips, move with the volatility.

So another approach I thought of, is to implement fibo ranges into blessing (using the GAF function?). Just an idea while I was outside riding bike with my daughter.

Yes, I am somehow still trying to find the swiss army knife. I hate to see that a set-up works for one pair (e.g. EURUSD) but kills the account with a different pair (e.g. AUDNZD). Because that means, that even the good set-up someday may encounter a similar situation that killed the other pair.

Goal would be to find code and settings that work for all pairs in all situations. If not profitable at least not blowing the account. And that code, in my humble opinion, can only work if based on percentages and not static distances.
  • #1,746
  • May 3, 2020 6:51pm May 3, 2020 6:51pm
Quoting ursinho4711
{quote} That's why I started -some time ago- to try this multipair Blessing version. I paused now as I have more important things to do, Corona messed up my "real work life". But on the other side that is a trigger for me to search even deeper for the holy grail... Blessing and many grid traders use points (or pips) for scaling. Since I started Forex trading I never liked to think in pips. For me it is not a scientific way to calculate. I know that many business administrators are not the best in mathematics, counting pips is easier than calculating...
Everyone needs to find their own preferences. It's a very inexact science (or art form.)
I like the simplicity and reduced workload of multi-pair robots. But I dislike having some pairs perform markedly better than others. I've been watching the 'Moving Day' robot from another site that does surprisingly well with one simple algorithm applied to all its pairs. Of course, the timeframe is high (hourly.) But it always seems to have its losing pairs. Those that you eventually leave out after you've run it long enough to see the patterns.

Blessing can trade almost any pair successfully. Its a little messy, and requires a fair bit of work to optimize, but its really solid. I no longer test it on demo. Everything goes straight to live accounts. Its that reliable. Though it took me quite a while to trust that.

You can optimize over years of history and expect decent results for quite a long time. Or you can tune it for just a few months of history and do much better.

I suspect automated optimization with a large number of pairs may produce a noticeably better end result.
  • #1,747
  • May 3, 2020 6:59pm May 3, 2020 6:59pm
Awesome EA, testing it now OP
  • #1,748
  • May 3, 2020 9:57pm May 3, 2020 9:57pm
Here's a set file to use for starting optimization for Blessing_3_v3.9.6.14. PotionPC set to 500. UseMM off. MaxTrades set to 1, good for unsettled markets.

Set Period to M5 (or whatever timeframe you prefer), Spread to 30, history to between 1 and four months and turn Genetic algorithm on. Set Model to Open prices only.

One pass, just a few minutes, should produce a winning setup.

Go through and optimize the various groups separately with Genetic algorithm off to make it better.

Groups:

1. ATRPeriods GAF
2. ProfitSet, MoveTP, TotalMoves
3. EEHoursPC, EEStartLevel, EELevelPC
4. MAPeriod, MADistance (if MAEntry is not disabled)
5. BollPeriod, BollDistance, BollDeviation (if BollingerEntry is not disabled)
6. FastPeriod, SlowPeriod, SignalPeriod (if MACDEntry is not disabled)

Keeping the groups to three or less entries keeps speed up. You can also set Start and Stop parameters to limit the range of each entry to about ten steps or so (five in either direction.)

Repeat steps 1 through 6 until no more changes are made, or until you get tired. :-)

Optimize for greatest Profit while watching that the total number of trades is reasonable. Also keep an eye on the Profit Factor. If one entry with a somewhat lower Profit has a noticeably higher Profit Factor consider choosing it.

Also keep a close eye on Average consecutive wins vs consecutive losses. I often get 15 or 20 to 1 there, though 5 to 1 is sometimes the best you can do. Good to have Average profit trade substantially larger than Average loss trade.

Ignore Modelling quality, we're not using ticks, fake or otherwise. The original Blessing (.9) took about ten times as long to optimize, without any better performance.


That's Blessing optimization in a nutshell. Piece of cake. Try things. See what you learn. Let us know if you find something good.

There are lots of good tutorials on optimizing MT4 Strategy Tester online. Please don't ask beginner questions here.


Good luck.


Edit: The filename is wrong, SL is on.


Attached File(s)
File Type: txt _optimize-start4-noSL-noMM.set.txt   10 KB | 514 downloads
  • #1,749
  • May 4, 2020 10:19am May 4, 2020 10:19am
Quoting richard96816
Here's a set file to use for starting optimization for Blessing_3_v3.9.6.14. PotionPC set to 500. UseMM off. MaxTrades set to 1, good for unsettled markets. Set Period to M5 (or whatever timeframe you prefer), Spread to 30, history to between 1 and four months and turn Genetic algorithm on. Set Model to Open prices only. One pass, just a few minutes, should produce a winning setup. Go through and optimize the various groups separately with Genetic algorithm off to make it better. Groups: 1. ATRPeriods GAF 2. ProfitSet, MoveTP, TotalMoves 3. EEHoursPC,...
Thanks a tonne Richard, highly appreciated!
  • #1,750
  • May 5, 2020 12:11pm May 5, 2020 12:11pm
Guys I have installed the EA and the presets.

I have not gotten it to work. No trades are being taken.

Can anyone help?
Attached Image (click to enlarge)
Click to Enlarge Name: Screen Shot 2020-05-05 at 12.10.06 PM.png Size: 88 KB
  • #1,751
  • May 5, 2020 2:18pm May 5, 2020 2:18pm
Quoting acehigh03
Guys I have installed the EA and the presets. I have not gotten it to work. No trades are being taken. Can anyone help? {image}
What are the entry conditions of your setup? Blessing can be very "slow", depending on the setup. Let it run on the tester to see if it opens trades with your settings.
  • #1,752
  • May 6, 2020 5:29pm May 6, 2020 5:29pm
Good evening!

My English is not the best, so I used the translator. I apologize for that!
I thank you in advance for anyone who can help me and answer my questions.

1 - Is version 3.9.6.14 better than the last official version, 3.9.6.09, in terms of performance?
I never switched to version 14 because I also like to analyze looking at the lines in visual mode, to make some adjustments later, but they don't work in the last version ... I would love to know, if this version is better than the last official one, if anyone would be interested in arranging the Display Overlay functionality, for a small fee?

2 - There is the Blessing 3 version v3.9.6.10a.MQL5 for MT5 published in this forum, but do not compile! Displays 100 errors and 2 warnings. Is there already a version for MT5?

3 - Can you indicate any youtube video or website where I can learn how to do backtesting?
I never got Modeling Quality higher than 25% and I can't know why!

4 - Regarding backtesting, if we are willing to put B3 v3.9.6.14 in an account and let it operate for us, why not put Stop Trade Percent = 100; PortionPC = 50 and MaxDDPercent = 100?

Thank you.
  • #1,753
  • May 6, 2020 6:04pm May 6, 2020 6:04pm
Quoting richard96816
Here's a set file to use for starting optimization for Blessing_3_v3.9.6.14. PotionPC set to 500. UseMM off. MaxTrades set to 1, good for unsettled markets. Set Period to M5 (or whatever timeframe you prefer), Spread to 30, history to between 1 and four months and turn Genetic algorithm on. Set Model to Open prices only. One pass, just a few minutes, should produce a winning setup. Go through and optimize the various groups separately with Genetic algorithm off to make it better. Groups: 1. ATRPeriods GAF 2. ProfitSet, MoveTP, TotalMoves 3. EEHoursPC,...
Blessing 3.9.6.14 on EURUSD M1

Is this a good result or a bad one?
Thank you!

P.S.: I'm noob
Attached File(s)
File Type: zip StrategyTester.zip   108 KB | 413 downloads
  • #1,754
  • May 7, 2020 12:15am May 7, 2020 12:15am
Quoting Gonzalez74
{quote} Blessing 3.9.6.14 on EURUSD M1 Is this a good result or a bad one? Thank you! P.S.: I'm noob {file}
Yikes. Average profit per trade? $0.93

Don't use ticks. Set Model to Open prices only and ignore Modeling quality.

You're getting there. Try again.
  • #1,755
  • May 8, 2020 6:57am May 8, 2020 6:57am
Note that the 1-minute timeframe is the noisiest and perhaps the least reliable.

I always liked the 1-minute when trading stocks, but that data is usually much more reliable. I mostly use the 5-minute as the lowest timeframe for currencies.

Unlike stocks and futures currencies don't have a single exchange as the authority on price. Each currency broker displays somewhat different data from the rest, and the smaller the timeframe the greater the difference. This is why you must optimize on your broker's data.
  • #1,756
  • May 8, 2020 8:34am May 8, 2020 8:34am
Quoting richard96816
Note that the 1-minute timeframe is the noisiest and perhaps the least reliable. I always liked the 1-minute when trading stocks, but that data is usually much more reliable. I mostly use the 5-minute as the lowest timeframe for currencies. Unlike stocks and futures currencies don't have a single exchange as the authority on price. Each currency broker displays somewhat different data from the rest, and the smaller the timeframe the greater the difference. This is why you must optimize on your broker's data.
Hi richard96816,
I really appreciate your feedback on my Blessing questions! For me, they are very valuable, coming from someone with a lot of experience that you have with this EA, so, thank you very much for the time you dedicate to them.
Having said that, in your experience, should I continue with version 3.9.6.14?
In the absence of feddback on this in particular, I assume so.
M1 is a timeframe with a lot of noise: OK!
Which one should I use? M5, or is it better to dedicate myself to improving settings for H1?
Quoting richard96816
This is why you must optimize on your broker's data
How to do this? Either we do baktesting from a short time window or the data is downloaded from MetaQuotes, whatever the broker I have in MT4. Any suggestions coming from you will be highly appreciated.
Thank you and stay safe!
  • #1,757
  • May 8, 2020 5:54pm May 8, 2020 5:54pm
Quoting Gonzalez74
{quote} Hi richard96816, I really appreciate your feedback on my Blessing questions! For me, they are very valuable, coming from someone with a lot of experience that you have with this EA, so, thank you very much for the time you dedicate to them. Having said that, in your experience, should I continue with version 3.9.6.14? In the absence of feedback on this in particular, I assume so. M1 is a timeframe with a lot of noise: OK! Which one should I use? M5, or is it better to dedicate myself...
It's good to see someone working through this.

I currently use 3.9.6.14, but I'm biased. I built many of the improvements, because I felt they were necessary. The base version (.9) in the #1 post is very slow and missing important stuff. Version .14 has been cleaned up to use '#property strict' so some of the work to migrating to MT5 has been done. EnableOncePerBar was added to allow use of Opens price only making optimizing way faster without sacrificing profit. And to make things more compatible with your brokers data. Added UseMinMarginPercent to allow better utilization of capital. That's a huge improvement in my opinion. A number of old bugs were also fixed, though more probably remain. Enhanced the PortionPC behavior which you can read about elsewhere.

EnableOncePerBar should always be on. It remains accessible for historical reasons, so people can see for themselves the difference.

Choice of timeframe has all kinds of issues. Smaller timeframes provide more trades and possibly more opportunities for profit even in unsettled markets. But smaller timeframes usually mean smaller profits per trade, though not always.

Trading multiple instances and currencies simultaneously has a number of advantages, from diversification to more opportunities to profit. There is a sweet spot somewhere between multiple instances and timeframes which I'm still looking for. I'm tending towards higher timeframes and more currencies, but a mix of timeframes might be advantageous.

You do not want to use MetaQuote data, or any other third-party data. As I've stated many times you MUST use your broker's data. Any other data will reduce your profits substantially, especially in smaller timeframes.

Don't ever use tick data with an EA that employs traditional indicators based on bars. People that sell tick data or extol it's virtues are not your friends. Modeling quality is not something to care about. Ignore it! That's for fake tick data. We don't want tick data at all.

Every broker's data is different. If you've downloaded data through the MT4 History Center you need to delete it all and refresh.

I've written extensively on all these issues before in this thread. Reading it might be helpful.

Good luck!
  • #1,758
  • May 9, 2020 9:54pm May 9, 2020 9:54pm
Hello richard96816,
many thanks for all your work. It is very much appreciated. I've learnt a lot by simple reading!
  • #1,759
  • May 10, 2020 6:31am May 10, 2020 6:31am
To the question of what timeframe to use, I think the 1-hour has some advantages for Blessing.

1. Faster to optimize.
2. Easier to find a good equity curve with many pairs.
3. Possibly the ideal timeframe for quite a few pairs.

The time necessary to test multiple timeframes can be excessive. Just trying one or two makes more sense. Or perhaps just standardizing on the 1-hour.

If you're learning how to optimize Blessing then the time saved will be significant.

If you want more trades then use more pairs together.
  • #1,760
  • May 11, 2020 4:54am May 11, 2020 4:54am
Hi Richard, having said that what should we do when EA start hedging while pair's trend reversed ? because on 15th april usdcad spiked up , and blessing starts hedged and my whole balance blown
6 traders viewing now