Blessing 2 - page 72
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
Backtest / Demo
You can run backtests until you are blue in the face.
You really should put together some demos, so you
can see how Blessing behaves in real market conditions, with data it has not seen. This will give you a realistic
perspective regarding margin, draw-down, profitability,
etc. This is where the rubber meets the road. You will never feel comfortable, until you walk Blessing forward
on a live feed.
Keep it simple. Start a demo with just usdjpy and
proper set file, on D1 TF, with .01 starting lots.
Recommended balance should be at least $500 on
1-cent pips, or $5,000 on 10-cent pips. For added
safety, I would double the starting balance.
A second demo could be run at the same time, with
two pairs and the appropriate increase in balance.
Then you can compare the performance of two pairs
against your single pair account, in order to assess
any changes in margin, DD, and profitability.
RobRob,
That is not the point.
I do run Blessing since last week already in 2 forward tests (2 different brokers) because of the points you mentiioned.
I would like to know why the backtest does not work anymore.
In the meantime I found out that it has something to do with the dowloaded historical data from Alpari.
Very strange and frustating.
But again maybe someone had similar problems and can help me out before my face turns deep blue
.
Best
Antomi
In the meantime I found out that it has something to do with the dowloaded historical data from Alpari.
Have you tried downloading from Metaquotes?
Interbank FX is introducing 5th decimal pricing!
Just saw this email, does anyone know if this will affect this ea?
Interbank FX is introducing 5th decimal pricing!
As part of our continuing commitment to delivering the optimal customer experience we are introducing 5th decimal pricing on our currency pairs. The 5th decimal prices will be available on our demo servers on Sunday afternoon at 6pm EST on April 12th, 2009 and will be available on the live trading servers on Sunday afternoon at 6pm EST on May 17, 2009.
Why 5th decimal pricing?
Adding the 5th decimal point can reduce the spread, or difference between the buy and sell prices of a currency pair. You may notice tighter spreads as we enable the 5th decimal pricing on our trading platforms.
How will this effect Expert Advisors (EAs)?
Most EAs will be affected by the introduction of the 5th decimal point. Please make sure that you have adjusted all pip value related parameters, such as take profits or stop losses, to allow for the additional decimal point. For instance, if you are currently using a stop loss level of 20 on a four decimal pair, you would have to change the value to 200 to get the same execution on a pair with a 5th decimal. Please be aware that some EAs will need fundamental changes to ensure that they will work as you expect. We would suggest that you use your EAs on our demo accounts before using them on your live account to ensure they are working properly.
How does this effect Stop Losses, Take Profits and Maximum Deviation?
All take profit, stop loss and maximum deviation values that are expressed in points should be multiplied by 10. For example, if you would like to enter a Maximum Deviation of 5 pips, you would enter 50 into the Maximum Deviation field.
How does this effect the profit / loss values expressed in the Trade tab in the Terminal?
All profit & loss values already incorporate the 5th decimal place adjustment when they are expressed as Points in the Profit column under the Trade tab in the Terminal window in MetaTrader.
Does this effect Trailing Stops?
Yes. When you are setting a trailing stop value you will multiply it by 10.
Who do I contact if I have additional questions?
You can contact our award-winning customer service team from Sunday at 5:00pm EST to Friday at 5:00pm EST by email, phone or live chat.
Toll Free:
US and Canada: 866.468.3739
Australia 1.800.884.912
Indonesia 001.803.017.9112
Malaysia 1.800.813.776
New Zealand 0800.445647
Singapore 800.101.2097
United Kingdom 0.808.120.1966
International: +1.801.930.6800
Email:
help@ibfx.com
Live Support:
Click here
- The IBFX Team
* Disclaimer - Interbank FX LLC / IBFX.com is a Futures Commissions Merchant. Interbank FX LLC / IBFX.com is not an advisory firm. The contents of this email are not to be construed as a recommendation to buy or sell. Only investors who are aware of the risks inherent in margined currency trading should use this information. Interbank FX LLC / IBFX.com accepts no liability for losses a client might incur while trading foreign currencies.
Blessing
Hi Iron,
Thank you for the insightful comments.
Which 3 blessing pairs are you using?
Thanks!
All,
Quite a spirited discussion, I see. Anyway, fixing something that ain't broke is tough for me to do. BTW, Fifth will post v5.0 on the first post shortly. In the mean time, go back a few posts to get v5.0. If you already downloaded v5.0, do it again since I put a new v5.0 changing thw warning display.
First of all, backtesting/optimizing must be done on a tester platform that allows micro lots. Alpari does not allow micro lots if you simply download the platform. You must call them and have them enable your demo to trade micro otherwise you will get ordersend error 131.
Second, Blessing 2 needs daily timeframes to operate correctly. TF corelation is not simply a factor of 24. It's about noise, data streams, price quotes, etc. Too many factors in fact that daily TFs are simply the best. H4 is even pushing it!
To get Blessing 2 to work properly, you should backtest for a good MA period first with the multiplier set to 1.4, autocal false, GAF = 1 and equity protection set to 50%. Test it on a standard platform with $2500K as the autolot logic uses $2500K to get .01 lots. I recommend backtesting only 8-12 weeks. Once you get an MA that makes it through that timeframe, try adjusting the autocal to true then optimizing for the GAF from .2 - 1.5. If that doesn't work, stick with the manual grid. Finally, optimize for the mutlipler, 1.4 - 1.5 in .01 increments. That's it! Get your set file that way.
USDJPY, AUDCAD, EURUSD and USDCAD are the only pairs that have stood the test of time. I did not develop set files for any other pair. Blessing might work on others but that is up to you.
Finally, I do not post personal account information but I will offer this. I use IBFX. I use 3 Blessing pairs only and portion each with a value of 3. My account is micro (penny a pip). I use Piplite and one other personally developed EA only. I portion Piplite to 4 and my other EA to 3 and 4 (I use my other EA on 2 pairs. With this setup, I get about $30-100 per day on my $2500 capital base. No, my other EA is not available but Blessing and Piplite are on this forum. Read up on portion control and what it can do for you as far as account protection goes. You will understand what I mean if you try it on a demo.
Happy Trading
v/rBlessing Info
Iron,
Thank you for your information. That was realy helpful.
Monday I will call Alpari to change my demo account.
But nevertheless I have the same problem with IBFX and that irritates me.
Rob,
no I didn't try the data from Metaquotes, since everybody is praising the Alpari data.
Antomi
Ok Antomi
thanks for the long explanation
I understand your thoughs now
Backtests are always a problem because of the lack of real data
Most of the bars we find in historical data are etrapolated data...because of insufficient real data.
THat's a problem all traders know, that's why i most of the time rely mainly on my forward tests rather than on backtests. But in the other hand the problem of forwardtest is that you never have long enough tests
What would you propose to improve this wonderfull ea, without the MA filter idea that i don't like because it would obviously change the ea logic and reliability, i'm sure of that.
Any idea would be interesting indeed.
My personal idea of improvment is : bigger balance per pair, nothing else really...
Hi Alex,
I guess 1 day has 24 h.
If you backtest 1 week on daily basis you only have 5 bars, but if you do that on a 1 h base you have 120 bars (5x24=120).
Going to 5 min TF you have already 1440 bars.
I'm not sure whether I can explain what I mean (english is not my native language, sorry).
You can look from different perspectives, the makro and micro one.
If we assume that the Forex market is random than a price action trading system should work similar in every TF.
The smaller the TF the more data available and the more reliable results.
Every system developer has the intension to check his/her system over the most possible time period.
If I test Blessing on a daily basis over a 2 year period I have only about 520 bars (for the sake of order that is not 100% correct but helps to get the picture).
Who tells me that this is sufficiant? Actually that is nothing.
But If I check Blessing over the same 2 years with a TF of 1h I have 12,480 bars. And even going down to 5 min TF I have 149,760 bars.
If I can find the parameters for Blessing to survive these bars I'm on a very safe side an would throw all my money in it (maybe not all
).
If the statement is true that Blessing can only survive on a 4h/1d TF (and that is according to my information only based on 2 years backtests, or 520 bars) this would mean to me, that the time will come when the account crashes.
Or in other words, up to now (mainly) Iron proofed that with his settings Blessing had survived about 2 years backtests on 4h/1d TF but at the same time that means also that it didn't survive on more data (because a smaller TF is nothing else than more data, from the micro perspective).
I'm pretty sure if we would backtest Blessing with the given settings on 1d TF for 48 years (48 years backtest with 1d TF equals 2 years backtest with 1h TF) it would not survive.
Because if you hit a worst-case scenario in your 2 years 1h TF test you will have it also in your 48 years backtest with 1d TF.
Oh, you migh say 48 years is a long time and than I will be already retired.
The bad thing is only, that statistically the worst thing can happen every day, even every bar.
What I'm trying to say is, that 520 bars successfull tests are just not enough for me and that I urge us to make the system as much as possible bullet proof.
But these are just some basic thoughts, much more other factors have to be taken into account.
By the way - and I said that before - Blessing is the best EA I've seen so far and therefore thank you FiFtHeLeMeNt for coming up with it and Ironto make it even better.
You both
Wish you all a great weekend.
Best
AntomiBlessing works best with some draw down. If the pair
has a decent amount of chop, it will often see 2-6
progressions. The key is chop, not trend. You want
pairs that basically trade within a channel, but with
good price movement. Make sure your account has
plenty of room for neg float and don't use a multiplier higher than 1.4.
RobHi Rob
I've been looking through my trading account and noticed that Blessing has been closing in profit upto level 5 (0.02, 0.03, 0.04, 0.06, 0.08) . Once it gets above level 5 it closes at a loss because its setting the TP at entry 2 levels back. I'm not sure if this is what its supposed to do, but it doesn't look right. Maybe other people can check this as well.
Thanks for sharing. I tested the system , the system is profitable, very interesting.
This question on 5 decimals has been asked a couple of times now but no one has answered it. If I am using a set file called "eurusd m240 4 decimal.set", this suggests it is set up for 4 decimal brokers and there is a 5 decimal file somehere. I would appreciate if someone could answer this.
This is the best EA I have ever used. 10k to 26K in 4 months on the eurusd only. Big thanks to all.
This question on 5 decimals has been asked a couple of times now but no one has answered it. If I am using a set file called "eurusd m240 4 decimal.set", this suggests it is set up for 4 decimal brokers and there is a 5 decimal file somehere. I would appreciate if someone could answer this. This is the best EA I have ever used. 10k to 26K in 4 months on the eurusd only. Big thanks to all.
Hi Tonyd,
Could you maybe share your settings?
Rgds
Eddie