Great EA in backtest! - page 82
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
my test results gave move trades with a lower value peroids.
what value peroids are you using?
I'll run a value peroid of 23 with the rest of the settings the same and post results.Have you tried to use the version 1.88 with higher valueperiod counts?
Something like 60,120, 240, 480 to make you processor need some fresh air?
Just returned from a great vacation to the Zion Canyon Narrows.
This thread seems to have taken some new turns with some new insights...
Dave would you be so kind as to post the EA again which is giving you the best results with the settings you are using? TF/pair etc. I would like to see if I can duplicate your new findings.
Have you tried to use the version 1.88 with higher valueperiod counts? Something like 60,120, 240, 480 to make you processor need some fresh air?
1.88 is an unfinished version. I will not use it. FXspeedster is one of the developers and he said its bugged and not ready for use. I personaly found several things wrong in its code.
I have also posted several times that it is a bugged unfinished version. But nobody wants to listen. That is probly another reason nobody can reproduce my results. Argorn is getting close and he uses the 85f/85g versions like I do. Same with Kat. she gets good results and uses the good versions.
I will try running a few back tests with the high value peroids, but I already know what I'll find. (not to be conceded but my settings will probly be best.) But I am willing to give it a test.
If you want my results I suggest you do what I do, and use what I use. That would make the most sence
Dave
P.S. Argorn...I'll post my version in a few.
Here it is.
edit: holy cow...using 120 value peroid it moves slightly faster than real market speed. This could take days to run out a test.
Sorry man
I posted the version I am using with my settings a few pages back.
I don't use this on a demo. I run it on a real money account.
Demo's are on a different server and price feeds are different. So demo trading is useless. What works on demo might not work same on real.
Back tests were different too when I was doing back tests on demos.
This is probly why nobody can get my results. Your all running them on demos and I am running on real account.
DaveI saw your settings and stuff, i was just curious because you posted that you changed the cci to 50 instead of 100... i was just curious to see how you did that.
Sorry mate... didn't mean to upset you,
B
I saw your settings and stuff, i was just curious because you posted that you changed the cci to 50 instead of 100... i was just curious to see how you did that.
Sorry mate... didn't mean to upset you,
BI wasn't upset sry if I seemed that way. I had a little to drink.
the above version has the CCI changed in it.
--------------------------------------------------------------------------
Kamyar has tested the JPY settings and gets bad results. On north finance account.
I was thinking...We don't realize how much brokers have controle of our platforms. When I changed from backtesting on a demo to backtesting on a real account my tests were different. This shouldn't be. Because the charts are offline, and its all the same Alpari data. But the differance shows up in testing CT because it evaluates each and every tick. The tick have to be getting manipulated by the different brokers platform. On line and off line. The manipulation must be built in or pre set in each indavidual platform, as the broker desires. Just like the way they can see what indacaters we use on our charts so the risk assesment department can manipulate and lag them to make trading hard. This proves that they use the platforms to manipulate price. Too sublte for the naked eye. But enough for CT to be affected.
Some of you can get good gbp$ results and I cannot.
Edit: I also know this: FXDD doesn't seem to magnet to my s/l like some brokers do. MG does, so does GFT and fxcm. I have had accounts with all 3. If the price comes with in 5 pips of your stop it will surly get hit.
But I have had my s/l get with in 1 pip on fxdd and turn and go to profits. On several occations.
Any broker that garentees fill on all orders will grab s/l's.
gmt 10 for alpari what version you using?
I am using 1.85h which is the same as 1.85g but with the news blackout code added.
I am using 1.85h which is the same as 1.85g but with the news blackout code added.
This is as frustrating for me as it is youguys. I am trying to be helpfull but what I do comes out different from person to person.
try down loading the version I just posted. set gmt=10 and no trade times 00,01,04,06 just like my tests. Put it on $jpy 1 hr.
Make sure the value peroids are 7 and pivot=fasle and CCI=false. s/l is 15 and trailingstop=false.I think I made those the defalts.
start the backtest peroid at august 11th 05. end it in sept 29th 06 same as mine.
every one do this exactly as mine on your own brokers and see what comes up.
I don't think its the EA I think its the different brokers. the EA makes the same calculations in the cyberiadecitions. The only thing that can be different is the way each brokers platform is processing the ticks.
then everyone should e-mail the makers of the metatrader and ask what the heck is going on. And demand an explanation.
who added the news block? Are you still here?
A guy on my yahoo group figured out how to import DLL's and have an EA read web pages. It has the potential to read the news times from forex factory's site. I can forward the e-mail that better explains this and it has 2 EA's attached that he made to read web pages and display info on the chart.
We need some one that can impliment this into CT so it can read the news times from forex factory calander, and block trade hours each day automaticly. You can be able to det it to block only certian news, like euro news and U.S. news.
try down loading the version I just posted. set gmt=10 and no trade times 00,01,04,06 just like my tests. Put it on $jpy 1 hr. Make sure the value peroids are 7 and pivot=fasle and CCI=false. s/l is 15 and trailingstop=false.I think I made those the defalts.
here You go David,
two backtests,
one with the version You posted - default settings +hours disabled and GMT like in Your tests;
second with the version i use.
data from alpari, tests taken on IBFX real account.
i do not trust in backtests anymore. sry but i will not share Your enthusiasm until i get some longterm fwd testing results close to any of our backtests.