The Pouria EA Development Thread | Page 23 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

The Pouria Method 564 replies

Lessons Learned during System Development 29 replies

(pouria method II on the test) 218 replies

System Development Secrets 5 replies

Metatrader 4 Development Course 3 replies

Attachments: The Pouria EA Development Thread
Exit Attachments

The Pouria EA Development Thread

  • #441
  • Mar 26, 2007 11:07am Mar 26, 2007 11:07am
One thing on the MACD indicators. I have them set to the 15, 26, 1, and then have them applied to the LOW.

What I messed up on with the new demo was using the "custom" MACD without the "applied to" setting this time around. I'm hoping that is the reason for the bad start. Like I said, I fixed them and went 4 for 5 last night....the lone LOSER was of course the GBPUSD.
  • #442
  • Mar 26, 2007 8:34pm Mar 26, 2007 8:34pm
hi all,

yesterday 26/3 there was a good cross on the eur/usd but the ea didn't buy any ideas why?
thnx in advance
  • #443
  • Mar 27, 2007 9:57am Mar 27, 2007 9:57am
I don't know what's wrong, but things aren't getting better. The account is now down 25%. GBPUSD continues to be a big loser.

I'm wondering if something was messed up when I lost all of my charts and data when I started the second demo. Having to reupload everything may have thrown it out of whack. I just have no idea why it's suddenly doing so bad when before it was doing so well. Went 2 for 5 yesterday.
  • #444
  • Mar 28, 2007 10:43am Mar 28, 2007 10:43am
It's now 1 for 8!! The demo account ($1M) is down $244,285. The GbpUSD pair alone count for $209,333 of that!

I'm trying to figure out how a pair can be such a big winner for this EA on a demo all of last month, and then such a HUGE loser in the last week on a new demo. Any ideas?
  • #445
  • Mar 28, 2007 11:06am Mar 28, 2007 11:06am
I just looked, and in the last demo the GBPUSD pair went 23 and 5, an 82% positive ratio.

This demo has gone 1 and 7, a 12.5% positive ratio. can't figure it out. It's about to go 1 and 8!!
  • #446
  • Mar 28, 2007 7:09pm Mar 28, 2007 7:09pm
That's why I stopped trading the GBP with beta 3. It had too many losers. I was confused when you kept on having winners. Then we thought it might be the feed...I was using ibfx and you were using migfx...

Who knows....but thanks for all the testing.

BP
  • #447
  • Mar 29, 2007 3:08pm Mar 29, 2007 3:08pm
Quoting 1Sundevil
I just looked, and in the last demo the GBPUSD pair went 23 and 5, an 82% positive ratio. This demo has gone 1 and 7, a 12.5% positive ratio. can't figure it out. It's about to go 1 and 8!!
Sundevil -

And you were all set for Easy Street. This sucks...

I haven't looked at the charts, so this is why I ask - is there any kind of difference in the price action, i.e., ranging vs. trending?

Is the stop/loss at the same level? I believe you've used both 40 and 50 for S/L - and you may have changed in mid-stream. Only you would know for sure.

Perhaps you were getting lucky with the change to 40 S/L? How much are your recent trades missing the mark by?

If nothing turns up there - I'd do the obvious thing. Stop trading the pair and see if the EA can continue to profit from the remaining pairs.

This could be a problem that migrates from pair to pair, essentially rendering the EA useless.

However, if the performance for all of the other pairs is roughly consistent from month to month - you've probably got a subtle flaw or change somewhere within the code that you're overlooking.

P.S. Here's another bright idea... I believe that there are updated versions of this EA available. Look into running both of them. I know this will make your record keeping a lot tougher - but you've got to get to the bottom of this.
  • #448
  • Mar 29, 2007 9:51pm Mar 29, 2007 9:51pm
Quoting BearPaw
....It had too many losers. I was confused when you kept on having winners. ...

BP
exactly like me.I used Alpari demo and got bad results on all pairs except eurchf, nzdusd and eurjpy.But realy can I trust and run EA on real account on these three pairs? I think there is no guarantee for being real like demo!!
  • #449
  • Mar 30, 2007 1:28pm Mar 30, 2007 1:28pm
I dumped that last demo and reopened another one and reset all of my charts etc.. So far the EA has dropped me into 14 trades and I'm 9 for 14. More importantly the GBPUSD is now 3 for 3. The GBPCHF is 0 for 3. This was one of my "new" currencies that I wanted to test. If it doesn't look good through next week I'll drop it.

Another thing I've been thinking. I've been trying to use 6 different currencies. Why? I have no idea. I may just find the best four and see how that effects my account.

I don't know what happened with the last demo, but things "appear" to be back on track with the new one. We'll see. My "live" MIG account has been apporved, but I want to test this a little more to find the best combinatio of pairs and settings first.
  • #450
  • Mar 30, 2007 1:43pm Mar 30, 2007 1:43pm
Is there some reason you thought your demo was invalid? I would caution against ignoring a "bad test". That is really why we test these things. I know that we all really want the EA to be successful, but we shouldn't ignore a bad run. They happen too often on LIVE accounts.
Don Life is expensive, but includes a free trip around the sun.
  • #451
  • Edited 2:49pm Mar 30, 2007 2:35pm | Edited 2:49pm
Quoting traderone
Is there some reason you thought your demo was invalid? I would caution against ignoring a "bad test". That is really why we test these things. I know that we all really want the EA to be successful, but we shouldn't ignore a bad run. They happen too often on LIVE accounts.
When I opened up a second demo, after one day the account froze and all of my charts disappeared. So I opened up another demo and had to re-do everything, and that is the one that did terrible. The telling sign for me was the GBPUSD pair. It went 25 for 28 in the first 30 day test and 1 for 9 in the second 3 day test. That's not a coincidence.

I'm not ignoring it, but it was so far to the other end of the spectrum that it just screamed "something's wrong". That's why I closed it and opened up a new demo to further test this EA. If this one doesn't do well then we'll have to tinker some more. So far the GBPUSD has gone 3 for 3. 1 day down, 29 to go.
  • #452
  • Apr 2, 2007 8:55am Apr 2, 2007 8:55am
Well, the new demo isn't as bad as the last one, but certainly ot as good as the first. Now, I am using different pairs and different SL's and TP's, so we'll see, but this also has me thinking.

I sent Mikhaildgreat a private message and asked him about the possibility of creating an "Anti-Pouria EA". Basically everything would be the same except where the EA is programmed to "buy" right now it would "sell" and vice versa. I can see many times where the EA has dropped me into "sell" trades at the bottom of a move, or "buy" trades at the top. It hits my SL with ease and was never a threat to hit my TP.

Given that it appears easy to find pairs that DON'T work and hit the -40 SL way before they would ever hit a +15 or +20 TP, it stands to reason that if we reversed the two (+40 TP and -20 SL) we would have a big winner with little drawdown. Sometimes finding an EA that consistantly doesn't work can be just as good as finding one that does.
What do you guys think about this idea?
  • #453
  • Apr 2, 2007 10:57am Apr 2, 2007 10:57am
Heres your Anti-Pouria sundevil... Enjoy!

Mikhail
Attached File(s)
File Type: mq4 Anti- Pouria EA Beta 3.mq4   12 KB | 867 downloads
  • #454
  • Apr 2, 2007 2:51pm Apr 2, 2007 2:51pm
Quoting 1Sundevil
Well, the new demo isn't as bad as the last one, but certainly ot as good as the first. Now, I am using different pairs and different SL's and TP's, so we'll see, but this also has me thinking.

I sent Mikhaildgreat a private message and asked him about the possibility of creating an "Anti-Pouria EA". Basically everything would be the same except where the EA is programmed to "buy" right now it would "sell" and vice versa. I can see many times where the EA has dropped me into "sell" trades at the bottom of a move, or "buy" trades at the top. It hits my SL with ease and was never a threat to hit my TP.

Given that it appears easy to find pairs that DON'T work and hit the -40 SL way before they would ever hit a +15 or +20 TP, it stands to reason that if we reversed the two (+40 TP and -20 SL) we would have a big winner with little drawdown. Sometimes finding an EA that consistantly doesn't work can be just as good as finding one that does.
What do you guys think about this idea?
Good idea Sundevil....lets see how she does.

BP
  • #455
  • Apr 2, 2007 6:30pm Apr 2, 2007 6:30pm
Quoting 1Sundevil
Basically everything would be the same except where the EA is programmed to "buy" right now it would "sell" and vice versa.
What do you guys think about this idea?
I have no great expectations - other than be greatly amused.

If successful, this will be one of the great feats in FF history - running an EA backwards and producing profit.

This is going to be great. Let the testing begin, SunDevil!
  • #456
  • Apr 2, 2007 11:47pm Apr 2, 2007 11:47pm
Quoting brentmack
I have no great expectations - other than be greatly amused.

If successful, this will be one of the great feats in FF history - running an EA backwards and producing profit.

This is going to be great. Let the testing begin, SunDevil!
This should be interesting to say the least! I opened an Interbank FX demo account, as I couldn't figure out how to open another MIG account to run simultaniously with the regular Pouria EA that I'm still testing. No worries as I earlier had terrible results with Interbank, so this should be fun.

Everything is up and running. I'll see what it did overnight when I wake up. I'm using +40 for TP and -20 for SL and have set the "confirm" to true, so everything is like the regular demo testing. I'm using the GBPUSD, EURUSD, USDCHF, USDJPY, GBPJPY and GBPCHF. I think the GBPUSD and the USDCHF are the wildcards. If they don't work well I'll exchange one for the EURAUD as it's been doing bad on the other test.

If this works I will laugh my a$$ off! If it doesn't here, I'll retest on MIG.
  • #457
  • Apr 3, 2007 10:28am Apr 3, 2007 10:28am
well, the Anti-Pouria took two trades last night that are both positive right now. here's a chart of the EURUSD, it's a thing of beauty, just scroll to the right:

http://i141.photobucket.com/albums/r...l520/chart.gif
  • #458
  • Apr 4, 2007 12:27pm Apr 4, 2007 12:27pm
Had some computer problems...but was going to set up anti and pouria together....

Hopefully will get to it by tomorrow.

BP
  • #459
  • Apr 4, 2007 2:44pm Apr 4, 2007 2:44pm
Quoting BearPaw
Had some computer problems...but was going to set up anti and pouria together....

Hopefully will get to it by tomorrow.

BP
I have the anti-pouria running on both MIG and Interbank demo's to compare. This should be very interesting as I'm still adjusting the TP and SL to find the optimal setting.
  • #460
  • Sep 23, 2007 10:03pm Sep 23, 2007 10:03pm
I to have an idea, to make that this 01 EA alone buys or venda time alone, to each crossing on the 02 red lines, I to think that it goes to give certain, that they find? MIKHAILDGREAT
it could make this?
0 traders viewing now