Caillou: another always-in auto-trader by SteveHopwood | Page 6 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Another Day, Another Chair, Another Day in the Office 203 replies

SSS EA v1.0 - Auto Trader Bot 623 replies

Candle Stick Pattern Auto Trader 12 replies

Auto trader for Currenex? 2 replies

SteveHopwood EAs 4 replies

Attachments: Caillou: another always-in auto-trader by SteveHopwood
Exit Attachments

Caillou: another always-in auto-trader by SteveHopwood

  • #101
  • May 10, 2011 11:00am May 10, 2011 11:00am
Hi guys,
the counter trend version worked well today.
I stopped it for now, i'll start it again tomorrow at London open.

Will see if it works on long run or not.

Cheers
gspe

http://www.myfxbook.com/widgets/110665/small.jpg
  • #102
  • May 10, 2011 11:04am May 10, 2011 11:04am
Quoting gspe
Hi guys,
the counter trend version worked well today.
I stopped it for now, i'll start it again tomorrow at London open.

Will see if it works on long run or not.

Cheers
gspe

http://www.myfxbook.com/widgets/110665/small.jpg

Great results. Will change my demo (still at 18% DD), and load your countertrend version.

Thank you
  • #103
  • May 10, 2011 11:31am May 10, 2011 11:31am
Dear Members,

I'm a newbie, and came across an EA that might have good potential? Can you guys please check it out. It was posted by GoogG@bi at MQL4 Code Base (http://codebase.mql4.com/7405).

The following is an extract of his post:

"Back-testing/Forward-testing details:
Period where EA is successful
Settings 1

Date tested: Jan 01, 2010 to Aug 31, 2010. Currency: EURUSD only. Periodicity: H1. Expert Properties: Minimum deposit: $50, Position: Short only, Lotsize: Auto. Draw-down (max): $603.00 or 76.83%. Draw-down (relative): 76.83% or $603.00. Positions won: 57. Positons Loss: 0.
Result for the period above settings show: ROI: 1824% (e.g. $50 became 912.37 from jan/01/2010 to aug/31/2010)


Settings 2
Date tested: Dec 01, 2010 to March 08, 2011. Currency: EURUSD only. Periodicity: H1. Expert Properties: Minimum deposit: $50, Position: Long only, Lotsize: Auto. Draw-down (max): $328.00 or 29.55%. Draw-down (relative): 47.18% or $119.24. Positions won: 25. Positons Loss: 0.
Result for the period above: ROI: 2382% (e.g. $50 became 1191.19 from dec/01/2010 to mar/08/2011)

This EA (GoodG@bi) has potential of making it to the top 5 most profitable all-time expert advisors in history. I have come to understand the following:

 

  1. That from Jan 01, 2010 to aug 31, 2010, the price was trending downward (bearish), that is why 'short only' position was ideal for trade and no loss was incured by the EA.
  2. That from sep 01, 2010 to nov 01, 2010, the price has not decided whether to continue downward or reverse upward, that is why the chances of 'zero loss' was slim. Therefore, NO TRADE because result will be full of + and -.
  3. That from dec 01, 2010 till mar 08, 2011, the price action decided for an upward trend (bullish), that is why 'long only' position was ideal for trade and EA had ZERO

LOSS. Therefore, if there is some kind of way by which we can (from the previous 6 months period) predict if the trend was bullish, bearish, or neutral, then we can have a 'perfect' expert advisor that trades safely and without LOSS.


If you are a programmer, I will want you to contribute to this project in the following ways (and other ways which I may not have anticipated)

 

  1. Add 'lines' that will make the EA detech trend over a wide period past (6 months or more).
  2. Add 'lines' that will switch EA from 'short only' trading mode to 'long only' trading mode and vice versa, depending on the prevailing long term trend.
  3. Add 'lines' that can carefully detect 'flat trend' and set EA to 'no trade'.
  4. Add 'lines' that are hitherto not inluded, and that will make the EA better and safer (having in mind the ZERO LOSS campaign)".


I have backtested the EA and it really seems like it has potential. Can the above be programmed? Please let me know.

Regards
saguy

  • #104
  • May 10, 2011 11:34am May 10, 2011 11:34am
I attached herewith the mq4 file.

Regards
saguy
Attached File(s)
File Type: mq4 [email protected]   8 KB | 278 downloads
  • #105
  • May 10, 2011 12:24pm May 10, 2011 12:24pm
the worst case scenario comes when the 1st trade is placed and a symmetrical broadening triangle starts to develop. I really hate these patterns in manual trading. This bot doesn't like them at all. Unpredictable, ergo unavoidable. Maybe a time/trade count based kind of "stop" would help, maybe smaller profit target / breakeven / or even a relatively small loss should be taken if the basket runs too long.
Attached Image (click to enlarge)
Click to Enlarge Name: caillou.png Size: 23 KB
:-anything
  • #106
  • May 10, 2011 12:27pm May 10, 2011 12:27pm
Quoting gspe
Hi guys,
the counter trend version worked well today.
I stopped it for now, i'll start it again tomorrow at London open.

Will see if it works on long run or not.

Cheers
gspe

http://www.myfxbook.com/widgets/110665/small.jpg
Hi again gspe,

You have created a monster, 5 pops in 1 hour.....
So the Countertrend EA opens a Buy when all Sell orders are in profit? I'm not sure about its rules......
  • #107
  • Edited 2:03pm May 10, 2011 12:49pm | Edited 2:03pm
cool, exactly what I mean after my backtests, thnaks


EDIT: can't increase lot multiplier, any bug ?

Quoting gspe
Hi guys,
i was playing with the EA this morning and i thought that maybe a counter trend version could works as well.

So attached you find a version in which i simply inverted the buy and sell Orders Condition i'm sure that we need more option in order to make the EA works better.

So play with it and if you find that this option could be useful we can try to improve it.

Cheers
gspe
  • #108
  • May 10, 2011 1:14pm May 10, 2011 1:14pm
Quoting Caillou
Hi again gspe,

You have created a monster, 5 pops in 1 hour.....
So the Countertrend EA opens a Buy when all Sell orders are in profit? I'm not sure about its rules......
If I'm understanding gspes mod rules correctly, It makes sense that the regular version is profitable in trending markets and this gspe modified version in ranging markets.

How about implementing some sort of "ranging" or "trending" filter to determine which style to trade? the usual 45 > RSI 20 > 55 comes to mind. Just a thought
  • #109
  • May 10, 2011 1:33pm May 10, 2011 1:33pm
Quoting Caillou
Hi again gspe,

You have created a monster, 5 pops in 1 hour.....
So the Countertrend EA opens a Buy when all Sell orders are in profit? I'm not sure about its rules......
Holy Lord ! This thing goes nuts ... have you guys tried setting the RSI TF to 1 (yes, M1) and the period to something like 20-30 or so ? If we do ultra fast scalping, the daily RSI is quite useless imho. Set it to fast also ...

An now, Steve or whoever ... please implement MaxTradesAllowed setting and a pipstep, maybe SL ... and off it'll take us maybe ... it survies BT from 3rd January until today, and that may mean something

Can't wait ...
"We are all just prisoners ... of our own device ..." (Hotel California)
  • #110
  • May 10, 2011 2:03pm May 10, 2011 2:03pm
Quoting acostafulano
If I'm understanding gspes mod rules correctly, It makes sense that the regular version is profitable in trending markets and this gspe modified version in ranging markets.

How about implementing some sort of "ranging" or "trending" filter to determine which style to trade? the usual 45 > RSI 20 > 55 comes to mind. Just a thought
Hi,
i was thinking about the same thing.
Maybe we can start trying 45 > RSI 20 > 55 filter

gspe
  • #111
  • May 10, 2011 2:09pm May 10, 2011 2:09pm
Quoting acostafulano
How about implementing some sort of "ranging" or "trending" filter to determine which style to trade?
If you can know that, you 'll be billionaire very soon
  • #112
  • May 10, 2011 2:38pm May 10, 2011 2:38pm
Quoting gspe
Hi,
i was thinking about the same thing.
Maybe we can start trying 45 > RSI 20 > 55 filter

gspe
as I just opened my eyes on the settings, there IS Steve's usual rsi filter already built in ?

I could limit my 100 allowed trades a little by twisting the 45/55 settings in the sourcecode to 70/30, but it still does open too much trades. Gotta limit that, but I am not quite sure what a pipstep would do to the strategy. Hopefully we'll soon see ^^
"We are all just prisoners ... of our own device ..." (Hotel California)
  • #113
  • May 10, 2011 2:56pm May 10, 2011 2:56pm
Quoting gspe
Hi,
i was thinking about the same thing.
Maybe we can start trying 45 > RSI 20 > 55 filter

gspe
This would require that the code integrates both its original trading style and the one proposed by you, and chooses among them based on the RSI reading. Could you be able to code something like this?
  • #114
  • May 10, 2011 3:45pm May 10, 2011 3:45pm
gspe,

there an bug in your invert caillou, it made only sells after made an first buy, saw on caillou's fxbook too
  • #115
  • May 10, 2011 3:54pm May 10, 2011 3:54pm
Quoting gregart
gspe,

there an bug in your invert caillou, it made only sells after made an first buy, saw on caillou's fxbook too
yeah, i know i just inverted the buy with sell not all the logic, was just a quick change to see what's happens
  • #116
  • May 11, 2011 2:34am May 11, 2011 2:34am
Good morning,

CounterTrend Version is still doing well, account up 4,62%, since yesterday 19h00 (GMT+1, or +2, donīt remember), in Asian Session too.

Any movement has a retracement, if we have enough orders in this retracement, the basket will be closed with profit. DD, sure, as always, but how much? Letīs see....
  • #117
  • May 11, 2011 2:52am May 11, 2011 2:52am
Quoting Caillou
Good morning,

CounterTrend Version is still doing well, account up 4,62%, since yesterday 19h00 (GMT+1, or +2, donīt remember), in Asian Session too.

Any movement has a retracement, if we have enough orders in this retracement, the basket will be closed with profit. DD, sure, as always, but how much? Letīs see....
Hi,
yeah let's see.
I did some backtest yesterday and no one was positive, but i don't believe at mt4 backtest, so let's see in forwarding test.

Said that, i'm pretty sure that we need more filter.

On the other hands we always stop the EA manually when we see any change in the market conditions.

gspe
  • #118
  • May 11, 2011 6:47am May 11, 2011 6:47am
After this miraculous recover i stopped the robot until the US news
Attached Image (click to enlarge)
Click to Enlarge Name: img_001.jpg Size: 79 KB
  • #119
  • May 11, 2011 9:15am May 11, 2011 9:15am
Lost $520 yesterday with original EA.

Was using it on GBP/USD and EUR/USD wih 0.1 lot.
  • #120
  • May 11, 2011 9:31am May 11, 2011 9:31am
Quoting piptastic7
Lost $520 yesterday with original EA.

Was using it on GBP/USD and EUR/USD wih 0.1 lot.
I'm runningthe original on renko using a RSI period 9 on renko period on 4 or 5 pairs


Seems to work better than on normal bars. Had a 2k drawdown on nornal bars in one night which is a bit much when targetting $10

rgds
Peter
0 traders viewing now