Caillou: another always-in auto-trader by SteveHopwood | Page 12 | 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

  • #221
  • May 19, 2011 9:29pm May 19, 2011 9:29pm
Quoting Hardpike
after 36 hour in 1D and 4H (gbp pairs) timeframe. 1.05 version, 20 pairs

basket 40
single 30-40 depending on pair volatility
24 hours trading
lot 0,1
18 TP HIT (!!)

closed 700 USD aprox
current DD 200 USD

seems to confirm backtesting predictions
wins far more per day than i have expected...great

regards
on 20 Pairs...that requires lot of margin isnt it... what was the maximum DD which you have seen. Wanna share the results on myfxbook or mt4i?
  • #222
  • May 20, 2011 2:27am May 20, 2011 2:27am
Quoting reachjj
on 20 Pairs...that requires lot of margin isnt it... what was the maximum DD which you have seen. Wanna share the results on myfxbook or mt4i?
currently I saw 350

i am on 5000 account...so we will see. the TP hit percent is great.

I am thinking that this can work "weekly" I mean, closing all remaining orders on Fridays doing most of the weeks profitable and in this way reseting the DD

time will tell
  • #223
  • May 20, 2011 5:03am May 20, 2011 5:03am
some tests
Attached File(s)
File Type: zip Caillou v1.07.zip   94 KB | 483 downloads
  • #224
  • May 20, 2011 5:47am May 20, 2011 5:47am
Quoting Hardpike
currently I saw 350

i am on 5000 account...so we will see. the TP hit percent is great.

I am thinking that this can work "weekly" I mean, closing all remaining orders on Fridays doing most of the weeks profitable and in this way reseting the DD

time will tell

Hardpike,

I just thought your dd will even be reduced if the trend on H4 will adhere with the weekly trend. I appreciate if you could run a backtest to prove my point.

Kind regards,

Edd
  • #225
  • May 20, 2011 5:51am May 20, 2011 5:51am
Quoting Edd Ganuelas
Hardpike,

I just thought your dd will even be reduced if the trend on H4 will adhere with the weekly trend. I appreciate if you could run a backtest to prove my point.

Kind regards,

Edd
sure, but the point is, how I identify the trend? I can try to do some test in the afternoon.
you mean just let the EA work on sell or on buy?
  • #226
  • Edited 3:18pm May 20, 2011 6:03am | Edited 3:18pm
Quoting Hardpike
sure, but the point is, how I identify the trend? I can try to do some test in the afternoon.
you mean just let the EA work on sell or on buy?
Yeah, I mean, if the EA must be tweak a little so that the H4 and W1 are singing together. Partly of Brian's concept of 3 auto trading has adopted this line if I am not mistaken using RSI.

Thanks and regards,

Edd
  • #227
  • May 20, 2011 6:34am May 20, 2011 6:34am
If you guys want to code and test that, I also do think RSI(20) is one of the best tools to spot a trend. It is also Steve's favourite.
RSI 20 > 55 uptrend
RSI 20 < 45 downtrend
In between 45 and 55: Ranging

Thanks for all the hard work and testing guys!
  • #228
  • May 20, 2011 6:38am May 20, 2011 6:38am
thank you

I will perform some tests...

will report results

regards
  • #229
  • May 20, 2011 6:57am May 20, 2011 6:57am
by the way I readed many times that backtesting on weekends cant be performed because high spreads.
there is a easy trick for this.

shut down MT4 before get close to friday close.
if you swictch on again with no internet connection will keep same spreads that the moment you shout it down.

all weekend with low spreads

oh yeah
  • #230
  • May 20, 2011 7:28am May 20, 2011 7:28am
oh one thing for Steve

I would be able to post in your threads, I think I show myself useful here, and I can be much more useful in other threads, but I do not the way to get in (an then, one day, I got in....)

So I put my knee on the ground lower my head and kindly ask for allowance.

regards
  • #231
  • May 20, 2011 9:22am May 20, 2011 9:22am
Quoting Edd Ganuelas
Yeah, I mean, if the EA must be tweak a little so that the H4 and W1 is singing together. Partly of Brian's concept of 3 auto trading has adopted this line if I am not mistaken using RSI.

Thanks and regards,

Edd
CONFIRMED in a hard way
i did tested audusd from 1 2009 until today
one case long only, following general trend.

long&short max DD 3000
only long DD 148 (!!!)

0,01 lot

I have to think on consequences of this...

similar results with other pairs with a general clear trend.....

regards
  • #232
  • May 20, 2011 11:03am May 20, 2011 11:03am
Hi Guys,

Sorry if I jump on this thread, but I found the original idea very nice.

I first would like to thank Steve for his EA and I would like to offer my idea on how to improve the performances.

I backtested the EA and I had very promising results on H4 different pairs. I tried also other TFs but H4 produced the best results.

Instead of trying to use trailing stops, RSIs, MAs, why not trying to edge the loosing position in order to limit the potential drawdown?
Something like: if the position if loosing more than X pips, than an opposite trade is open. The basket will be closed based on all the open trades taking into account shorts and longs until the basket profit has been reached.

Again, Just an idea. Please feel free to comment.
Cheers
Ogeid
  • #233
  • May 20, 2011 4:07pm May 20, 2011 4:07pm
Quoting ogeid
Hi Guys,

Sorry if I jump on this thread, but I found the original idea very nice.

I first would like to thank Steve for his EA and I would like to offer my idea on how to improve the performances....
I think H4 tests good because it hasn't got a lot of bars. I did some backtesting (although with some very low quality), it always ends up in margin call, just depends on how many bars it's testing over.

Opening oposite trades here isn't the awnser either i guess: it's just admitting your wrong. Making equal lots oposite trades is stupid, it would be better to just close the existing ones and open in different direction, doubling lots is an option but i always doubed that that tactic.

I guess you are going to have to admit the basket is wrong at some point, this just always leads to disaster somewhere in time. Or be able to delay that as much as possible so you are able to make profit and get it out of the account in time.

Just my thoughts about this one
  • #234
  • May 21, 2011 3:57am May 21, 2011 3:57am
Hi guys,

myabe for help I find an EA based on fractal but it's bugged and made only buys, but the system is as we want, maybe we can correct it

thanks a lot

Greg

http://codebase.mql4.com/573
  • #235
  • May 21, 2011 4:48am May 21, 2011 4:48am
Quoting gregart
Hi guys,

myabe for help I find an EA based on fractal but it's bugged and made only buys, but the system is as we want, maybe we can correct it

thanks a lot

Greg

http://codebase.mql4.com/573
There is a typo in the code, if you want a quick fix Click Modify Expert in Strategy Tester and scroll down near the bottom you will find

if(trend == 1)
ExtMapBuffer1[PU] = FU;
else
ExtMapBuffer1[PD] = FD;
// order section

you need to change the "trend" to "Trend" to get it to work correctly as it is "Trend" through rest of code. I did a quick test but get Ordersend Error 130 so MT4 is being a pain, i'm using on Alpari UK Demo. Not got time today to look into modifying code today.

Hope this helps

Mike
  • #236
  • May 21, 2011 4:50am May 21, 2011 4:50am
Here is the mod
Attached File(s)
File Type: mq4 Fractal ZigZag EA v1.01.mq4   7 KB | 257 downloads
  • #237
  • May 21, 2011 5:05am May 21, 2011 5:05am
thanks Mike, but it continue to make buys and open lot of orders not only with fractals, sorry for inconviences

Greg

Quoting magft
Here is the mod
  • #238
  • May 21, 2011 7:16pm May 21, 2011 7:16pm
Quoting Edd Ganuelas
Mike,

If you have spare time, can you please code a separate ea base on my request in post #50 for the higher time frame (D1). It will send a trade on brake-out. We then compare what is profitable. "Buy when price closes above the open price of previous down candle. Sell when price closes below the open price of previous up candle. User option to trade long or short or both".



Thanks,

Edd
Ok, i've had a go at implementing Edd's idea and current backtesting seems to have reduced DD from 61% to 12%.

I have also tried to correct the HiLo Stoploss for Greg but it still gives the ordersend error 130 so if someone with more experience in this could have a look that would be great.

So Updated dated version based on gspe's CT caillou is v1.08 and Edd's version is v1.08 Edd. I've added the option to allow the user to select only buy, sell or both. Also a CounterTrend option to compare original caillou against gspe caillou methods is added but option disabled as it doesnt seem to work properly yet. Default settings gave good results on a $1000 demo account 100:1.

I need sleep now, let me know if i've managed to break anything in the process of updating.

Mike
Attached File(s)
File Type: mq4 Caillou v1.08.mq4   21 KB | 295 downloads
File Type: mq4 Caillou v1.08 Edd.mq4   22 KB | 305 downloads
  • #239
  • May 22, 2011 5:03am May 22, 2011 5:03am
Thank you!
I will backtest both and shre my thoughts

Regards
  • #240
  • May 22, 2011 7:08am May 22, 2011 7:08am
Quoting magft
Ok, i've had a go at implementing Edd's idea and current backtesting seems to have reduced DD from 61% to 12%.

I have also tried to correct the HiLo Stoploss for Greg but it still gives the ordersend error 130 so if someone with more experience in this could have a look that would be great.

So Updated dated version based on gspe's CT caillou is v1.08 and Edd's version is v1.08 Edd. I've added the option to allow the user to select only buy, sell or both....
Sorry, none of them backtest for me....:-(
0 traders viewing now