Caillou: another always-in auto-trader by SteveHopwood | Page 11 | 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

  • #201
  • May 18, 2011 4:30pm May 18, 2011 4:30pm
Quoting gregart
OK, you're right Daily TF are terrific results but as regarding my 4 years experience in forex, I'll never trader again whitout stop, I ask to somebody to put an stop on V1.05 on the previous Daily Hight or Low bar. As the % of winning trade is 93 %, I think it can decrease significatly the DD wihtout decrease much profit and will secure the trades.

I made several backtests and will post them with new V1.05 with stop

thanks again

Greg
I have updated the EA with several modifications as requested. This version has a FractalCandleLookback which as far as i can tell is set to 2 for v1.04 and 3 for v1.05. I've added the min margin percent and max trades options as in v1.02. I've also added a HiLo stoploss that is based on HiLoShift number of bars of the current tf, if trend is up it uses Low[HiLoShift] and if trend is down High[HiLoShift] to set stoploss.

I've had a play around with H1 EU, as i only have 99% model quality data for this pair for last year. The results in the EA give max profit for lowest DD so far, still 61%! I need some sleep so will continue tomorrow, have fun.

Mike
Attached File(s)
File Type: mq4 Caillou v1.07.mq4   20 KB | 426 downloads
  • #202
  • May 18, 2011 4:44pm May 18, 2011 4:44pm
thanks so much magft, I will test it now
good night

Quoting magft
I have updated the EA with several modifications as requested. This version has a FractalCandleLookback which as far as i can tell is set to 2 for v1.04 and 3 for v1.05. I've added the min margin percent and max trades options as in v1.02. I've also added a HiLo stoploss that is based on HiLoShift number of bars of the current tf, if trend is up it uses Low[HiLoShift] and if trend is down High[HiLoShift] to set stoploss.

I've had a play around with H1 EU, as i only have 99% model quality data for this pair for last year. The results in the EA...
  • #203
  • May 18, 2011 6:40pm May 18, 2011 6:40pm
magft, I think there are an problem on the stop maybe cause sometimes the entry is too near to the stop or maybe due to an gap, I am not sure but not work properly.
but your MM work perfectly

thanks
Attached Image (click to enlarge)
Click to Enlarge Name: caillouv1.07.gif Size: 58 KB
  • #204
  • May 19, 2011 2:47am May 19, 2011 2:47am
Quoting magft
I have updated the EA with several modifications as requested. This version has a FractalCandleLookback which as far as i can tell is set to 2 for v1.04 and 3 for v1.05. I've added the min margin percent and max trades options as in v1.02. I've also added a HiLo stoploss that is based on HiLoShift number of bars of the current tf, if trend is up it uses Low[HiLoShift] and if trend is down High[HiLoShift] to set stoploss.

I've had a play around with H1 EU, as i only have 99% model quality data for this pair for last year. The results in the EA...
hi magft,

the robot looks promising. but in my opinion since it is an anti trend robot, it is bound to blow account in strong trends. have you considered hedging trades , as in the original, but , instead of trading every high low of candle, entering at fractal breakout?
that ways we always be with the trend, limit positions.
  • #205
  • May 19, 2011 3:13am May 19, 2011 3:13am
Quoting pmacel
Hardpike
did you keep all the other settings as default?

yes
but 24H trading

4H timeframe is good also in some pairs, mostly GBP ones... I dont know the reason....

I am glad you find high timeframes useful

I have ongoing a forward testing with 20 pairs, mostly 1D some 4H, is going very nice... we will see, is a slow burn testing...

regards

thank you Caillou for your amazing work
  • #206
  • May 19, 2011 4:14am May 19, 2011 4:14am
Quoting gregart
OK, you're right Daily TF are terrific results but as regarding my 4 years experience in forex, I'll never trader again whitout stop, I ask to somebody to put an stop on V1.05 on the previous Daily Hight or Low bar. As the % of winning trade is 93 %, I think it can decrease significatly the DD wihtout decrease much profit and will secure the trades.

I made several backtests and will post them with new V1.05 with stop

thanks again

Greg
the point is we need the orders to breath a lot.... this system works with drawdown...
even on daily, some strong trends (like eur/cad) blow the account (i always use 5000 usd for testing).
the point is in this case, time is slow, and we can react to strange market situations, edge or whatever....
and looking the backtestings, by the time that a 3000usd drawdown appear, you will have 15.000 on your account... thats the way.... thats the point

regards
  • #207
  • May 19, 2011 4:35am May 19, 2011 4:35am
Quoting Hardpike
the point is we need the orders to breath a lot.... this system works with drawdown...
even on daily, some strong trends (like eur/cad) blow the account (i always use 5000 usd for testing).
the point is in this case, time is slow, and we can react to strange market situations, edge or whatever....
and looking the backtestings, by the time that a 3000usd drawdown appear, you will have 15.000 on your account... thats the way.... thats the point

regards

Totally agree.....
  • #208
  • May 19, 2011 4:44am May 19, 2011 4:44am
I can't agree with this point, I had blow up too many accounts thinking DD could'nt be resolved and it's an part of the game.
One time is enough , only one, experiences tell that.
Anyway, I am confident with the system but wait for backtest with an stop

regards

Greg


Quoting Hardpike
the point is we need the orders to breath a lot.... this system works with drawdown...
even on daily, some strong trends (like eur/cad) blow the account (i always use 5000 usd for testing).
the point is in this case, time is slow, and we can react to strange market situations, edge or whatever....
and looking the backtestings, by the time that a 3000usd drawdown appear, you will have 15.000 on your account... thats the way.... thats the point

regards
  • #209
  • May 19, 2011 5:02am May 19, 2011 5:02am
Quoting gregart
I can't agree with this point, I had blow up too many accounts thinking DD could'nt be resolved and it's an part of the game.
One time is enough , only one, experiences tell that.
Anyway, I am confident with the system but wait for backtest with an stop

regards

Greg
Good. Be confident.
I blow accounts in the past too.
the point is, you have to win enough in a period of time, to get the drawdown-profit relation under control.
if statistically this relation is on your favor, you are on the bright side. The daily backtestings says that statistic is on our side. But as always in forex...strange things happens...all the time...
and in this case the exposure is always quite low, and things will happen slow, so there is time to take actions.

I like how this looks...
  • #210
  • May 19, 2011 5:44am May 19, 2011 5:44am
Quoting magft
I have updated the EA with several modifications as requested. This version has a FractalCandleLookback which as far as i can tell is set to 2 for v1.04 and 3 for v1.05. I've added the min margin percent and max trades options as in v1.02. I've also added a HiLo stoploss that is based on HiLoShift number of bars of the current tf, if trend is up it uses Low[HiLoShift] and if trend is down High[HiLoShift] to set stoploss.

I've had a play around with H1 EU, as i only have 99% model quality data for this pair for last year. The results in the EA...
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
  • #211
  • May 19, 2011 6:09am May 19, 2011 6:09am
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
Edd i'll have a look at this at some point this week not sure if it will be before the weekend, work is getting in the way!

Greg i had noticed the SL was not quite right, sometimes it is spot on other times it isnt. I have had a quick look this morning and it is possibly to do with crims min stoploss. I started to put a check in to change the value to allow the trade to go through but no joy. I realised that this EA doesnt have all the fancy stuff coded by Steve to overcome "OrderSend Error 130" in MT4 so to get this working i need to extract that code. Quickest option maybe to just take the whole mptm code in other EAs and strap in the triggers from this one. We then get all the other options as well, i need to have a look in detail at the code so that wont be til tonight.

I want to address the DD issue as if we get this working i cannot afford a massive account to run this that is why i also test with $500 to see if it has legs, that is the max i could start an account with. Realistically a $100 account to start is preferable to make sure it is profitable like TOB now TB. I just setup a live account for this and planning to set going next week.

Interesting times ahead, thanks for all the support and ideas and of course thanks Steve for all your hard work. Your code has helped my understand of coding EAs exponentially :-)

Mike
  • #212
  • May 19, 2011 6:27am May 19, 2011 6:27am
Hi,

I'm wondering something.

The original rules state
Quote
3.- Then at the close of each bar, if price is above start price and all previous buy orders are in profit, it opens a new buy order. If price is below start price and all previous sell orders are in profit, it opens a new sell order.

However, since version 0.4 of the modified EA, I see in the code
Inserted Code
if (AreBuysAllInProfit() && UpFract != 0)
{
if (UseRsi && trend == up && OpenTrades > 0) SendLot = Lot * WithTrendLotMultiplier;
if (OpenTrades == 1 && upl > 0 && SingleTradeProfitTarget > 0) return;
RefreshRates();
while(IsTradeContextBusy() ) Sleep(100);
ticket = OrderSend(Symbol(), OP_SELL, SendLot, Bid, 0, 0, 0, TradeComment, MagicNumber, 0, Red); 
}//if (AreBuysAllInProfit())         

if (AreSellsAllInProfit() && DnFract != 0)
{
if (UseRsi && trend == down && OpenTrades > 0) SendLot = Lot * WithTrendLotMultiplier;
if (OpenTrades == 1 && upl > 0 && SingleTradeProfitTarget > 0) return;
RefreshRates();
while(IsTradeContextBusy() ) Sleep(100);
ticket = OrderSend(Symbol(), OP_BUY, SendLot, Ask, 0, 0, 0, TradeComment, MagicNumber, 0, Green);
}//if (AreBuysAllInProfit())

shouldn't the OrderSends not be reversed? Now it looks like an OP_BUY is triggered when the sells are in profit.

Thanks
  • #213
  • May 19, 2011 6:42am May 19, 2011 6:42am
Since version 1.01 gspe reversed the original method and it proved to be more successful. Steve has not done any updates so the original rules apply to the EA in post #1. In the next version i was going to add the option to select which version to use but countertrending seems more promising.

Hope this helps

Mike
  • #214
  • May 19, 2011 6:44am May 19, 2011 6:44am
Yes, that helps :-)

thanks!
  • #215
  • May 19, 2011 6:55am May 19, 2011 6:55am
hi,
version 1.07 doesnt take any trades.
does someone else has the same problem?

cheers
  • #216
  • May 19, 2011 7:36am May 19, 2011 7:36am
Quoting Sgt.forex
hi,
version 1.07 doesnt take any trades.
does someone else has the same problem?

cheers
I've put it on a new EU H1 chart different broker and it enter a sell trade. Rsi is on by default at mo and trend is none, i still think this needs to be reviewed what to do during trending and ranging. I'll have a think later, better get on with work lol.
  • #217
  • May 19, 2011 7:39am May 19, 2011 7:39am
Quoting magft
I've put it on a new EU H1 chart different broker and it enter a sell trade. Rsi is on by default at mo and trend is none, i still think this needs to be reviewed what to do during trending and ranging. I'll have a think later, better get on with work lol.
i`ve got it on the 1min chart but changed to h1 now... lets see what happens but version 1.05 is still performing very well. got it on 4 charts at 1min tf...
have fun working :-P

cheers
  • #218
  • May 19, 2011 11:34am May 19, 2011 11:34am
Quoting Sgt.forex
i`ve got it on the 1min chart but changed to h1 now... lets see what happens ...
the problem was the money management funktion. there is a mistake at the lotsize calulation! i turned it off and it began trading

cheers
  • #219
  • May 19, 2011 1:28pm May 19, 2011 1:28pm
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
  • #220
  • May 19, 2011 5:44pm May 19, 2011 5:44pm
Quoting magft
Edd i'll have a look at this at some point this week not sure if it will be before the weekend, work is getting in the way!

Greg i had noticed the SL was not quite right, sometimes it is spot on other times it isnt. I have had a quick look this morning and it is possibly to do with crims min stoploss. I started to put a check in to change the value to allow the trade to go through but no joy. I realised that this EA doesnt have all the fancy stuff coded by Steve to overcome "OrderSend Error 130" in MT4 so to get this working i need to extract...
Thanks Mike,

Edd
0 traders viewing now